Authors: Jakub Olczak, Max Gordon
Abstract: Background: Radiography (X-rays) is the dominant modality in orthopedics, and improving the interpretation of radiographs is clinically relevant. Machine learning (ML) has revolutionized data analysis and has been applied to medicine, with some success, in the form of natural language processing (NLP) and artificial neural networks (ANN). Latent Dirichlet allocation (LDA) is an NLP method that automatically categorizes documents into topics. Successfully applying ML to orthopedic radiography could enable the creation of computer-aided decision systems for use in the clinic. We studied how an automated ML pipeline could classify orthopedic trauma radiographs from radiologist reports. Methods: Wrist and ankle radiographs from Danderyd Hospital in Sweden taken between 2002 and 2015, with radiologist reports. LDA was used to create image labels for radiographs from the radiologist reports. Radiographs and labels were used to train an image recognition ANN. The ANN outcomes were manually reviewed to get an accurate estimate of the method's utility and accuracy. Results: Image Labels generated via LDA could successfully train the ANN. The ANN reached an accuracy between 91% and 60% compared to a gold standard, depending on the label. Conclusions: We found that LDA was unsuited to label orthopedic radiographs from reports with high accuracy. However, despite this, the ANN could learn to detect some features in radiographs with high accuracy. The study also illustrates how ML and ANN can be applied to medical research.
Authors: Jiseung Hong, Changmin Lee, Gyusang Yu
Abstract: TL;DR Perform 3D object editing selectively by disentangling it from the background scene. Instruct-NeRF2NeRF (in2n) is a promising method that enables editing of 3D scenes composed of Neural Radiance Field (NeRF) using text prompts. However, it is challenging to perform geometrical modifications such as shrinking, scaling, or moving on both the background and object simultaneously. In this project, we enable geometrical changes of objects within the 3D scene by selectively editing the object after separating it from the scene. We perform object segmentation and background inpainting respectively, and demonstrate various examples of freely resizing or moving disentangled objects within the three-dimensional space.
Authors: Rishabh Srivastava, Addrish Roy
Abstract: Our study delves into the fusion of abstract art interpretation and text-to-image synthesis, addressing the challenge of achieving precise spatial control over image composition solely through textual prompts. Leveraging the capabilities of ControlNet, we empower users with finer control over the synthesis process, enabling enhanced manipulation of synthesized imagery. Inspired by the minimalist forms found in abstract artworks, we introduce a novel condition crafted from geometric primitives such as triangles.
Authors: Taigo Sakai, Kazuhiro Hotta
Abstract: Deep learning has excelled in image recognition tasks through neural networks inspired by the human brain. However, the necessity for large models to improve prediction accuracy introduces significant computational demands and extended training times.Conventional methods such as fine-tuning, knowledge distillation, and pruning have the limitations like potential accuracy drops. Drawing inspiration from human neurogenesis, where neuron formation continues into adulthood, we explore a novel approach of progressively increasing neuron numbers in compact models during training phases, thereby managing computational costs effectively. We propose a method that reduces feature extraction biases and neuronal redundancy by introducing constraints based on neuron similarity distributions. This approach not only fosters efficient learning in new neurons but also enhances feature extraction relevancy for given tasks. Results on CIFAR-10 and CIFAR-100 datasets demonstrated accuracy improvement, and our method pays more attention to whole object to be classified in comparison with conventional method through Grad-CAM visualizations. These results suggest that our method's potential to decision-making processes.
Authors: Qi Qian, Juhua Hu
Abstract: Vision-language pre-training such as CLIP enables zero-shot transfer that can classify images according to the candidate class names. While CLIP demonstrates an impressive zero-shot performance on diverse downstream tasks, the distribution from the target data has not been leveraged sufficiently. In this work, we study a novel online zero-shot transfer scenario, where each image arrives in a random order for classification and is visited only once to obtain prediction immediately without storing its representation. Compared with the vanilla zero-shot classification, the proposed framework preserves its flexibility for online service while considering the statistics of the arrived images as the side information to capture the distribution of target data, which can help improve the performance of real-world applications. To tackle the challenge of effective online optimization, we first develop online label learning to model the target data distribution. Then, the proxy of each class in the vision space is further optimized with the proposed online proxy learning method to mitigate the modality gap between images and text. The convergence of both online strategies can be theoretically guaranteed. By combining the predicted label from the online label learning and proxy learning, our online zero-shot transfer method (OnZeta) achieves $78.94\%$ accuracy on ImageNet without accessing the entire data set. Moreover, extensive experiments on other 13 downstream tasks with different vision encoders show a more than $3\%$ improvement on average, which demonstrates the effectiveness of our proposal. Code is available at \url{https://github.com/idstcv/OnZeta}.
Authors: Zitao Shuai, Chenwei Wu, Zhengxu Tang, Bowen Song, Liyue Shen
Abstract: Diffusion Transformers (DiTs) have achieved remarkable success in diverse and high-quality text-to-image(T2I) generation. However, how text and image latents individually and jointly contribute to the semantics of generated images, remain largely unexplored. Through our investigation of DiT's latent space, we have uncovered key findings that unlock the potential for zero-shot fine-grained semantic editing: (1) Both the text and image spaces in DiTs are inherently decomposable. (2) These spaces collectively form a disentangled semantic representation space, enabling precise and fine-grained semantic control. (3) Effective image editing requires the combined use of both text and image latent spaces. Leveraging these insights, we propose a simple and effective Extract-Manipulate-Sample (EMS) framework for zero-shot fine-grained image editing. Our approach first utilizes a multi-modal Large Language Model to convert input images and editing targets into text descriptions. We then linearly manipulate text embeddings based on the desired editing degree and employ constrained score distillation sampling to manipulate image embeddings. We quantify the disentanglement degree of the latent space of diffusion models by proposing a new metric. To evaluate fine-grained editing performance, we introduce a comprehensive benchmark incorporating both human annotations, manual evaluation, and automatic metrics. We have conducted extensive experimental results and in-depth analysis to thoroughly uncover the semantic disentanglement properties of the diffusion transformer, as well as the effectiveness of our proposed method. Our annotated benchmark dataset is publicly available at https://anonymous.com/anonymous/EMS-Benchmark, facilitating reproducible research in this domain.
Authors: Qi Qian, Yuanhong Xu, Juhua Hu
Abstract: Deep features extracted from certain layers of a pre-trained deep model show superior performance over the conventional hand-crafted features. Compared with fine-tuning or linear probing that can explore diverse augmentations, \eg, random crop/flipping, in the original input space, the appropriate augmentations for learning with fixed deep features are more challenging and have been less investigated, which degenerates the performance. To unleash the potential of fixed deep features, we propose a novel semantic adversarial augmentation (SeA) in the feature space for optimization. Concretely, the adversarial direction implied by the gradient will be projected to a subspace spanned by other examples to preserve the semantic information. Then, deep features will be perturbed with the semantic direction, and augmented features will be applied to learn the classifier. Experiments are conducted on $11$ benchmark downstream classification tasks with $4$ popular pre-trained models. Our method is $2\%$ better than the deep features without SeA on average. Moreover, compared to the expensive fine-tuning that is expected to give good performance, SeA shows a comparable performance on $6$ out of $11$ tasks, demonstrating the effectiveness of our proposal in addition to its efficiency. Code is available at \url{https://github.com/idstcv/SeA}.
Authors: Guangzong Chen, Zhi-Hong Mao, Mingui Sun, Kangni Liu, Wenyan Jia
Abstract: Traditional dietary assessment methods heavily rely on self-reporting, which is time-consuming and prone to bias. Recent advancements in Artificial Intelligence (AI) have revealed new possibilities for dietary assessment, particularly through analysis of food images. Recognizing foods and estimating food volumes from images are known as the key procedures for automatic dietary assessment. However, both procedures required large amounts of training images labeled with food names and volumes, which are currently unavailable. Alternatively, recent studies have indicated that training images can be artificially generated using Generative Adversarial Networks (GANs). Nonetheless, convenient generation of large amounts of food images with known volumes remain a challenge with the existing techniques. In this work, we present a simple GAN-based neural network architecture for conditional food image generation. The shapes of the food and container in the generated images closely resemble those in the reference input image. Our experiments demonstrate the realism of the generated images and shape-preserving capabilities of the proposed framework.
Authors: Zhenyuan Liu, Yu Guo, Xinyuan Li, Bernd Bickel, Ran Zhang
Abstract: We present Bidirectional Gaussian Primitives, an image-based novel view synthesis technique designed to represent and render 3D objects with surface and volumetric materials under dynamic illumination. Our approach integrates light intrinsic decomposition into the Gaussian splatting framework, enabling real-time relighting of 3D objects. To unify surface and volumetric material within a cohesive appearance model, we adopt a light- and view-dependent scattering representation via bidirectional spherical harmonics. Our model does not use a specific surface normal-related reflectance function, making it more compatible with volumetric representations like Gaussian splatting, where the normals are undefined. We demonstrate our method by reconstructing and rendering objects with complex materials. Using One-Light-At-a-Time (OLAT) data as input, we can reproduce photorealistic appearances under novel lighting conditions in real time.
Authors: Zhenyu Zhang, Guangyao Chen, Yixiong Zou, Yuhua Li, Ruixuan Li
Abstract: Few-shot open-set recognition (FSOR) is a challenging task that requires a model to recognize known classes and identify unknown classes with limited labeled data. Existing approaches, particularly Negative-Prototype-Based methods, generate negative prototypes based solely on known class data. However, as the unknown space is infinite while the known space is limited, these methods suffer from limited representation capability. To address this limitation, we propose a novel approach, termed \textbf{D}iversified \textbf{N}egative \textbf{P}rototypes \textbf{G}enerator (DNPG), which adopts the principle of "learning unknowns from unknowns." Our method leverages the unknown space information learned from base classes to generate more representative negative prototypes for novel classes. During the pre-training phase, we learn the unknown space representation of the base classes. This representation, along with inter-class relationships, is then utilized in the meta-learning process to construct negative prototypes for novel classes. To prevent prototype collapse and ensure adaptability to varying data compositions, we introduce the Swap Alignment (SA) module. Our DNPG model, by learning from the unknown space, generates negative prototypes that cover a broader unknown space, thereby achieving state-of-the-art performance on three standard FSOR datasets.
Authors: Hyo Jong Chung, Byungkon Kang, Yoonseok Yang
Abstract: Driver motion recognition is a principal factor in ensuring the safety of driving systems. This paper presents a novel system for learning and predicting driver motions and an event-based high-resolution (1280x720) dataset, N-DriverMotion, newly collected to train on a neuromorphic vision system. The system comprises an event-based camera that generates the first high-resolution driver motion dataset representing spike inputs and efficient spiking neural networks (SNNs) that are effective in training and predicting the driver's gestures. The event dataset consists of 13 driver motion categories classified by direction (front, side), illumination (bright, moderate, dark), and participant. A novel simplified four-layer convolutional spiking neural network (CSNN) that we proposed was directly trained using the high-resolution dataset without any time-consuming preprocessing. This enables efficient adaptation to on-device SNNs for real-time inference on high-resolution event-based streams. Compared with recent gesture recognition systems adopting neural networks for vision processing, the proposed neuromorphic vision system achieves comparable accuracy, 94.04\%, in recognizing driver motions with the CSNN architecture. Our proposed CSNN and the dataset can be used to develop safer and more efficient driver monitoring systems for autonomous vehicles or edge devices requiring an efficient neural network architecture.
Authors: Zhenyu Zhang, Guangyao Chen, Yixiong Zou, Zhimeng Huang, Yuhua Li, Ruixuan Li
Abstract: Humans exhibit a remarkable ability to learn quickly from a limited number of labeled samples, a capability that starkly contrasts with that of current machine learning systems. Unsupervised Few-Shot Learning (U-FSL) seeks to bridge this divide by reducing reliance on annotated datasets during initial training phases. In this work, we first quantitatively assess the impacts of Masked Image Modeling (MIM) and Contrastive Learning (CL) on few-shot learning tasks. Our findings highlight the respective limitations of MIM and CL in terms of discriminative and generalization abilities, which contribute to their underperformance in U-FSL contexts. To address these trade-offs between generalization and discriminability in unsupervised pretraining, we introduce a novel paradigm named Masked Image Contrastive Modeling (MICM). MICM creatively combines the targeted object learning strength of CL with the generalized visual feature learning capability of MIM, significantly enhancing its efficacy in downstream few-shot learning inference. Extensive experimental analyses confirm the advantages of MICM, demonstrating significant improvements in both generalization and discrimination capabilities for few-shot learning. Our comprehensive quantitative evaluations further substantiate the superiority of MICM, showing that our two-stage U-FSL framework based on MICM markedly outperforms existing leading baselines.
Authors: Yangyang Xu, Wenqi Shao, Yong Du, Haiming Zhu, Yang Zhou, Ping Luo, Shengfeng He
Abstract: Recent advancements in text-guided diffusion models have unlocked powerful image manipulation capabilities, yet balancing reconstruction fidelity and editability for real images remains a significant challenge. In this work, we introduce \textbf{T}ask-\textbf{O}riented \textbf{D}iffusion \textbf{I}nversion (\textbf{TODInv}), a novel framework that inverts and edits real images tailored to specific editing tasks by optimizing prompt embeddings within the extended \(\mathcal{P}^*\) space. By leveraging distinct embeddings across different U-Net layers and time steps, TODInv seamlessly integrates inversion and editing through reciprocal optimization, ensuring both high fidelity and precise editability. This hierarchical editing mechanism categorizes tasks into structure, appearance, and global edits, optimizing only those embeddings unaffected by the current editing task. Extensive experiments on benchmark dataset reveal TODInv's superior performance over existing methods, delivering both quantitative and qualitative enhancements while showcasing its versatility with few-step diffusion model.
Authors: Xuran Hu, Mingzhe Zhu, Zhenpeng Feng, Milo\v{s} Dakovi\'c, Ljubi\v{s}a Stankovi\'c
Abstract: The inherent "black box" nature of deep neural networks (DNNs) compromises their transparency and reliability. Recently, explainable AI (XAI) has garnered increasing attention from researchers. Several perturbation-based interpretations have emerged. However, these methods often fail to adequately consider feature dependencies. To solve this problem, we introduce a perturbation-based interpretation guided by feature coalitions, which leverages deep information of network to extract correlated features. Then, we proposed a carefully-designed consistency loss to guide network interpretation. Both quantitative and qualitative experiments are conducted to validate the effectiveness of our proposed method. Code is available at github.com/Teriri1999/Perturebation-on-Feature-Coalition.
Authors: Rui Zhang, Yaosen Chen, Yuegen Liu, Wei Wang, Xuming Wen, Hongxia Wang
Abstract: Transition videos play a crucial role in media production, enhancing the flow and coherence of visual narratives. Traditional methods like morphing often lack artistic appeal and require specialized skills, limiting their effectiveness. Recent advances in diffusion model-based video generation offer new possibilities for creating transitions but face challenges such as poor inter-frame relationship modeling and abrupt content changes. We propose a novel training-free Transition Video Generation (TVG) approach using video-level diffusion models that addresses these limitations without additional training. Our method leverages Gaussian Process Regression ($\mathcal{GPR}$) to model latent representations, ensuring smooth and dynamic transitions between frames. Additionally, we introduce interpolation-based conditional controls and a Frequency-aware Bidirectional Fusion (FBiF) architecture to enhance temporal control and transition reliability. Evaluations of benchmark datasets and custom image pairs demonstrate the effectiveness of our approach in generating high-quality smooth transition videos. The code are provided in https://sobeymil.github.io/tvg.com.
Authors: Wenhao Li, Yichao Cao, Xie Su, Xi Lin, Shan You, Mingkai Zheng, Yi Chen, Chang Xu
Abstract: Video generation models hold substantial potential in areas such as filmmaking. However, current video diffusion models need high computational costs and produce suboptimal results due to high complexity of video generation task. In this paper, we propose \textbf{ConFiner}, an efficient high-quality video generation framework that decouples video generation into easier subtasks: structure \textbf{con}trol and spatial-temporal re\textbf{fine}ment. It can generate high-quality videos with chain of off-the-shelf diffusion model experts, each expert responsible for a decoupled subtask. During the refinement, we introduce coordinated denoising, which can merge multiple diffusion experts' capabilities into a single sampling. Furthermore, we design ConFiner-Long framework, which can generate long coherent video with three constraint strategies on ConFiner. Experimental results indicate that with only 10\% of the inference cost, our ConFiner surpasses representative models like Lavie and Modelscope across all objective and subjective metrics. And ConFiner-Long can generate high-quality and coherent videos with up to 600 frames.
Authors: Junjie Liu
Abstract: Large-scale face clustering has achieved significant progress, with many efforts dedicated to learning to cluster large-scale faces with supervised-learning. However, complex model design and tedious clustering processes are typical in existing methods. Such limitations result in infeasible clustering in real-world applications. Reasonable and efficient model design and training need to be taken into account. Besides, developing unsupervised face clustering algorithms is crucial, which are more realistic in real-world applications. In this paper, we propose a novel unsupervised face clustering algorithm FC-ES and a novel supervised face clustering algorithm FC-ESER to address these issues. An efficient and effective neighbor-based edge probability and a novel early stopping strategy are proposed in FC-ES, guaranteeing the accuracy and recall of large-scale face clustering simultaneously. Furthermore, to take advantage of supervised learning, a novel edge recall strategy is proposed in FC-ESER to further recall the edge connections that are not connected in FC-ES. Extensive experiments on multiple benchmarks for face, person, and vehicle clustering show that our proposed FC-ES and FC-ESER significantly outperform previous state-of-the-art methods. Our code will be available at https://github.com/jumptoliujj/FC-ESER.
Authors: Aditya Taparia, Som Sagar, Ransalu Senanayake
Abstract: Concept-based explanations have become a popular choice for explaining deep neural networks post-hoc because, unlike most other explainable AI techniques, they can be used to test high-level visual "concepts" that are not directly related to feature attributes. For instance, the concept of "stripes" is important to classify an image as a zebra. Concept-based explanation methods, however, require practitioners to guess and collect multiple candidate concept image sets, which can often be imprecise and labor-intensive. Addressing this limitation, in this paper, we frame concept image set creation as an image generation problem. However, since naively using a generative model does not result in meaningful concepts, we devise a reinforcement learning-based preference optimization algorithm that fine-tunes the vision-language generative model from approximate textual descriptions of concepts. Through a series of experiments, we demonstrate the capability of our method to articulate complex, abstract concepts that are otherwise challenging to craft manually. In addition to showing the efficacy and reliability of our method, we show how our method can be used as a diagnostic tool for analyzing neural networks.
Authors: Chao Chen, Ruoyu Wang, Yuliang Guo, Cheng Zhao, Xinyu Huang, Chen Feng, Liu Ren
Abstract: Autonomous driving in complex urban scenarios requires 3D perception to be both comprehensive and precise. Traditional 3D perception methods focus on object detection, resulting in sparse representations that lack environmental detail. Recent approaches estimate 3D occupancy around vehicles for a more comprehensive scene representation. However, dense 3D occupancy prediction increases computational demands, challenging the balance between efficiency and resolution. High-resolution occupancy grids offer accuracy but demand substantial computational resources, while low-resolution grids are efficient but lack detail. To address this dilemma, we introduce AdaOcc, a novel adaptive-resolution, multi-modal prediction approach. Our method integrates object-centric 3D reconstruction and holistic occupancy prediction within a single framework, performing highly detailed and precise 3D reconstruction only in regions of interest (ROIs). These high-detailed 3D surfaces are represented in point clouds, thus their precision is not constrained by the predefined grid resolution of the occupancy map. We conducted comprehensive experiments on the nuScenes dataset, demonstrating significant improvements over existing methods. In close-range scenarios, we surpass previous baselines by over 13% in IOU, and over 40% in Hausdorff distance. In summary, AdaOcc offers a more versatile and effective framework for delivering accurate 3D semantic occupancy prediction across diverse driving scenarios.
Authors: Chen Rao, Guangyuan Li, Zehua Lan, Jiakai Sun, Junsheng Luan, Wei Xing, Lei Zhao, Huaizhong Lin, Jianfeng Dong, Dalong Zhang
Abstract: Current video deblurring methods have limitations in recovering high-frequency information since the regression losses are conservative with high-frequency details. Since Diffusion Models (DMs) have strong capabilities in generating high-frequency details, we consider introducing DMs into the video deblurring task. However, we found that directly applying DMs to the video deblurring task has the following problems: (1) DMs require many iteration steps to generate videos from Gaussian noise, which consumes many computational resources. (2) DMs are easily misled by the blurry artifacts in the video, resulting in irrational content and distortion of the deblurred video. To address the above issues, we propose a novel video deblurring framework VD-Diff that integrates the diffusion model into the Wavelet-Aware Dynamic Transformer (WADT). Specifically, we perform the diffusion model in a highly compact latent space to generate prior features containing high-frequency information that conforms to the ground truth distribution. We design the WADT to preserve and recover the low-frequency information in the video while utilizing the high-frequency information generated by the diffusion model. Extensive experiments show that our proposed VD-Diff outperforms SOTA methods on GoPro, DVD, BSD, and Real-World Video datasets.
Authors: Jiwei Guan, Tianyu Ding, Longbing Cao, Lei Pan, Chen Wang, Xi Zheng
Abstract: Vision-language pretraining (VLP) with transformers has demonstrated exceptional performance across numerous multimodal tasks. However, the adversarial robustness of these models has not been thoroughly investigated. Existing multimodal attack methods have largely overlooked cross-modal interactions between visual and textual modalities, particularly in the context of cross-attention mechanisms. In this paper, we study the adversarial vulnerability of recent VLP transformers and design a novel Joint Multimodal Transformer Feature Attack (JMTFA) that concurrently introduces adversarial perturbations in both visual and textual modalities under white-box settings. JMTFA strategically targets attention relevance scores to disrupt important features within each modality, generating adversarial samples by fusing perturbations and leading to erroneous model predictions. Experimental results indicate that the proposed approach achieves high attack success rates on vision-language understanding and reasoning downstream tasks compared to existing baselines. Notably, our findings reveal that the textual modality significantly influences the complex fusion processes within VLP transformers. Moreover, we observe no apparent relationship between model size and adversarial robustness under our proposed attacks. These insights emphasize a new dimension of adversarial robustness and underscore potential risks in the reliable deployment of multimodal AI systems.
Authors: Hongwu Li, Zhenliang Zhang, Wei Wang
Abstract: Human Action Recognition (HAR) is a very crucial task in computer vision. It helps to carry out a series of downstream tasks, like understanding human behaviors. Due to the complexity of human behaviors, many highly valuable behaviors are not yet encompassed within the available datasets for HAR, e.g., human habitual behaviors (HHBs). HHBs hold significant importance for analyzing a person's personality, habits, and psychological changes. To solve these problems, in this work, we build a novel video dataset to demonstrate various HHBs. These behaviors in the proposed dataset are able to reflect internal mental states and specific emotions of the characters, e.g., crossing arms suggests to shield oneself from perceived threats. The dataset contains 30 categories of habitual behaviors including more than 300,000 frames and 6,899 action instances. Since these behaviors usually appear at small local parts of human action videos, it is difficult for existing action recognition methods to handle these local features. Therefore, we also propose a two-stream model using both human skeletons and RGB appearances. Experimental results demonstrate that our proposed method has much better performance in action recognition than the existing methods on the proposed dataset.
Authors: Jinchao Ge, Zeyu Zhang, Minh Hieu Phan, Bowen Zhang, Akide Liu, Yang Zhao
Abstract: Active learning enhances annotation efficiency by selecting the most revealing samples for labeling, thereby reducing reliance on extensive human input. Previous methods in semantic segmentation have centered on individual pixels or small areas, neglecting the rich patterns in natural images and the power of advanced pre-trained models. To address these challenges, we propose three key contributions: Firstly, we introduce Entity-Superpixel Annotation (ESA), an innovative and efficient active learning strategy which utilizes a class-agnostic mask proposal network coupled with super-pixel grouping to capture local structural cues. Additionally, our method selects a subset of entities within each image of the target domain, prioritizing superpixels with high entropy to ensure comprehensive representation. Simultaneously, it focuses on a limited number of key entities, thereby optimizing for efficiency. By utilizing an annotator-friendly design that capitalizes on the inherent structure of images, our approach significantly outperforms existing pixel-based methods, achieving superior results with minimal queries, specifically reducing click cost by 98% and enhancing performance by 1.71%. For instance, our technique requires a mere 40 clicks for annotation, a stark contrast to the 5000 clicks demanded by conventional methods.
Authors: Keon-Hee Park, Hakyung Lee, Kyungwoo Song, Gyeong-Moon Park
Abstract: With the advancement of deep neural networks in computer vision, artificial intelligence (AI) is widely employed in real-world applications. However, AI still faces limitations in mimicking high-level human capabilities, such as novel category discovery, for practical use. While some methods utilizing offline continual learning have been proposed for novel category discovery, they neglect the continuity of data streams in real-world settings. In this work, we introduce Online Continuous Generalized Category Discovery (OCGCD), which considers the dynamic nature of data streams where data can be created and deleted in real time. Additionally, we propose a novel method, DEAN, Discovery via Energy guidance and feature AugmentatioN, which can discover novel categories in an online manner through energy-guided discovery and facilitate discriminative learning via energy-based contrastive loss. Furthermore, DEAN effectively pseudo-labels unlabeled data through variance-based feature augmentation. Experimental results demonstrate that our proposed DEAN achieves outstanding performance in proposed OCGCD scenario.
Authors: Juan E. Tapia, Sebastian Gonzalez, Daniel Benalcazar, Christoph Busch
Abstract: In the last few years, face morphing has been shown to be a complex challenge for Face Recognition Systems (FRS). Thus, the evaluation of other biometric modalities such as fingerprint, iris, and others must be explored and evaluated to enhance biometric systems. This work proposes an end-to-end framework to produce iris morphs at the image level, creating morphs from Periocular iris images. This framework considers different stages such as pair subject selection, segmentation, morph creation, and a new iris recognition system. In order to create realistic morphed images, two approaches for subject selection are explored: random selection and similar radius size selection. A vulnerability analysis and a Single Morphing Attack Detection algorithm were also explored. The results show that this approach obtained very realistic images that can confuse conventional iris recognition systems.
Authors: Yixuan Li, Zan Wang, Wei Liang
Abstract: We propose Reasoning to Ground (R2G), a neural symbolic model that grounds the target objects within 3D scenes in a reasoning manner. In contrast to prior works, R2G explicitly models the 3D scene with a semantic concept-based scene graph; recurrently simulates the attention transferring across object entities; thus makes the process of grounding the target objects with the highest probability interpretable. Specifically, we respectively embed multiple object properties within the graph nodes and spatial relations among entities within the edges, utilizing a predefined semantic vocabulary. To guide attention transferring, we employ learning or prompting-based methods to analyze the referential utterance and convert it into reasoning instructions within the same semantic space. In each reasoning round, R2G either (1) merges current attention distribution with the similarity between the instruction and embedded entity properties or (2) shifts the attention across the scene graph based on the similarity between the instruction and embedded spatial relations. The experiments on Sr3D/Nr3D benchmarks show that R2G achieves a comparable result with the prior works while maintaining improved interpretability, breaking a new path for 3D language grounding.
Authors: Adil Meric, Umut Kocasari, Matthias Nie{\ss}ner, Barbara Roessle
Abstract: Neural Radiance Fields (NeRF) have emerged as a powerful tool for creating highly detailed and photorealistic scenes. Existing methods for NeRF-based 3D style transfer need extensive per-scene optimization for single or multiple styles, limiting the applicability and efficiency of 3D style transfer. In this work, we overcome the limitations of existing methods by rendering stylized novel views from a NeRF without the need for per-scene or per-style optimization. To this end, we take advantage of a generalizable NeRF model to facilitate style transfer in 3D, thereby enabling the use of a single learned model across various scenes. By incorporating a hypernetwork into a generalizable NeRF, our approach enables on-the-fly generation of stylized novel views. Moreover, we introduce a novel flow-based multi-view consistency loss to preserve consistency across multiple views. We evaluate our method across various scenes and artistic styles and show its performance in generating high-quality and multi-view consistent stylized images without the need for a scene-specific implicit model. Our findings demonstrate that this approach not only achieves a good visual quality comparable to that of per-scene methods but also significantly enhances efficiency and applicability, marking a notable advancement in the field of 3D style transfer.
Authors: Ying Jin, Jinlong Peng, Qingdong He, Teng Hu, Hao Chen, Jiafu Wu, Wenbing Zhu, Mingmin Chi, Jun Liu, Yabiao Wang, Chengjie Wang
Abstract: The performance of anomaly inspection in industrial manufacturing is constrained by the scarcity of anomaly data. To overcome this challenge, researchers have started employing anomaly generation approaches to augment the anomaly dataset. However, existing anomaly generation methods suffer from limited diversity in the generated anomalies and struggle to achieve a seamless blending of this anomaly with the original image. In this paper, we overcome these challenges from a new perspective, simultaneously generating a pair of the overall image and the corresponding anomaly part. We propose DualAnoDiff, a novel diffusion-based few-shot anomaly image generation model, which can generate diverse and realistic anomaly images by using a dual-interrelated diffusion model, where one of them is employed to generate the whole image while the other one generates the anomaly part. Moreover, we extract background and shape information to mitigate the distortion and blurriness phenomenon in few-shot image generation. Extensive experiments demonstrate the superiority of our proposed model over state-of-the-art methods in terms of both realism and diversity. Overall, our approach significantly improves the performance of downstream anomaly detection tasks, including anomaly detection, anomaly localization, and anomaly classification tasks.
Authors: Yujin Lee, Seoyoon Jang, Hyunsoo Yoon
Abstract: Few-shot Anomaly Detection (FAD) poses significant challenges due to the limited availability of training samples and the frequent absence of abnormal samples. Previous approaches often rely on annotations or true abnormal samples to improve detection, but such textual or visual cues are not always accessible. To address this, we introduce AnoPLe, a multi-modal prompt learning method designed for anomaly detection without prior knowledge of anomalies. AnoPLe simulates anomalies and employs bidirectional coupling of textual and visual prompts to facilitate deep interaction between the two modalities. Additionally, we integrate a lightweight decoder with a learnable multi-view signal, trained on multi-scale images to enhance local semantic comprehension. To further improve performance, we align global and local semantics, enriching the image-level understanding of anomalies. The experimental results demonstrate that AnoPLe achieves strong FAD performance, recording 94.1% and 86.2% Image AUROC on MVTec-AD and VisA respectively, with only around a 1% gap compared to the SoTA, despite not being exposed to true anomalies. Code is available at https://github.com/YoojLee/AnoPLe.
Authors: Giuseppe Macario
Abstract: The metaverse has received much attention in the literature and industry in the last few years, but the lack of an open and cross-platform architecture has led to many distinct metaverses that cannot communicate with each other. This work proposes a WebXR-based cross-platform architecture for developing spatial web apps using the A-Frame and Networked-Aframe frameworks with a view to an open and interoperable metaverse, accessible from both the web and extended reality devices. A prototype was implemented and evaluated, supporting the capability of the technology stack to enable immersive experiences across different platforms and devices. Positive feedback on ease of use of the immersive environment further corroborates the proposed approach, underscoring its effectiveness in facilitating engaging and interactive virtual spaces. By adhering to principles of interoperability and inclusivity, it lives up to Tim Berners-Lee's vision of the World Wide Web as an open platform that transcends geographical and technical boundaries.
Authors: Ammar Ahmed, Ali Shariq Imran, Zenun Kastrati, Sher Muhammad Daudpota, Mohib Ullah, Waheed Noord
Abstract: Wrist pathologies, {particularly fractures common among children and adolescents}, present a critical diagnostic challenge. While X-ray imaging remains a prevalent diagnostic tool, the increasing misinterpretation rates highlight the need for more accurate analysis, especially considering the lack of specialized training among many surgeons and physicians. Recent advancements in deep convolutional neural networks offer promise in automating pathology detection in trauma X-rays. However, distinguishing subtle variations between {pediatric} wrist pathologies in X-rays remains challenging. Traditional manual annotation, though effective, is laborious, costly, and requires specialized expertise. {In this paper, we address the challenge of pediatric wrist pathology recognition with a fine-grained approach, aimed at automatically identifying discriminative regions in X-rays without manual intervention. We refine our fine-grained architecture through ablation analysis and the integration of LION.} Leveraging Grad-CAM, an explainable AI technique, we highlight these regions. Despite using limited data, reflective of real-world medical study constraints, our method consistently outperforms state-of-the-art image recognition models on both augmented and original (challenging) test sets. {Our proposed refined architecture achieves an increase in accuracy of 1.06% and 1.25% compared to the baseline method, resulting in accuracies of 86% and 84%, respectively. Moreover, our approach demonstrates the highest fracture sensitivity of 97%, highlighting its potential to enhance wrist pathology recognition. The implementation code can be found at https://github.com/ammarlodhi255/fine-grained-approach-to-wrist-pathology-recognition
URLs: https://github.com/ammarlodhi255/fine-grained-approach-to-wrist-pathology-recognition
Authors: Huy Hoang Nguyen, Cuong Nhat Nguyen, Xuan Tung Dao, Quoc Trung Duong, Dzung Pham Thi Kim, Minh-Tan Pham
Abstract: This paper aims to conduct a comparative analysis of contemporary Variational Autoencoder (VAE) architectures employed in anomaly detection, elucidating their performance and behavioral characteristics within this specific task. The architectural configurations under consideration encompass the original VAE baseline, the VAE with a Gaussian Random Field prior (VAE-GRF), and the VAE incorporating a vision transformer (ViT-VAE). The findings reveal that ViT-VAE exhibits exemplary performance across various scenarios, whereas VAE-GRF may necessitate more intricate hyperparameter tuning to attain its optimal performance state. Additionally, to mitigate the propensity for over-reliance on results derived from the widely used MVTec dataset, this paper leverages the recently-public MiAD dataset for benchmarking. This deliberate inclusion seeks to enhance result competitiveness by alleviating the impact of domain-specific models tailored exclusively for MVTec, thereby contributing to a more robust evaluation framework. Codes is available at https://github.com/endtheme123/VAE-compare.git.
Authors: Hao Yang, Qianyu Zhou, Haijia Sun, Xiangtai Li, Fengqi Liu, Xuequan Lu, Lizhuang Ma, Shuicheng Yan
Abstract: Domain Generalization (DG) has been recently explored to improve the generalizability of point cloud classification (PCC) models toward unseen domains. However, they often suffer from limited receptive fields or quadratic complexity due to the use of convolution neural networks or vision Transformers. In this paper, we present the first work that studies the generalizability of state space models (SSMs) in DG PCC and find that directly applying SSMs into DG PCC will encounter several challenges: the inherent topology of the point cloud tends to be disrupted and leads to noise accumulation during the serialization stage. Besides, the lack of designs in domain-agnostic feature learning and data scanning will introduce unanticipated domain-specific information into the 3D sequence data. To this end, we propose a novel framework, PointDGMamba, that excels in strong generalizability toward unseen domains and has the advantages of global receptive fields and efficient linear complexity. PointDGMamba consists of three innovative components: Masked Sequence Denoising (MSD), Sequence-wise Cross-domain Feature Aggregation (SCFA), and Dual-level Domain Scanning (DDS). In particular, MSD selectively masks out the noised point tokens of the point cloud sequences, SCFA introduces cross-domain but same-class point cloud features to encourage the model to learn how to extract more generalized features. DDS includes intra-domain scanning and cross-domain scanning to facilitate information exchange between features. In addition, we propose a new and more challenging benchmark PointDG-3to1 for multi-domain generalization. Extensive experiments demonstrate the effectiveness and state-of-the-art performance of our presented PointDGMamba.
Authors: G\"orkay Aydemir, Weidi Xie, Fatma G\"uney
Abstract: Large-scale vision foundation models have demonstrated remarkable success across various tasks, underscoring their robust generalization capabilities. While their proficiency in two-view correspondence has been explored, their effectiveness in long-term correspondence within complex environments remains unexplored. To address this, we evaluate the geometric awareness of visual foundation models in the context of point tracking: (i) in zero-shot settings, without any training; (ii) by probing with low-capacity layers; (iii) by fine-tuning with Low Rank Adaptation (LoRA). Our findings indicate that features from Stable Diffusion and DINOv2 exhibit superior geometric correspondence abilities in zero-shot settings. Furthermore, DINOv2 achieves performance comparable to supervised models in adaptation settings, demonstrating its potential as a strong initialization for correspondence learning.
Authors: Jinming Chai, Qin Ma, Junpei Zhang, Licheng Jiao, Fang Liu
Abstract: Video object segmentation is a challenging task that serves as the cornerstone of numerous downstream applications, including video editing and autonomous driving. In this technical report, we briefly introduce the solution of our team "yuanjie" for video object segmentation in the 6-th LSVOS Challenge VOS Track at ECCV 2024. We believe that our proposed CSS-Segment will perform better in videos of complex object motion and long-term presentation. In this report, we successfully validated the effectiveness of the CSS-Segment in video object segmentation. Finally, our method achieved a J\&F score of 80.84 in and test phases, and ultimately ranked 2nd in the 6-th LSVOS Challenge VOS Track at ECCV 2024.
Authors: Jianing Song, Nabil Aouf, Duarte Rondao, Christophe Honvault, Luis Mansilla
Abstract: The Lunar landing has drawn great interest in lunar exploration in recent years, and autonomous lunar landing navigation is fundamental to this task. AI is expected to play a critical role in autonomous and intelligent space missions, yet human experts question the reliability of AI solutions. Thus, the \gls{xai} for vision-based lunar landing is studied in this paper, aiming at providing transparent and understandable predictions for intelligent lunar landing. Attention-based Darknet53 is proposed as the feature extraction structure. For crater detection and navigation tasks, attention-based YOLOv3 and attention-Darknet53-LSTM are presented respectively. The experimental results show that the offered networks provide competitive performance on relative crater detection and pose estimation during the lunar landing. The explainability of the provided networks is achieved by introducing an attention mechanism into the network during model building. Moreover, the PCC is utilised to quantitively evaluate the explainability of the proposed networks, with the findings showing the functions of various convolutional layers in the network.
Authors: Sakhinana Sagar Srinivas, Geethan Sannidhi, Sreeja Gangasani, Chidaksh Ravuru, Venkataramana Runkana
Abstract: Characterizing materials using electron micrographs is crucial in areas such as semiconductors and quantum materials. Traditional classification methods falter due to the intricatestructures of these micrographs. This study introduces an innovative architecture that leverages the generative capabilities of zero-shot prompting in Large Language Models (LLMs) such as GPT-4(language only), the predictive ability of few-shot (in-context) learning in Large Multimodal Models (LMMs) such as GPT-4(V)ision, and fuses knowledge across image based and linguistic insights for accurate nanomaterial category prediction. This comprehensive approach aims to provide a robust solution for the automated nanomaterial identification task in semiconductor manufacturing, blending performance, efficiency, and interpretability. Our method surpasses conventional approaches, offering precise nanomaterial identification and facilitating high-throughput screening.
Authors: Yitong Yang, Yinglin Wang, Jing Wang, Tian Zhang
Abstract: Text-driven diffusion models have achieved remarkable success in image editing, but a crucial component in these models-text embeddings-has not been fully explored. The entanglement and opacity of text embeddings present significant challenges to achieving precise image editing. In this paper, we provide a comprehensive and in-depth analysis of text embeddings in Stable Diffusion XL, offering three key insights. First, while the 'aug_embedding' captures the full semantic content of the text, its contribution to the final image generation is relatively minor. Second, 'BOS' and 'Padding_embedding' do not contain any semantic information. Lastly, the 'EOS' holds the semantic information of all words and contains the most style features. Each word embedding plays a unique role without interfering with one another. Based on these insights, we propose a novel approach for controllable image editing using a free-text embedding control method called PSP (Prompt-Softbox-Prompt). PSP enables precise image editing by inserting or adding text embeddings within the cross-attention layers and using Softbox to define and control the specific area for semantic injection. This technique allows for obejct additions and replacements while preserving other areas of the image. Additionally, PSP can achieve style transfer by simply replacing text embeddings. Extensive experimental results show that PSP achieves significant results in tasks such as object replacement, object addition, and style transfer.
Authors: Bharatesh Chakravarthi, Aayush Atul Verma, Kostas Daniilidis, Cornelia Fermuller
Abstract: Event-based vision, inspired by the human visual system, offers transformative capabilities such as low latency, high dynamic range, and reduced power consumption. This paper presents a comprehensive survey of event cameras, tracing their evolution over time. It introduces the fundamental principles of event cameras, compares them with traditional frame cameras, and highlights their unique characteristics and operational differences. The survey covers various event camera models from leading manufacturers, key technological milestones, and influential research contributions. It explores diverse application areas across different domains and discusses essential real-world and synthetic datasets for research advancement. Additionally, the role of event camera simulators in testing and development is discussed. This survey aims to consolidate the current state of event cameras and inspire further innovation in this rapidly evolving field. To support the research community, a \href{https://github.com/chakravarthi589/Event-based-Vision_Resources}{GitHub} page categorizes past and future research articles and consolidates valuable resources.
URLs: https://github.com/chakravarthi589/Event-based-Vision_Resources
Authors: Johannes H\"agerlind, Jonas Hentati-Sundberg, Bastian Wandt
Abstract: This paper deals with 3D reconstruction of seabirds which recently came into focus of environmental scientists as valuable bio-indicators for environmental change. Such 3D information is beneficial for analyzing the bird's behavior and physiological shape, for example by tracking motion, shape, and appearance changes. From a computer vision perspective birds are especially challenging due to their rapid and oftentimes non-rigid motions. We propose an approach to reconstruct the 3D pose and shape from monocular videos of a specific breed of seabird - the common murre. Our approach comprises a full pipeline of detection, tracking, segmentation, and temporally consistent 3D reconstruction. Additionally, we propose a temporal loss that extends current single-image 3D bird pose estimators to the temporal domain. Moreover, we provide a real-world dataset of 10000 frames of video observations on average capture nine birds simultaneously, comprising a large variety of motions and interactions, including a smaller test set with bird-specific keypoint labels. Using our temporal optimization, we achieve state-of-the-art performance for the challenging sequences in our dataset.
Authors: Ameer Majeed, Hossein Hassani
Abstract: Many languages have vast amounts of handwritten texts, such as ancient scripts about folktale stories and historical narratives or contemporary documents and letters. Digitization of those texts has various applications, such as daily tasks, cultural studies, and historical research. Syriac is an ancient, endangered, and low-resourced language that has not received the attention it requires and deserves. This paper reports on a research project aimed at developing a optical character recognition (OCR) model based on the handwritten Syriac texts as a starting point to build more digital services for this endangered language. A dataset was created, KHAMIS (inspired by the East Syriac poet, Khamis bar Qardahe), which consists of handwritten sentences in the East Syriac script. We used it to fine-tune the Tesseract-OCR engine's pretrained Syriac model on handwritten data. The data was collected from volunteers capable of reading and writing in the language to create KHAMIS. KHAMIS currently consists of 624 handwritten Syriac sentences collected from 31 university students and one professor, and it will be partially available online and the whole dataset available in the near future for development and research purposes. As a result, the handwritten OCR model was able to achieve a character error rate of 1.097-1.610% and 8.963-10.490% on both training and evaluation sets, respectively, and both a character error rate of 18.89-19.71% and a word error rate of 62.83-65.42% when evaluated on the test set, which is twice as better than the default Syriac model of Tesseract.
Authors: Lukas Picek, Klara Janouskova, Milan Sulc, Jiri Matas
Abstract: We introduce a new, highly challenging benchmark and a dataset -- FungiTastic -- based on data continuously collected over a twenty-year span. The dataset originates in fungal records labeled and curated by experts. It consists of about 350k multi-modal observations that include more than 650k photographs from 5k fine-grained categories and diverse accompanying information, e.g., acquisition metadata, satellite images, and body part segmentation. FungiTastic is the only benchmark that includes a test set with partially DNA-sequenced ground truth of unprecedented label reliability. The benchmark is designed to support (i) standard close-set classification, (ii) open-set classification, (iii) multi-modal classification, (iv) few-shot learning, (v) domain shift, and many more. We provide baseline methods tailored for almost all the use-cases. We provide a multitude of ready-to-use pre-trained models on HuggingFace and a framework for model training. A comprehensive documentation describing the dataset features and the baselines are available at https://bohemianvra.github.io/FungiTastic/ and https://www.kaggle.com/datasets/picekl/fungitastic.
URLs: https://bohemianvra.github.io/FungiTastic/, https://www.kaggle.com/datasets/picekl/fungitastic.
Authors: Jing Yuan, Tania Stathaki
Abstract: Scribble supervision, a common form of weakly supervised learning, involves annotating pixels using hand-drawn curve lines, which helps reduce the cost of manual labelling. This technique has been widely used in medical image segmentation tasks to fasten network training. However, scribble supervision has limitations in terms of annotation consistency across samples and the availability of comprehensive groundtruth information. Additionally, it often grapples with the challenge of accommodating varying scale targets, particularly in the context of medical images. In this paper, we propose three novel methods to overcome these challenges, namely, 1) the cross-shape scribble annotation method; 2) the pseudo mask method based on cross shapes; and 3) the size-aware multi-branch method. The parameter and structure design are investigated in depth. Experimental results show that the proposed methods have achieved significant improvement in mDice scores across multiple polyp datasets. Notably, the combination of these methods outperforms the performance of state-of-the-art scribble supervision methods designed for medical image segmentation.
Authors: Nada Osman, Marwan Torki
Abstract: Anomaly action detection and localization play an essential role in security and advanced surveillance systems. However, due to the tremendous amount of surveillance videos, most of the available data for the task is unlabeled or semi-labeled with the video class known, but the location of the anomaly event is unknown. In this work, we target anomaly localization in semi-supervised videos. While the mainstream direction in addressing this task is focused on segment-level multi-instance learning and the generation of pseudo labels, we aim to explore a promising yet unfulfilled direction to solve the problem by learning the temporal relations within videos in order to locate anomaly events. To this end, we propose a hierarchical transformer model designed to evaluate the significance of observed actions in anomalous videos with a divide-and-conquer strategy along the temporal axis. Our approach segments a parent video hierarchically into multiple temporal children instances and measures the influence of the children nodes in classifying the abnormality of the parent video. Evaluating our model on two well-known anomaly detection datasets, UCF-crime and ShanghaiTech, proves its ability to interpret the observed actions within videos and localize the anomalous ones. Our proposed approach outperforms previous works relying on segment-level multiple-instance learning approaches while reaching a promising performance compared to the more recent pseudo-labeling-based approaches.
Authors: Jing Yuan, Tania Stathaki, Guangyu Ren
Abstract: The design of pedestrian detectors seldom considers the unique characteristics of this task and usually follows the common strategies for general object detection. To explore the potential of these characteristics, we take the perspective effect in pedestrian datasets as an example and propose the mean height aided suppression for post-processing. This method rejects predictions that fall at levels with a low possibility of containing any pedestrians or that have an abnormal height compared to the average. To achieve this, the existence score and mean height generators are proposed. Comprehensive experiments on various datasets and detectors are performed; the choice of hyper-parameters is discussed in depth. The proposed method is easy to implement and is plug-and-play. Results show that the proposed methods significantly improve detection accuracy when applied to different existing pedestrian detectors and datasets. The combination of mean height aided suppression with particular detectors outperforms state-of-the-art pedestrian detectors on Caltech and Citypersons datasets.
Authors: Sakhinana Sagar Srinivas, Geethan Sannidhi, Venkataramana Runkana
Abstract: Characterizing materials with electron micrographs is a crucial task in fields such as semiconductors and quantum materials. The complex hierarchical structure of micrographs often poses challenges for traditional classification methods. In this study, we propose an innovative backbone architecture for analyzing electron micrographs. We create multi-modal representations of the micrographs by tokenizing them into patch sequences and, additionally, representing them as vision graphs, commonly referred to as patch attributed graphs. We introduce the Hierarchical Network Fusion (HNF), a multi-layered network structure architecture that facilitates information exchange between the multi-modal representations and knowledge integration across different patch resolutions. Furthermore, we leverage large language models (LLMs) to generate detailed technical descriptions of nanomaterials as auxiliary information to assist in the downstream task. We utilize a cross-modal attention mechanism for knowledge fusion across cross-domain representations(both image-based and linguistic insights) to predict the nanomaterial category. This multi-faceted approach promises a more comprehensive and accurate representation and classification of micrographs for nanomaterial identification. Our framework outperforms traditional methods, overcoming challenges posed by distributional shifts, and facilitating high-throughput screening.
Authors: Keqiang Sun, Amin Jourabloo, Riddhish Bhalodia, Moustafa Meshry, Yu Rong, Zhengyu Yang, Thu Nguyen-Phuoc, Christian Haene, Jiu Xu, Sam Johnson, Hongsheng Li, Sofien Bouaziz
Abstract: Photo-realistic and controllable 3D avatars are crucial for various applications such as virtual and mixed reality (VR/MR), telepresence, gaming, and film production. Traditional methods for avatar creation often involve time-consuming scanning and reconstruction processes for each avatar, which limits their scalability. Furthermore, these methods do not offer the flexibility to sample new identities or modify existing ones. On the other hand, by learning a strong prior from data, generative models provide a promising alternative to traditional reconstruction methods, easing the time constraints for both data capture and processing. Additionally, generative methods enable downstream applications beyond reconstruction, such as editing and stylization. Nonetheless, the research on generative 3D avatars is still in its infancy, and therefore current methods still have limitations such as creating static avatars, lacking photo-realism, having incomplete facial details, or having limited drivability. To address this, we propose a text-conditioned generative model that can generate photo-realistic facial avatars of diverse identities, with more complete details like hair, eyes and mouth interior, and which can be driven through a powerful non-parametric latent expression space. Specifically, we integrate the generative and editing capabilities of latent diffusion models with a strong prior model for avatar expression driving. Our model can generate and control high-fidelity avatars, even those out-of-distribution. We also highlight its potential for downstream applications, including avatar editing and single-shot avatar reconstruction.
Authors: George Tang, William Zhao, Logan Ford, David Benhaim, Paul Zhang
Abstract: We propose Segment Any Mesh (SAMesh), a novel zero-shot method for mesh part segmentation that overcomes the limitations of shape analysis-based, learning-based, and current zero-shot approaches. SAMesh operates in two phases: multimodal rendering and 2D-to-3D lifting. In the first phase, multiview renders of the mesh are individually processed through Segment Anything 2 (SAM2) to generate 2D masks. These masks are then lifted into a mesh part segmentation by associating masks that refer to the same mesh part across the multiview renders. We find that applying SAM2 to multimodal feature renders of normals and shape diameter scalars achieves better results than using only untextured renders of meshes. By building our method on top of SAM2, we seamlessly inherit any future improvements made to 2D segmentation. We compare our method with a robust, well-evaluated shape analysis method, Shape Diameter Function (ShapeDiam), and show our method is comparable to or exceeds its performance. Since current benchmarks contain limited object diversity, we also curate and release a dataset of generated meshes and use it to demonstrate our method's improved generalization over ShapeDiam via human evaluation. We release the code and dataset at https://github.com/gtangg12/samesh
Authors: Yingjian Chen, Lei Zhang, Yakun Niu, Pei Chen, Lei Tan, Jing Zhou
Abstract: The rise of generative models has sparked concerns about image authenticity online, highlighting the urgent need for an effective and general detector. Recent methods leveraging the frozen pre-trained CLIP-ViT model have made great progress in deepfake detection. However, these models often rely on visual-general features directly extracted by the frozen network, which contain excessive information irrelevant to the task, resulting in limited detection performance. To address this limitation, in this paper, we propose an efficient Guided and Fused Frozen CLIP-ViT (GFF), which integrates two simple yet effective modules. The Deepfake-Specific Feature Guidance Module (DFGM) guides the frozen pre-trained model in extracting features specifically for deepfake detection, reducing irrelevant information while preserving its generalization capabilities. The Multi-Stage Fusion Module (FuseFormer) captures low-level and high-level information by fusing features extracted from each stage of the ViT. This dual-module approach significantly improves deepfake detection by fully leveraging CLIP-ViT's inherent advantages. Extensive experiments demonstrate the effectiveness and generalization ability of GFF, which achieves state-of-the-art performance with optimal results in only 5 training epochs. Even when trained on only 4 classes of ProGAN, GFF achieves nearly 99% accuracy on unseen GANs and maintains an impressive 97% accuracy on unseen diffusion models.
Authors: Lanhu Wu, Miao Zhang, Yongri Piao, Zhenyan Yao, Weibing Sun, Feng Tian, Huchuan Lu
Abstract: Automatic and precise medical image segmentation (MIS) is of vital importance for clinical diagnosis and analysis. Current MIS methods mainly rely on the convolutional neural network (CNN) or self-attention mechanism (Transformer) for feature modeling. However, CNN-based methods suffer from the inaccurate localization owing to the limited global dependency while Transformer-based methods always present the coarse boundary for the lack of local emphasis. Although some CNN-Transformer hybrid methods are designed to synthesize the complementary local and global information for better performance, the combination of CNN and Transformer introduces numerous parameters and increases the computation cost. To this end, this paper proposes a CNN-Transformer rectified collaborative learning (CTRCL) framework to learn stronger CNN-based and Transformer-based models for MIS tasks via the bi-directional knowledge transfer between them. Specifically, we propose a rectified logit-wise collaborative learning (RLCL) strategy which introduces the ground truth to adaptively select and rectify the wrong regions in student soft labels for accurate knowledge transfer in the logit space. We also propose a class-aware feature-wise collaborative learning (CFCL) strategy to achieve effective knowledge transfer between CNN-based and Transformer-based models in the feature space by granting their intermediate features the similar capability of category perception. Extensive experiments on three popular MIS benchmarks demonstrate that our CTRCL outperforms most state-of-the-art collaborative learning methods under different evaluation metrics.
Authors: Cho-Ying Wu, Quankai Gao, Chin-Cheng Hsu, Te-Lin Wu, Jing-Wen Chen, Ulrich Neumann
Abstract: Indoor monocular depth estimation helps home automation, including robot navigation or AR/VR for surrounding perception. Most previous methods primarily experiment with the NYUv2 Dataset and concentrate on the overall performance in their evaluation. However, their robustness and generalization to diversely unseen types or categories for indoor spaces (spaces types) have yet to be discovered. Researchers may empirically find degraded performance in a released pretrained model on custom data or less-frequent types. This paper studies the common but easily overlooked factor-space type and realizes a model's performance variances across spaces. We present InSpaceType Dataset, a high-quality RGBD dataset for general indoor scenes, and benchmark 13 recent state-of-the-art methods on InSpaceType. Our examination shows that most of them suffer from performance imbalance between head and tailed types, and some top methods are even more severe. The work reveals and analyzes underlying bias in detail for transparency and robustness. We extend the analysis to a total of 4 datasets and discuss the best practice in synthetic data curation for training indoor monocular depth. Further, dataset ablation is conducted to find out the key factor in generalization. This work marks the first in-depth investigation of performance variances across space types and, more importantly, releases useful tools, including datasets and codes, to closely examine your pretrained depth models. Data and code: https://depthcomputation.github.io/DepthPublic/
Authors: Wenrui Li, Yapeng Mi, Fucheng Cai, Zhe Yang, Wangmeng Zuo, Xingtao Wang, Xiaopeng Fan
Abstract: Text-driven 3D scene generation has seen significant advancements recently. However, most existing methods generate single-view images using generative models and then stitch them together in 3D space. This independent generation for each view often results in spatial inconsistency and implausibility in the 3D scenes. To address this challenge, we proposed a novel text-driven 3D-consistent scene generation model: SceneDreamer360. Our proposed method leverages a text-driven panoramic image generation model as a prior for 3D scene generation and employs 3D Gaussian Splatting (3DGS) to ensure consistency across multi-view panoramic images. Specifically, SceneDreamer360 enhances the fine-tuned Panfusion generator with a three-stage panoramic enhancement, enabling the generation of high-resolution, detail-rich panoramic images. During the 3D scene construction, a novel point cloud fusion initialization method is used, producing higher quality and spatially consistent point clouds. Our extensive experiments demonstrate that compared to other methods, SceneDreamer360 with its panoramic image generation and 3DGS can produce higher quality, spatially consistent, and visually appealing 3D scenes from any text prompt. Our codes are available at \url{https://github.com/liwrui/SceneDreamer360}.
Authors: Wenrui Li, Wei Han, Yandu Chen, Yeyu Chai, Yidan Lu, Xingtao Wang, Xiaopeng Fan
Abstract: Due to the challenges in acquiring paired Text-3D data and the inherent irregularity of 3D data structures, combined representation learning of 3D point clouds and text remains unexplored. In this paper, we propose a novel Riemann-based Multi-scale Attention Reasoning Network (RMARN) for text-3D retrieval. Specifically, the extracted text and point cloud features are refined by their respective Adaptive Feature Refiner (AFR). Furthermore, we introduce the innovative Riemann Local Similarity (RLS) module and the Global Pooling Similarity (GPS) module. However, as 3D point cloud data and text data often possess complex geometric structures in high-dimensional space, the proposed RLS employs a novel Riemann Attention Mechanism to reflect the intrinsic geometric relationships of the data. Without explicitly defining the manifold, RMARN learns the manifold parameters to better represent the distances between text-point cloud samples. To address the challenges of lacking paired text-3D data, we have created the large-scale Text-3D Retrieval dataset T3DR-HIT, which comprises over 3,380 pairs of text and point cloud data. T3DR-HIT contains coarse-grained indoor 3D scenes and fine-grained Chinese artifact scenes, consisting of 1,380 and over 2,000 text-3D pairs, respectively. Experiments on our custom datasets demonstrate the superior performance of the proposed method. Our code and proposed datasets are available at \url{https://github.com/liwrui/RMARN}.
Authors: Jack Saunders, Vinay Namboodiri
Abstract: Speech-driven facial animation is important for many applications including TV, film, video games, telecommunication and AR/VR. Recently, transformers have been shown to be extremely effective for this task. However, we identify two issues with the existing transformer-based models. Firstly, they are difficult to adapt to new personalised speaking styles and secondly, they are slow to run for long sentences due to the quadratic complexity of the transformer. We propose TalkLoRA to address both of these issues. TalkLoRA uses Low-Rank Adaptation to effectively and efficiently adapt to new speaking styles, even with limited data. It does this by training an adaptor with a small number of parameters for each subject. We also utilise a chunking strategy to reduce the complexity of the underlying transformer, allowing for long sentences at inference time. TalkLoRA can be applied to any transformer-based speech-driven animation method. We perform extensive experiments to show that TalkLoRA archives state-of-the-art style adaptation and that it allows for an order-of-complexity reduction in inference times without sacrificing quality. We also investigate and provide insights into the hyperparameter selection for LoRA fine-tuning of speech-driven facial animation models.
Authors: Abu Saleh Musa Miah, Najmul Hassan, Md. Maniruzzaman, Nobuyoshi Asai, Jungpil Shin
Abstract: Surface electromyography (EMG) serves as a pivotal tool in hand gesture recognition and human-computer interaction, offering a non-invasive means of signal acquisition. This study presents a novel methodology for classifying hand gestures using EMG signals. To address the challenges associated with feature extraction where, we explored 23 distinct morphological, time domain and frequency domain feature extraction techniques. However, the substantial size of the features may increase the computational complexity issues that can hinder machine learning algorithm performance. We employ an efficient feature selection approach, specifically an extra tree classifier, to mitigate this. The selected potential feature fed into the various machine learning-based classification algorithms where our model achieved 97.43\% accuracy with the KNN algorithm and selected feature. By leveraging a comprehensive feature extraction and selection strategy, our methodology enhances the accuracy and usability of EMG-based hand gesture recognition systems. The higher performance accuracy proves the effectiveness of the proposed model over the existing system. \keywords{EMG signal, machine learning approach, hand gesture recognition.
Authors: Rundong Luo, Haoran Geng, Congyue Deng, Puhao Li, Zan Wang, Baoxiong Jia, Leonidas Guibas, Siyuang Huang
Abstract: Interactable objects are ubiquitous in our daily lives. Recent advances in 3D generative models make it possible to automate the modeling of these objects, benefiting a range of applications from 3D printing to the creation of robot simulation environments. However, while significant progress has been made in modeling 3D shapes and appearances, modeling object physics, particularly for interactable objects, remains challenging due to the physical constraints imposed by inter-part motions. In this paper, we tackle the problem of physically plausible part completion for interactable objects, aiming to generate 3D parts that not only fit precisely into the object but also allow smooth part motions. To this end, we propose a diffusion-based part generation model that utilizes geometric conditioning through classifier-free guidance and formulates physical constraints as a set of stability and mobility losses to guide the sampling process. Additionally, we demonstrate the generation of dependent parts, paving the way toward sequential part generation for objects with complex part-whole hierarchies. Experimentally, we introduce a new metric for measuring physical plausibility based on motion success rates. Our model outperforms existing baselines over shape and physical metrics, especially those that do not adequately model physical constraints. We also demonstrate our applications in 3D printing, robot manipulation, and sequential part generation, showing our strength in realistic tasks with the demand for high physical plausibility.
Authors: Haizhao Jing, Liuwei Wan, Xizhe Xue, Haokui Zhang, Ying Li
Abstract: Recently, the Vision Transformer (ViT) model has replaced the classical Convolutional Neural Network (ConvNet) in various computer vision tasks due to its superior performance. Even in hyperspectral image (HSI) classification field, ViT-based methods also show promising potential. Nevertheless, ViT encounters notable difficulties in processing HSI data. Its self-attention mechanism, which exhibits quadratic complexity, escalates computational costs. Additionally, ViT's substantial demand for training samples does not align with the practical constraints posed by the expensive labeling of HSI data. To overcome these challenges, we propose a 3D relational ConvNet named 3D-RCNet, which inherits both strengths of ConvNet and ViT, resulting in high performance in HSI classification. We embed the self-attention mechanism of Transformer into the convolutional operation of ConvNet to design 3D relational convolutional operation and use it to build the final 3D-RCNet. The proposed 3D-RCNet maintains the high computational efficiency of ConvNet while enjoying the flexibility of ViT. Additionally, the proposed 3D relational convolutional operation is a plug-and-play operation, which can be inserted into previous ConvNet-based HSI classification methods seamlessly. Empirical evaluations on three representative benchmark HSI datasets show that the proposed model outperforms previous ConvNet-based and ViT-based HSI approaches.
Authors: Chaowei Chen, Li Yu, Shiquan Min, Shunfang Wang
Abstract: State Space Models (SSMs), especially Mamba, have shown great promise in medical image segmentation due to their ability to model long-range dependencies with linear computational complexity. However, accurate medical image segmentation requires the effective learning of both multi-scale detailed feature representations and global contextual dependencies. Although existing works have attempted to address this issue by integrating CNNs and SSMs to leverage their respective strengths, they have not designed specialized modules to effectively capture multi-scale feature representations, nor have they adequately addressed the directional sensitivity problem when applying Mamba to 2D image data. To overcome these limitations, we propose a Multi-Scale Vision Mamba UNet model for medical image segmentation, termed MSVM-UNet. Specifically, by introducing multi-scale convolutions in the VSS blocks, we can more effectively capture and aggregate multi-scale feature representations from the hierarchical features of the VMamba encoder and better handle 2D visual data. Additionally, the large kernel patch expanding (LKPE) layers achieve more efficient upsampling of feature maps by simultaneously integrating spatial and channel information. Extensive experiments on the Synapse and ACDC datasets demonstrate that our approach is more effective than some state-of-the-art methods in capturing and aggregating multi-scale feature representations and modeling long-range dependencies between pixels.
Authors: Xu Zhang, Zhipeng Xie, Haiyang Yu, Qitong Wang, Peng Wang, Wei Wang
Abstract: Handling varying computational resources is a critical issue in modern AI applications. Adaptive deep networks, featuring the dynamic employment of multiple classifier heads among different layers, have been proposed to address classification tasks under varying computing resources. Existing approaches typically utilize the last classifier supported by the available resources for inference, as they believe that the last classifier always performs better across all classes. However, our findings indicate that earlier classifier heads can outperform the last head for certain classes. Based on this observation, we introduce the Collaborative Decision Making (CDM) module, which fuses the multiple classifier heads to enhance the inference performance of adaptive deep networks. CDM incorporates an uncertainty-aware fusion method based on evidential deep learning (EDL), that utilizes the reliability (uncertainty values) from the first c-1 classifiers to improve the c-th classifier' accuracy. We also design a balance term that reduces fusion saturation and unfairness issues caused by EDL constraints to improve the fusion quality of CDM. Finally, a regularized training strategy that uses the last classifier to guide the learning process of early classifiers is proposed to further enhance the CDM module's effect, called the Guided Collaborative Decision Making (GCDM) framework. The experimental evaluation demonstrates the effectiveness of our approaches. Results on ImageNet datasets show CDM and GCDM obtain 0.4% to 2.8% accuracy improvement (under varying computing resources) on popular adaptive networks. The code is available at the link https://github.com/Meteor-Stars/GCDM_AdaptiveNet.
Authors: Zhaoyang Li, Yuan Wang, Wangkai Li, Rui Sun, Tianzhu Zhang
Abstract: Point cloud few-shot semantic segmentation (PC-FSS) aims to segment targets of novel categories in a given query point cloud with only a few annotated support samples. The current top-performing prototypical learning methods employ prototypes originating from support samples to direct the classification of query points. However, the inherent fragility of point-level matching and the prevalent intra-class diversity pose great challenges to this cross-instance matching paradigm, leading to erroneous background activations or incomplete target excavation. In this work, we propose a simple yet effective framework in the spirit of Decoupled Localization and Expansion (DLE). The proposed DLE, including a structural localization module (SLM) and a self-expansion module (SEM), enjoys several merits. First, structural information is injected into the matching process through the agent-level correlation in SLM, and the confident target region can thus be precisely located. Second, more reliable intra-object similarity is harnessed in SEM to derive the complete target, and the conservative expansion strategy is introduced to reasonably constrain the expansion. Extensive experiments on two challenging benchmarks under different settings demonstrate that DLE outperforms previous state-of-the-art approaches by large margins.
Authors: Jayakanth Kunhoth, Somaya Al-Maadeed, Moutaz Saleh, Younes Akbari
Abstract: Developmental dysgraphia is a neurological disorder that hinders children's writing skills. In recent years, researchers have increasingly explored machine learning methods to support the diagnosis of dysgraphia based on offline and online handwriting. In most previous studies, the two types of handwriting have been analysed separately, which does not necessarily lead to promising results. In this way, the relationship between online and offline data cannot be explored. To address this limitation, we propose a novel multimodal machine learning approach utilizing both online and offline handwriting data. We created a new dataset by transforming an existing online handwritten dataset, generating corresponding offline handwriting images. We considered only different types of word data (simple word, pseudoword & difficult word) in our multimodal analysis. We trained SVM and XGBoost classifiers separately on online and offline features as well as implemented multimodal feature fusion and soft-voted ensemble. Furthermore, we proposed a novel ensemble with conditional feature fusion method which intelligently combines predictions from online and offline classifiers, selectively incorporating feature fusion when confidence scores fall below a threshold. Our novel approach achieves an accuracy of 88.8%, outperforming SVMs for single modalities by 12-14%, existing methods by 8-9%, and traditional multimodal approaches (soft-vote ensemble and feature fusion) by 3% and 5%, respectively. Our methodology contributes to the development of accurate and efficient dysgraphia diagnosis tools, requiring only a single instance of multimodal word/pseudoword data to determine the handwriting impairment. This work highlights the potential of multimodal learning in enhancing dysgraphia diagnosis, paving the way for accessible and practical diagnostic tools.
Authors: Shi Hezi, Jiang Luo, Zheng Jianmin, Zeng Jun
Abstract: This paper addresses the challenges of designing mesh convolution neural networks for 3D mesh dense prediction. While deep learning has achieved remarkable success in image dense prediction tasks, directly applying or extending these methods to irregular graph data, such as 3D surface meshes, is nontrivial due to the non-uniform element distribution and irregular connectivity in surface meshes which make it difficult to adapt downsampling, upsampling, and convolution operations. In addition, commonly used multiresolution networks require repeated high-to-low and then low-to-high processes to boost the performance of recovering rich, high-resolution representations. To address these challenges, this paper proposes a self-parameterization-based multi-resolution convolution network that extends existing image dense prediction architectures to 3D meshes. The novelty of our approach lies in two key aspects. First, we construct a multi-resolution mesh pyramid directly from the high-resolution input data and propose area-aware mesh downsampling/upsampling operations that use sequential bijective inter-surface mappings between different mesh resolutions. The inter-surface mapping redefines the mesh, rather than reshaping it, which thus avoids introducing unnecessary errors. Second, we maintain the high-resolution representation in the multi-resolution convolution network, enabling multi-scale fusions to exchange information across parallel multi-resolution subnetworks, rather than through connections of high-to-low resolution subnetworks in series. These features differentiate our approach from most existing mesh convolution networks and enable more accurate mesh dense predictions, which is confirmed in experiments.
Authors: Fengshun Wang, Qiurui Wang, Yuting Wang
Abstract: The present few-shot temporal action localization model can't handle the situation where videos contain multiple action instances. So the purpose of this paper is to achieve manifold action instances localization in a lengthy untrimmed query video using limited trimmed support videos. To address this challenging problem effectively, we proposed a novel solution involving a spatial-channel relation transformer with probability learning and cluster refinement. This method can accurately identify the start and end boundaries of actions in the query video, utilizing only a limited number of labeled videos. Our proposed method is adept at capturing both temporal and spatial contexts to effectively classify and precisely locate actions in videos, enabling a more comprehensive utilization of these crucial details. The selective cosine penalization algorithm is designed to suppress temporal boundaries that do not include action scene switches. The probability learning combined with the label generation algorithm alleviates the problem of action duration diversity and enhances the model's ability to handle fuzzy action boundaries. The interval cluster can help us get the final results with multiple instances situations in few-shot temporal action localization. Our model achieves competitive performance through meticulous experimentation utilizing the benchmark datasets ActivityNet1.3 and THUMOS14. Our code is readily available at https://github.com/ycwfs/FMI-TAL.
Authors: Miguel Tjia, Artem Kim, Elaine Wynette Wijaya, Hanna Tefara, Kevin Zhu
Abstract: 7,651 cases of Search and Rescue Missions (SAR) were reported by the United States Coast Guard in 2024, with over 1322 SAR helicopters deployed in the 6 first months alone. Through the utilizations of YOLO, we were able to run different weather conditions and lighting from our augmented dataset for training. YOLO then utilizes CNNs to apply a series of convolutions and pooling layers to the input image, where the convolution layers are able to extract the main features of the image. Through this, our YOLO model is able to learn to differentiate different objects which may considerably improve its accuracy, possibly enhancing the efficiency of SAR operations through enhanced detection accuracy. This paper aims to improve the model's accuracy of human detection in maritime SAR by evaluating a robust datasets containing various elevations and geological locations, as well as through data augmentation which simulates different weather and lighting. We observed that models trained on augmented datasets outperformed their non-augmented counterparts in which the human recall scores ranged from 0.891 to 0.911 with an improvement rate of 3.4\% on the YOLOv5l model. Results showed that these models demonstrate greater robustness to real-world conditions in varying of weather, brightness, tint, and contrast.
Authors: Chuanrui Zhang, Yingshuang Zou, Zhuoling Li, Minmin Yi, Haoqian Wang
Abstract: Compared with previous 3D reconstruction methods like Nerf, recent Generalizable 3D Gaussian Splatting (G-3DGS) methods demonstrate impressive efficiency even in the sparse-view setting. However, the promising reconstruction performance of existing G-3DGS methods relies heavily on accurate multi-view feature matching, which is quite challenging. Especially for the scenes that have many non-overlapping areas between various views and contain numerous similar regions, the matching performance of existing methods is poor and the reconstruction precision is limited. To address this problem, we develop a strategy that utilizes a predicted depth confidence map to guide accurate local feature matching. In addition, we propose to utilize the knowledge of existing monocular depth estimation models as prior to boost the depth estimation precision in non-overlapping areas between views. Combining the proposed strategies, we present a novel G-3DGS method named TranSplat, which obtains the best performance on both the RealEstate10K and ACID benchmarks while maintaining competitive speed and presenting strong cross-dataset generalization ability. Our code, and demos will be available at: https://xingyoujun.github.io/transplat.
Authors: Xin Zhang, Teodor Boyadzhiev, Jinglei Shi, Jufeng Yang
Abstract: In this paper, we leverage image complexity as a prior for refining segmentation features to achieve accurate real-time semantic segmentation. The design philosophy is based on the observation that different pixel regions within an image exhibit varying levels of complexity, with higher complexities posing a greater challenge for accurate segmentation. We thus introduce image complexity as prior guidance and propose the Image Complexity prior-guided Feature Refinement Network (ICFRNet). This network aggregates both complexity and segmentation features to produce an attention map for refining segmentation features within an Image Complexity Guided Attention (ICGA) module. We optimize the network in terms of both segmentation and image complexity prediction tasks with a combined loss function. Experimental results on the Cityscapes and CamViD datasets have shown that our ICFRNet achieves higher accuracy with a competitive efficiency for real-time segmentation.
Authors: Fares Bougourzi, Fadi Dornaika, Chongsheng Zhang
Abstract: Text recognition in the wild is an important technique for digital maps and urban scene understanding, in which the natural resembling properties between glyphs is one of the major reasons that lead to wrong recognition results. To address this challenge, we introduce two extremely fine-grained visual recognition benchmark datasets that contain very challenging resembling glyphs (characters/letters) in the wild to be distinguished. Moreover, we propose a simple yet effective two-stage contrastive learning approach to the extremely fine-grained recognition task of resembling glyphs discrimination. In the first stage, we utilize supervised contrastive learning to leverage label information to warm-up the backbone network. In the second stage, we introduce CCFG-Net, a network architecture that integrates classification and contrastive learning in both Euclidean and Angular spaces, in which contrastive learning is applied in both supervised learning and pairwise discrimination manners to enhance the model's feature representation capability. Overall, our proposed approach effectively exploits the complementary strengths of contrastive learning and classification, leading to improved recognition performance on the resembling glyphs. Comparative evaluations with state-of-the-art fine-grained classification approaches under both Convolutional Neural Network (CNN) and Transformer backbones demonstrate the superiority of our proposed method.
Authors: Jia-Run Du, Kun-Yu Lin, Jingke Meng, Wei-Shi Zheng
Abstract: To address the zero-shot temporal action localization (ZSTAL) task, existing works develop models that are generalizable to detect and classify actions from unseen categories. They typically develop a category-agnostic action detector and combine it with the Contrastive Language-Image Pre-training (CLIP) model to solve ZSTAL. However, these methods suffer from incomplete action proposals generated for \textit{unseen} categories, since they follow a frame-level prediction paradigm and require hand-crafted post-processing to generate action proposals. To address this problem, in this work, we propose a novel model named Generalizable Action Proposal generator (GAP), which can interface seamlessly with CLIP and generate action proposals in a holistic way. Our GAP is built in a query-based architecture and trained with a proposal-level objective, enabling it to estimate proposal completeness and eliminate the hand-crafted post-processing. Based on this architecture, we propose an Action-aware Discrimination loss to enhance the category-agnostic dynamic information of actions. Besides, we introduce a Static-Dynamic Rectifying module that incorporates the generalizable static information from CLIP to refine the predicted proposals, which improves proposal completeness in a generalizable manner. Our experiments show that our GAP achieves state-of-the-art performance on two challenging ZSTAL benchmarks, i.e., Thumos14 and ActivityNet1.3. Specifically, our model obtains significant performance improvement over previous works on the two benchmarks, i.e., +3.2% and +3.4% average mAP, respectively.
Authors: Anmol Manjunath, Viola Negroni, Sara Mandelli, Daniel Moreira, Paolo Bestagini
Abstract: Recent breakthroughs in deep learning and generative systems have significantly fostered the creation of synthetic media, as well as the local alteration of real content via the insertion of highly realistic synthetic manipulations. Local image manipulation, in particular, poses serious challenges to the integrity of digital content and societal trust. This problem is not only confined to multimedia data, but also extends to biological images included in scientific publications, like images depicting Western blots. In this work, we address the task of localizing synthetic manipulations in Western blot images. To discriminate between pristine and synthetic pixels of an analyzed image, we propose a synthetic detector that operates on small patches extracted from the image. We aggregate patch contributions to estimate a tampering heatmap, highlighting synthetic pixels out of pristine ones. Our methodology proves effective when tested over two manipulated Western blot image datasets, one altered automatically and the other manually by exploiting advanced AI-based image manipulation tools that are unknown at our training stage. We also explore the robustness of our method over an external dataset of other scientific images depicting different semantics, manipulated through unseen generation techniques.
Authors: Shichao Dong, Ze Yang, Guosheng Lin
Abstract: Data augmentation plays a crucial role in deep learning, enhancing the generalization and robustness of learning-based models. Standard approaches involve simple transformations like rotations and flips for generating extra data. However, these augmentations are limited by their initial dataset, lacking high-level diversity. Recently, large models such as language models and diffusion models have shown exceptional capabilities in perception and content generation. In this work, we propose a new paradigm to automatically generate 3D labeled training data by harnessing the power of pretrained large foundation models. For each target semantic class, we first generate 2D images of a single object in various structure and appearance via diffusion models and chatGPT generated text prompts. Beyond texture augmentation, we propose a method to automatically alter the shape of objects within 2D images. Subsequently, we transform these augmented images into 3D objects and construct virtual scenes by random composition. This method can automatically produce a substantial amount of 3D scene data without the need of real data, providing significant benefits in addressing few-shot learning challenges and mitigating long-tailed class imbalances. By providing a flexible augmentation approach, our work contributes to enhancing 3D data diversity and advancing model capabilities in scene understanding tasks.
Authors: Xiaoyu Tang, Zeyu Chen, Jintao Cheng, Xieyuanli Chen, Jin Wu, Bohuan Xue
Abstract: In autonomous driving, accurately distinguishing between static and moving objects is crucial for the autonomous driving system. When performing the motion object segmentation (MOS) task, effectively leveraging motion information from objects becomes a primary challenge in improving the recognition of moving objects. Previous methods either utilized range view (RV) or bird's eye view (BEV) residual maps to capture motion information. Unlike traditional approaches, we propose combining RV and BEV residual maps to exploit a greater potential of motion information jointly. Thus, we introduce CV-MOS, a cross-view model for moving object segmentation. Novelty, we decouple spatial-temporal information by capturing the motion from BEV and RV residual maps and generating semantic features from range images, which are used as moving object guidance for the motion branch. Our direct and unique solution maximizes the use of range images and RV and BEV residual maps, significantly enhancing the performance of LiDAR-based MOS task. Our method achieved leading IoU(\%) scores of 77.5\% and 79.2\% on the validation and test sets of the SemanticKitti dataset. In particular, CV-MOS demonstrates SOTA performance to date on various datasets. The CV-MOS implementation is available at https://github.com/SCNU-RISLAB/CV-MOS
Authors: Seongmin Park, Minjae Lee, Junwon Choi, Jungwook Choi
Abstract: Pillar-based 3D object detection has gained traction in self-driving technology due to its speed and accuracy facilitated by the artificial densification of pillars for GPU-friendly processing. However, dense pillar processing fundamentally wastes computation since it ignores the inherent sparsity of pillars derived from scattered point cloud data. Motivated by recent embedded accelerators with native sparsity support, sparse pillar convolution methods like submanifold convolution (SubM-Conv) aimed to reduce these redundant computations by applying convolution only on active pillars but suffered considerable accuracy loss. Our research identifies that this accuracy loss is due to the restricted fine-grained spatial information flow (fSIF) of SubM-Conv in sparse pillar networks. To overcome this restriction, we propose a selectively dilated (SD-Conv) convolution that evaluates the importance of encoded pillars and selectively dilates the convolution output, enhancing the receptive field for critical pillars and improving object detection accuracy. To facilitate actual acceleration with this novel convolution approach, we designed SPADE+ as a cost-efficient augmentation to existing embedded sparse convolution accelerators. This design supports the SD-Conv without significant demands in area and SRAM size, realizing superior trade-off between the speedup and model accuracy. This strategic enhancement allows our method to achieve extreme pillar sparsity, leading to up to 18.1x computational savings and 16.2x speedup on the embedded accelerators, without compromising object detection accuracy.
Authors: Xiongwei Zhao, Congcong Wen, Yang Wang, Haojie Bai, Wenhao Dou
Abstract: LiDAR sensors are crucial for providing high-resolution 3D point cloud data in autonomous driving systems, enabling precise environmental perception. However, real-world adverse weather conditions, such as rain, fog, and snow, introduce significant noise and interference, degrading the reliability of LiDAR data and the performance of downstream tasks like semantic segmentation. Existing datasets often suffer from limited weather diversity and small dataset sizes, which restrict their effectiveness in training models. Additionally, current deep learning denoising methods, while effective in certain scenarios, often lack interpretability, complicating the ability to understand and validate their decision-making processes. To overcome these limitations, we introduce two large-scale datasets, Weather-KITTI and Weather-NuScenes, which cover three common adverse weather conditions: rain, fog, and snow. These datasets retain the original LiDAR acquisition information and provide point-level semantic labels for rain, fog, and snow. Furthermore, we propose a novel point cloud denoising model, TripleMixer, comprising three mixer layers: the Geometry Mixer Layer, the Frequency Mixer Layer, and the Channel Mixer Layer. These layers are designed to capture geometric spatial information, extract multi-scale frequency information, and enhance the multi-channel feature information of point clouds, respectively. Experiments conducted on the WADS dataset in real-world scenarios, as well as on our proposed Weather-KITTI and Weather-NuScenes datasets, demonstrate that our model achieves state-of-the-art denoising performance. Additionally, our experiments show that integrating the denoising model into existing segmentation frameworks enhances the performance of downstream tasks.The datasets and code will be made publicly available at https://github.com/Grandzxw/TripleMixer.
Authors: Tal Alter, Raz Lapid, Moshe Sipper
Abstract: Kolmogorov-Arnold Networks (KANs) have recently emerged as a novel approach to function approximation, demonstrating remarkable potential in various domains. Despite their theoretical promise, the robustness of KANs under adversarial conditions has yet to be thoroughly examined. In this paper, we explore the adversarial robustness of KANs, with a particular focus on image classification tasks. We assess the performance of KANs against standard white-box adversarial attacks, comparing their resilience to that of established neural network architectures. Further, we investigate the transferability of adversarial examples between KANs and Multilayer Perceptron (MLPs), deriving critical insights into the unique vulnerabilities of KANs. Our experiments use the MNIST, FashionMNIST, and KMNIST datasets, providing a comprehensive evaluation of KANs in adversarial scenarios. This work offers the first in-depth analysis of security in KANs, laying the groundwork for future research in this emerging field.
Authors: Ardhendu Sekhar, Ravi Kant Gupta, Amit Sethi
Abstract: This paper presents a study on few-shot classification in the context of histopathology images. While few-shot learning has been studied for natural image classification, its application to histopathology is relatively unexplored. Given the scarcity of labeled data in medical imaging and the inherent challenges posed by diverse tissue types and data preparation techniques, this research evaluates the performance of state-of-the-art few-shot learning methods for various scenarios on histology data. We have considered four histopathology datasets for few-shot histopathology image classification and have evaluated 5-way 1-shot, 5-way 5-shot and 5-way 10-shot scenarios with a set of state-of-the-art classification techniques. The best methods have surpassed an accuracy of 70%, 80% and 85% in the cases of 5-way 1-shot, 5-way 5-shot and 5-way 10-shot cases, respectively. We found that for histology images popular meta-learning approaches is at par with standard fine-tuning and regularization methods. Our experiments underscore the challenges of working with images from different domains and underscore the significance of unbiased and focused evaluations in advancing computer vision techniques for specialized domains, such as histology images.
Authors: Yuhong Jiao, Jiaqing Miao, Jinnan Gong, Hui He, Ping Liang, Cheng Luo, Ying Tan
Abstract: Schizophrenia is a serious psychiatric disorder. Its pathogenesis is not completely clear, making it difficult to treat patients precisely. Because of the complicated non-Euclidean network structure of the human brain, learning critical information from brain networks remains difficult. To effectively capture the topological information of brain neural networks, a novel multimodal graph attention network based on sparse interaction mechanism (Multi-SIGATnet) was proposed for SZ classification was proposed for SZ classification. Firstly, structural and functional information were fused into multimodal data to obtain more comprehensive and abundant features for patients with SZ. Subsequently, a sparse interaction mechanism was proposed to effectively extract salient features and enhance the feature representation capability. By enhancing the strong connections and weakening the weak connections between feature information based on an asymmetric convolutional network, high-order interactive features were captured. Moreover, sparse learning strategies were designed to filter out redundant connections to improve model performance. Finally, local and global features were updated in accordance with the topological features and connection weight constraints of the higher-order brain network, the features being projected to the classification target space for disorder classification. The effectiveness of the model is verified on the Center for Biomedical Research Excellence (COBRE) and University of California Los Angeles (UCLA) datasets, achieving 81.9\% and 75.8\% average accuracy, respectively, 4.6\% and 5.5\% higher than the graph attention network (GAT) method. Experiments showed that the Multi-SIGATnet method exhibited good performance in identifying SZ.
Authors: Zifan Chen, Xinyu Nan, Jiazheng Li, Jie Zhao, Haifeng Li, Zilin Lin, Haoshen Li, Heyun Chen, Yiting Liu, Bin Dong, Li Zhang, Lei Tang
Abstract: Volumetric segmentation is crucial for medical imaging but is often constrained by labor-intensive manual annotations and the need for scenario-specific model training. Furthermore, existing general segmentation models are inefficient due to their design and inferential approaches. Addressing this clinical demand, we introduce PropSAM, a propagation-based segmentation model that optimizes the use of 3D medical structure information. PropSAM integrates a CNN-based UNet for intra-slice processing with a Transformer-based module for inter-slice propagation, focusing on structural and semantic continuities to enhance segmentation across various modalities. Distinctively, PropSAM operates on a one-view prompt, such as a 2D bounding box or sketch mask, unlike conventional models that require two-view prompts. It has demonstrated superior performance, significantly improving the Dice Similarity Coefficient (DSC) across 44 medical datasets and various imaging modalities, outperforming models like MedSAM and SegVol with an average DSC improvement of 18.1%. PropSAM also maintains stable predictions despite prompt deviations and varying propagation configurations, confirmed by one-way ANOVA tests with P>0.5985 and P>0.6131, respectively. Moreover, PropSAM's efficient architecture enables faster inference speeds (Wilcoxon rank-sum test, P<0.001) and reduces user interaction time by 37.8% compared to two-view prompt models. Its ability to handle irregular and complex objects with robust performance further demonstrates its potential in clinical settings, facilitating more automated and reliable medical imaging analyses with minimal retraining.
Authors: Yuwen Pan, Rui Sun, Naisong Luo, Tianzhu Zhang, Yongdong Zhang
Abstract: Semantic segmentation of night-time images holds significant importance in computer vision, particularly for applications like night environment perception in autonomous driving systems. However, existing methods tend to parse night-time images from a day-time perspective, leaving the inherent challenges in low-light conditions (such as compromised texture and deceiving matching errors) unexplored. To address these issues, we propose a novel end-to-end optimized approach, named NightFormer, tailored for night-time semantic segmentation, avoiding the conventional practice of forcibly fitting night-time images into day-time distributions. Specifically, we design a pixel-level texture enhancement module to acquire texture-aware features hierarchically with phase enhancement and amplified attention, and an object-level reliable matching module to realize accurate association matching via reliable attention in low-light environments. Extensive experimental results on various challenging benchmarks including NightCity, BDD and Cityscapes demonstrate that our proposed method performs favorably against state-of-the-art night-time semantic segmentation methods.
Authors: Xuyi Yu
Abstract: Due to the high complexity and technical requirements of industrial production processes, surface defects will inevitably appear, which seriously affects the quality of products. Although existing lightweight detection networks are highly efficient, they are susceptible to false or missed detection of non-salient defects due to the lack of semantic information. In contrast, the diffusion model can generate higher-order semantic representations in the denoising process. Therefore, the aim of this paper is to incorporate the higher-order modelling capability of the diffusion model into the detection model, so as to better assist in the classification and localization of difficult targets. First, the denoising diffusion probabilistic model (DDPM) is pre-trained to extract the features of denoising process to construct as a feature repository. In particular, to avoid the potential bottleneck of memory caused by the dataloader loading high-dimensional features, a residual convolutional variational auto-encoder (ResVAE) is designed to further compress the feature repository. The image is fed into both image backbone and feature repository for feature extraction and querying respectively. The queried latent features are reconstructed and filtered to obtain high-dimensional DDPM features. A dynamic cross-fusion method is proposed to fully refine the contextual features of DDPM to optimize the detection model. Finally, we employ knowledge distillation to migrate the higher-order modelling capabilities back into the lightweight baseline model without additional efficiency cost. Experiment results demonstrate that our method achieves competitive results on several industrial datasets.
Authors: Keyi Zhou, Li Li, Wengang Zhou, Yonghui Wang, Hao Feng, Houqiang Li
Abstract: In video lane detection, there are rich temporal contexts among successive frames, which is under-explored in existing lane detectors. In this work, we propose LaneTCA to bridge the individual video frames and explore how to effectively aggregate the temporal context. Technically, we develop an accumulative attention module and an adjacent attention module to abstract the long-term and short-term temporal context, respectively. The accumulative attention module continuously accumulates visual information during the journey of a vehicle, while the adjacent attention module propagates this lane information from the previous frame to the current frame. The two modules are meticulously designed based on the transformer architecture. Finally, these long-short context features are fused with the current frame features to predict the lane lines in the current frame. Extensive quantitative and qualitative experiments are conducted on two prevalent benchmark datasets. The results demonstrate the effectiveness of our method, achieving several new state-of-the-art records. The codes and models are available at https://github.com/Alex-1337/LaneTCA
Authors: Jiamin Tang, Chao Zhang, Xudong Zhu, Mengchi Liu
Abstract: Significant advancements in Large Multimodal Models (LMMs) have enabled them to tackle complex problems involving visual-mathematical reasoning. However, their ability to identify geometric elements remains understudied. To bridge this gap, we introduce Tangram, a novel benchmark designed to evaluate the performance of LMMs on geometric element recognition. Tangram includes 1,080 diverse geometric diagrams sourced from primary and secondary school exams, competitions, and textbooks, covering from simple basic geometric shapes to complex combinations. Each diagram is associated with four questions, resulting in a total of 4,320 visual-question-answer pairs. Unlike existing benchmarks that seek higher-level cognition and reasoning, Tangram focuses on the understanding of geometric elements, requiring models to perform a "simple but interesting" counting task. Systematic evaluation of 10 prominent LMMs, such as GPT-4o and Claude 3.5 Sonnet, shows that even in the seemingly simple task, these models still face significant challenges. Notably, the overall accuracy of the top performer across all tested models is only 56.8%, marking a significant gap when compared to human performance. These findings highlight the limitations of current multimodal artificial intelligence systems in handling basic perception tasks, and will inspire the development of the next generation of expert-level multimodal foundational models. The Tangram and evaluation code will be available soon.
Authors: Minghao Liu, Le Zhang, Yingjie Tian, Xiaochao Qu, Luoqi Liu, Ting Liu
Abstract: Recent advances in text-to-image diffusion models have demonstrated impressive capabilities in image quality. However, complex scene generation remains relatively unexplored, and even the definition of `complex scene' itself remains unclear. In this paper, we address this gap by providing a precise definition of complex scenes and introducing a set of Complex Decomposition Criteria (CDC) based on this definition. Inspired by the artists painting process, we propose a training-free diffusion framework called Complex Diffusion (CxD), which divides the process into three stages: composition, painting, and retouching. Our method leverages the powerful chain-of-thought capabilities of large language models (LLMs) to decompose complex prompts based on CDC and to manage composition and layout. We then develop an attention modulation method that guides simple prompts to specific regions to complete the complex scene painting. Finally, we inject the detailed output of the LLM into a retouching model to enhance the image details, thus implementing the retouching stage. Extensive experiments demonstrate that our method outperforms previous SOTA approaches, significantly improving the generation of high-quality, semantically consistent, and visually diverse images for complex scenes, even with intricate prompts.
Authors: Amir Nazemi, Mohammad Hadi Sepanj, Nicholas Pellegrino, Chris Czarnecki, Paul Fieguth
Abstract: Current strategies for solving image-based inverse problems apply latent diffusion models to perform posterior sampling.However, almost all approaches make no explicit attempt to explore the solution space, instead drawing only a single sample from a Gaussian distribution from which to generate their solution. In this paper, we introduce a particle-filtering-based framework for a nonlinear exploration of the solution space in the initial stages of reverse SDE methods. Our proposed particle-filtering-based latent diffusion (PFLD) method and proposed problem formulation and framework can be applied to any diffusion-based solution for linear or nonlinear inverse problems. Our experimental results show that PFLD outperforms the SoTA solver PSLD on the FFHQ-1K and ImageNet-1K datasets on inverse problem tasks of super resolution, Gaussian debluring and inpainting.
Authors: Xiaoyu Guo, Pengzhi Zhong, Hao Zhang, Ling Huang, Defeng Huang, Shuiwang Li
Abstract: Visual tracking has seen remarkable advancements, largely driven by the availability of large-scale training datasets that have enabled the development of highly accurate and robust algorithms. While significant progress has been made in tracking general objects, research on more challenging scenarios, such as tracking camouflaged objects, remains limited. Camouflaged objects, which blend seamlessly with their surroundings or other objects, present unique challenges for detection and tracking in complex environments. This challenge is particularly critical in applications such as military, security, agriculture, and marine monitoring, where precise tracking of camouflaged objects is essential. To address this gap, we introduce the Camouflaged Object Tracking Dataset (COTD), a specialized benchmark designed specifically for evaluating camouflaged object tracking methods. The COTD dataset comprises 200 sequences and approximately 80,000 frames, each annotated with detailed bounding boxes. Our evaluation of 20 existing tracking algorithms reveals significant deficiencies in their performance with camouflaged objects. To address these issues, we propose a novel tracking framework, HiPTrack-MLS, which demonstrates promising results in improving tracking performance for camouflaged objects. COTD and code are avialable at https://github.com/openat25/HIPTrack-MLS.
Authors: Zhijian Huang, Tao Tang, Shaoxiang Chen, Sihao Lin, Zequn Jie, Lin Ma, Guangrun Wang, Xiaodan Liang
Abstract: Data-driven approaches for autonomous driving (AD) have been widely adopted in the past decade but are confronted with dataset bias and uninterpretability. Inspired by the knowledge-driven nature of human driving, recent approaches explore the potential of large language models (LLMs) to improve understanding and decision-making in traffic scenarios. They find that the pretrain-finetune paradigm of LLMs on downstream data with the Chain-of-Thought (CoT) reasoning process can enhance explainability and scene understanding. However, such a popular strategy proves to suffer from the notorious problems of misalignment between the crafted CoTs against the consequent decision-making, which remains untouched by previous LLM-based AD methods. To address this problem, we motivate an end-to-end decision-making model based on multimodality-augmented LLM, which simultaneously executes CoT reasoning and carries out planning results. Furthermore, we propose a reasoning-decision alignment constraint between the paired CoTs and planning results, imposing the correspondence between reasoning and decision-making. Moreover, we redesign the CoTs to enable the model to comprehend complex scenarios and enhance decision-making performance. We dub our proposed large language planners with reasoning-decision alignment as RDA-Driver. Experimental evaluations on the nuScenes and DriveLM-nuScenes benchmarks demonstrate the effectiveness of our RDA-Driver in enhancing the performance of end-to-end AD systems. Specifically, our RDA-Driver achieves state-of-the-art planning performance on the nuScenes dataset with 0.80 L2 error and 0.32 collision rate, and also achieves leading results on challenging DriveLM-nuScenes benchmarks with 0.82 L2 error and 0.38 collision rate.
Authors: Sensen Gao, Xiaojun Jia, Yihao Huang, Ranjie Duan, Jindong Gu, Yang Liu, Qing Guo
Abstract: Recently, Text-to-Image(T2I) models have achieved remarkable success in image generation and editing, yet these models still have many potential issues, particularly in generating inappropriate or Not-Safe-For-Work(NSFW) content. Strengthening attacks and uncovering such vulnerabilities can advance the development of reliable and practical T2I models. Most of the previous works treat T2I models as white-box systems, using gradient optimization to generate adversarial prompts. However, accessing the model's gradient is often impossible in real-world scenarios. Moreover, existing defense methods, those using gradient masking, are designed to prevent attackers from obtaining accurate gradient information. While some black-box jailbreak attacks have been explored, these typically rely on simply replacing sensitive words, leading to suboptimal attack performance. To address this issue, we introduce a two-stage query-based black-box attack method utilizing random search. In the first stage, we establish a preliminary prompt by maximizing the semantic similarity between the adversarial and target harmful prompts. In the second stage, we use this initial prompt to refine our approach, creating a detailed adversarial prompt aimed at jailbreaking and maximizing the similarity in image features between the images generated from this prompt and those produced by the target harmful prompt. Extensive experiments validate the effectiveness of our method in attacking the latest prompt checkers, post-hoc image checkers, securely trained T2I models, and online commercial models.
Authors: Haiwen Zhang, Zixi Yang, Yuanzhi Liu, Xinran Wang, Zheqi He, Kongming Liang, Zhanyu Ma
Abstract: Currently, large vision-language models have gained promising progress on many downstream tasks. However, they still suffer many challenges in fine-grained visual understanding tasks, such as object attribute comprehension. Besides, there have been growing efforts on the evaluations of large vision-language models, but lack of in-depth study of attribute comprehension and the visual language fine-tuning process. In this paper, we propose to evaluate the attribute comprehension ability of large vision-language models from two perspectives: attribute recognition and attribute hierarchy understanding. We evaluate three vision-language interactions, including visual question answering, image-text matching, and image-text cosine similarity. Furthermore, we explore the factors affecting attribute comprehension during fine-tuning. Through a series of quantitative and qualitative experiments, we introduce three main findings: (1) Large vision-language models possess good attribute recognition ability, but their hierarchical understanding ability is relatively limited. (2) Compared to ITC, ITM exhibits superior capability in capturing finer details, making it more suitable for attribute understanding tasks. (3) The attribute information in the captions used for fine-tuning plays a crucial role in attribute understanding. We hope this work can help guide future progress in fine-grained visual understanding of large vision-language models.
Authors: Li Li, Tanqiu Qiao, Hubert P. H. Shum, Toby P. Breckon
Abstract: 3D point clouds are essential for perceiving outdoor scenes, especially within the realm of autonomous driving. Recent advances in 3D LiDAR Object Detection focus primarily on the spatial positioning and distribution of points to ensure accurate detection. However, despite their robust performance in variable conditions, these methods are hindered by their sole reliance on coordinates and point intensity, resulting in inadequate isometric invariance and suboptimal detection outcomes. To tackle this challenge, our work introduces Transformation-Invariant Local (TraIL) features and the associated TraIL-Det architecture. Our TraIL features exhibit rigid transformation invariance and effectively adapt to variations in point density, with a design focus on capturing the localized geometry of neighboring structures. They utilize the inherent isotropic radiation of LiDAR to enhance local representation, improve computational efficiency, and boost detection performance. To effectively process the geometric relations among points within each proposal, we propose a Multi-head self-Attention Encoder (MAE) with asymmetric geometric features to encode high-dimensional TraIL features into manageable representations. Our method outperforms contemporary self-supervised 3D object detection approaches in terms of mAP on KITTI (67.8, 20% label, moderate) and Waymo (68.9, 20% label, moderate) datasets under various label ratios (20%, 50%, and 100%).
Authors: Yeji Park, Deokyeong Lee, Junsuk Choe, Buru Chang
Abstract: Hallucinations in Multimodal Large Language Models (MLLMs) where generated responses fail to accurately reflect the given image pose a significant challenge to their reliability. To address this, we introduce ConVis, a novel training-free contrastive decoding method. ConVis leverages a text-to-image (T2I) generation model to semantically reconstruct the given image from hallucinated captions. By comparing the contrasting probability distributions produced by the original and reconstructed images, ConVis enables MLLMs to capture visual contrastive signals that penalize hallucination generation. Notably, this method operates purely within the decoding process, eliminating the need for additional data or model updates. Our extensive experiments on five popular benchmarks demonstrate that ConVis effectively reduces hallucinations across various MLLMs, highlighting its potential to enhance model reliability.
Authors: Ali Asgarov, Samir Rustamov
Abstract: This research explores the development of multimodal vision-language models for image retrieval in low-resource languages, specifically Azerbaijani. Existing vision-language models primarily support high-resource languages, and fine-tuning them remains computationally demanding. To address challenges in vision-language retrieval for low-resource languages, we integrated the CLIP model architecture and employed several techniques to balance computational efficiency with performance. These techniques include synthetic data generation through machine translation, image augmentation, and further training the attention mechanisms of transformer-based models with domain-specific data. We integrated Multilingual BERT as a text encoder with image encoders like ResNet50, EfficientNet0, Vision Transformer (ViT), and Tiny Swin Transformer. Our study found that models like EfficientNet0 and Tiny Swin Transformer perform best on the datasets they were trained on, such as COCO, Flickr30k, and Flickr8k. Augmentation techniques boosted EfficientNet0 MAP on Flickr30k from 0.84 to 0.87 and ResNet50 MAP on MSCOCO from 0.70 to 0.80, contributing to a new state of the art in vision-language retrieval. We share our configurations and results to support further research. Code and pre-trained models are available at https://github.com/aliasgerovs/azclip.
Authors: Brandon Smart, Chuanxia Zheng, Iro Laina, Victor Adrian Prisacariu
Abstract: In this paper, we introduce Splatt3R, a pose-free, feed-forward method for in-the-wild 3D reconstruction and novel view synthesis from stereo pairs. Given uncalibrated natural images, Splatt3R can predict 3D Gaussian Splats without requiring any camera parameters or depth information. For generalizability, we start from a 'foundation' 3D geometry reconstruction method, MASt3R, and extend it to be a full 3D structure and appearance reconstructor. Specifically, unlike the original MASt3R which reconstructs only 3D point clouds, we predict the additional Gaussian attributes required to construct a Gaussian primitive for each point. Hence, unlike other novel view synthesis methods, Splatt3R is first trained by optimizing the 3D point cloud's geometry loss, and then a novel view synthesis objective. By doing this, we avoid the local minima present in training 3D Gaussian Splats from stereo views. We also propose a novel loss masking strategy that we empirically find is critical for strong performance on extrapolated viewpoints. We train Splatt3R on the ScanNet++ dataset and demonstrate excellent generalisation to uncalibrated, in-the-wild images. Splatt3R can reconstruct scenes at 4FPS at 512 x 512 resolution, and the resultant splats can be rendered in real-time.
Authors: Andrew Hou, Zhixin Shu, Xuaner Zhang, He Zhang, Yannick Hold-Geoffroy, Jae Shin Yoon, Xiaoming Liu
Abstract: Existing portrait relighting methods struggle with precise control over facial shadows, particularly when faced with challenges such as handling hard shadows from directional light sources or adjusting shadows while remaining in harmony with existing lighting conditions. In many situations, completely altering input lighting is undesirable for portrait retouching applications: one may want to preserve some authenticity in the captured environment. Existing shadow editing methods typically restrict their application to just the facial region and often offer limited lighting control options, such as shadow softening or rotation. In this paper, we introduce COMPOSE: a novel shadow editing pipeline for human portraits, offering precise control over shadow attributes such as shape, intensity, and position, all while preserving the original environmental illumination of the portrait. This level of disentanglement and controllability is obtained thanks to a novel decomposition of the environment map representation into ambient light and an editable gaussian dominant light source. COMPOSE is a four-stage pipeline that consists of light estimation and editing, light diffusion, shadow synthesis, and finally shadow editing. We define facial shadows as the result of a dominant light source, encoded using our novel gaussian environment map representation. Utilizing an OLAT dataset, we have trained models to: (1) predict this light source representation from images, and (2) generate realistic shadows using this representation. We also demonstrate comprehensive and intuitive shadow editing with our pipeline. Through extensive quantitative and qualitative evaluations, we have demonstrated the robust capability of our system in shadow editing.
Authors: Burak Sevsay, Erdem Akag\"und\"uz
Abstract: Quantization is one of the most popular techniques for reducing computation time and shrinking model size. However, ensuring the accuracy of quantized models typically involves calibration using training data, which may be inaccessible due to privacy concerns. In such cases, zero-shot quantization, a technique that relies on pretrained models and statistical information without the need for specific training data, becomes valuable. Exploring zero-shot quantization in the infrared domain is important due to the prevalence of infrared imaging in sensitive fields like medical and security applications. In this work, we demonstrate how to apply zero-shot quantization to an object detection model retrained with thermal imagery. We use batch normalization statistics of the model to distill data for calibration. RGB image-trained models and thermal image-trained models are compared in the context of zero-shot quantization. Our investigation focuses on the contributions of mean and standard deviation statistics to zero-shot quantization performance. Additionally, we compare zero-shot quantization with post-training quantization on a thermal dataset. We demonstrated that zero-shot quantization successfully generates data that represents the training dataset for the quantization of object detection models. Our results indicate that our zero-shot quantization framework is effective in the absence of training data and is well-suited for the infrared domain.
Authors: Lukas Picek, Christophe Botella, Maximilien Servajean, C\'esar Leblanc, R\'emi Palard, Th\'eo Larcher, Benjamin Deneu, Diego Marcos, Pierre Bonnet, Alexis Joly
Abstract: The difficulty of monitoring biodiversity at fine scales and over large areas limits ecological knowledge and conservation efforts. To fill this gap, Species Distribution Models (SDMs) predict species across space from spatially explicit features. Yet, they face the challenge of integrating the rich but heterogeneous data made available over the past decade, notably millions of opportunistic species observations and standardized surveys, as well as multi-modal remote sensing data. In light of that, we have designed and developed a new European-scale dataset for SDMs at high spatial resolution (10-50 m), including more than 10k species (i.e., most of the European flora). The dataset comprises 5M heterogeneous Presence-Only records and 90k exhaustive Presence-Absence survey records, all accompanied by diverse environmental rasters (e.g., elevation, human footprint, and soil) that are traditionally used in SDMs. In addition, it provides Sentinel-2 RGB and NIR satellite images with 10 m resolution, a 20-year time-series of climatic variables, and satellite time-series from the Landsat program. In addition to the data, we provide an openly accessible SDM benchmark (hosted on Kaggle), which has already attracted an active community and a set of strong baselines for single predictor/modality and multimodal approaches. All resources, e.g., the dataset, pre-trained models, and baseline methods (in the form of notebooks), are available on Kaggle, allowing one to start with our dataset literally with two mouse clicks.
Authors: Muhammad Rameez ur Rahman, Piero Simonetto, Anna Polato, Francesco Pasti, Luca Tonin, Sebastiano Vascon
Abstract: Open vocabulary 3D object detection (OV3D) allows precise and extensible object recognition crucial for adapting to diverse environments encountered in assistive robotics. This paper presents OpenNav, a zero-shot 3D object detection pipeline based on RGB-D images for smart wheelchairs. Our pipeline integrates an open-vocabulary 2D object detector with a mask generator for semantic segmentation, followed by depth isolation and point cloud construction to create 3D bounding boxes. The smart wheelchair exploits these 3D bounding boxes to identify potential targets and navigate safely. We demonstrate OpenNav's performance through experiments on the Replica dataset and we report preliminary results with a real wheelchair. OpenNav improves state-of-the-art significantly on the Replica dataset at mAP25 (+9pts) and mAP50 (+5pts) with marginal improvement at mAP. The code is publicly available at this link: https://github.com/EasyWalk-PRIN/OpenNav.
Authors: Xianghui Xie, Jan Eric Lenssen, Gerard Pons-Moll
Abstract: Tracking human object interaction from videos is important to understand human behavior from the rapidly growing stream of video data. Previous video-based methods require predefined object templates while single-image-based methods are template-free but lack temporal consistency. In this paper, we present a method to track human object interaction without any object shape templates. We decompose the 4D tracking problem into per-frame pose tracking and canonical shape optimization. We first apply a single-view reconstruction method to obtain temporally-inconsistent per-frame interaction reconstructions. Then, for the human, we propose an efficient autoencoder to predict SMPL vertices directly from the per-frame reconstructions, introducing temporally consistent correspondence. For the object, we introduce a pose estimator that leverages temporal information to predict smooth object rotations under occlusions. To train our model, we propose a method to generate synthetic interaction videos and synthesize in total 10 hour videos of 8.5k sequences with full 3D ground truth. Experiments on BEHAVE and InterCap show that our method significantly outperforms previous template-based video tracking and single-frame reconstruction methods. Our proposed synthetic video dataset also allows training video-based methods that generalize to real-world videos. Our code and dataset will be publicly released.
Authors: Jia Cheng Hu, Roberto Cavicchioli, Alessandro Capotondi
Abstract: Image Captioning is an important Language and Vision task that finds application in a variety of contexts, ranging from healthcare to autonomous vehicles. As many real-world applications rely on devices with limited resources, much effort in the field was put into the development of lighter and faster models. However, much of the current optimizations focus on the Transformer architecture in contrast to the existence of more efficient methods. In this work, we introduce SwiFTeR, an architecture almost entirely based on Fourier Transform and Retention, to tackle the main efficiency bottlenecks of current light image captioning models, being the visual backbone's onerosity, and the decoder's quadratic cost. SwiFTeR is made of only 20M parameters, and requires 3.1 GFLOPs for a single forward pass. Additionally, it showcases superior scalability to the caption length and its small memory requirements enable more images to be processed in parallel, compared to the traditional transformer-based architectures. For instance, it can generate 400 captions in one second. Although, for the time being, the caption quality is lower (110.2 CIDEr-D), most of the decrease is not attributed to the architecture but rather an incomplete training practice which currently leaves much room for improvements. Overall, SwiFTeR points toward a promising direction to new efficient architectural design. The implementation code will be released in the future.
Authors: Weiwei Cai, Weicai Ye, Peng Ye, Tong He, Tao Chen
Abstract: Dynamic scene reconstruction has garnered significant attention in recent years due to its capabilities in high-quality and real-time rendering. Among various methodologies, constructing a 4D spatial-temporal representation, such as 4D-GS, has gained popularity for its high-quality rendered images. However, these methods often produce suboptimal surfaces, as the discrete 3D Gaussian point clouds fail to align with the object's surface precisely. To address this problem, we propose DynaSurfGS to achieve both photorealistic rendering and high-fidelity surface reconstruction of dynamic scenarios. Specifically, the DynaSurfGS framework first incorporates Gaussian features from 4D neural voxels with the planar-based Gaussian Splatting to facilitate precise surface reconstruction. It leverages normal regularization to enforce the smoothness of the surface of dynamic objects. It also incorporates the as-rigid-as-possible (ARAP) constraint to maintain the approximate rigidity of local neighborhoods of 3D Gaussians between timesteps and ensure that adjacent 3D Gaussians remain closely aligned throughout. Extensive experiments demonstrate that DynaSurfGS surpasses state-of-the-art methods in both high-fidelity surface reconstruction and photorealistic rendering.
Authors: Shuai Fu, Xiequn Wang, Qiushi Huang, Yu Zhang
Abstract: With the prevalence of large-scale pretrained vision-language models (VLMs), such as CLIP, soft-prompt tuning has become a popular method for adapting these models to various downstream tasks. However, few works delve into the inherent properties of learnable soft-prompt vectors, specifically the impact of their norms to the performance of VLMs. This motivates us to pose an unexplored research question: ``Do we need to normalize the soft prompts in VLMs?'' To fill this research gap, we first uncover a phenomenon, called the \textbf{Low-Norm Effect} by performing extensive corruption experiments, suggesting that reducing the norms of certain learned prompts occasionally enhances the performance of VLMs, while increasing them often degrades it. To harness this effect, we propose a novel method named \textbf{N}ormalizing th\textbf{e} soft-pro\textbf{m}pt v\textbf{e}ctors of vi\textbf{si}on-language model\textbf{s} (\textbf{Nemesis}) to normalize soft-prompt vectors in VLMs. To the best of our knowledge, our work is the first to systematically investigate the role of norms of soft-prompt vector in VLMs, offering valuable insights for future research in soft-prompt tuning. The code is available at \texttt{\href{https://github.com/ShyFoo/Nemesis}{https://github.com/ShyFoo/Nemesis}}.
URLs: https://github.com/ShyFoo/Nemesis, https://github.com/ShyFoo/Nemesis
Authors: Daixun Li, Weiying Xie, Mingxiang Cao, Yunke Wang, Jiaqing Zhang, Yunsong Li, Leyuan Fang, Chang Xu
Abstract: Multimodal image fusion and segmentation enhance scene understanding in autonomous driving by integrating data from various sensors. However, current models struggle to efficiently segment densely packed elements in such scenes, due to the absence of comprehensive fusion features that can guide mid-process fine-tuning and focus attention on relevant areas. The Segment Anything Model (SAM) has emerged as a transformative segmentation method. It provides more effective prompts through its flexible prompt encoder, compared to transformers lacking fine-tuned control. Nevertheless, SAM has not been extensively studied in the domain of multimodal fusion for natural images. In this paper, we introduce SAM into multimodal image segmentation for the first time, proposing a novel framework that combines Latent Space Token Generation (LSTG) and Fusion Mask Prompting (FMP) modules to enhance SAM's multimodal fusion and segmentation capabilities. Specifically, we first obtain latent space features of the two modalities through vector quantization and embed them into a cross-attention-based inter-domain fusion module to establish long-range dependencies between modalities. Then, we use these comprehensive fusion features as prompts to guide precise pixel-level segmentation. Extensive experiments on several public datasets demonstrate that the proposed method significantly outperforms SAM and SAM2 in multimodal autonomous driving scenarios, achieving at least 3.9$\%$ higher segmentation mIoU than the state-of-the-art approaches.
Authors: Lu Wen, Wenxia Yin, Zhenghao Feng, Xi Wu, Deng Xiong, Yan Wang
Abstract: Radiation therapy is the mainstay treatment for cervical cancer, and its ultimate goal is to ensure the planning target volume (PTV) reaches the prescribed dose while reducing dose deposition of organs-at-risk (OARs) as much as possible. To achieve these clinical requirements, the medical physicist needs to manually tweak the radiotherapy plan repeatedly in a trial-anderror manner until finding the optimal one in the clinic. However, such trial-and-error processes are quite time-consuming, and the quality of plans highly depends on the experience of the medical physicist. In this paper, we propose an end-to-end Attentionbased Residual Adversarial Network with deep supervision, namely ARANet, to automatically predict the 3D dose distribution of cervical cancer. Specifically, given the computer tomography (CT) images and their corresponding segmentation masks of PTV and OARs, ARANet employs a prediction network to generate the dose maps. We also utilize a multi-scale residual attention module and deep supervision mechanism to enforce the prediction network to extract more valuable dose features while suppressing irrelevant information. Our proposed method is validated on an in-house dataset including 54 cervical cancer patients, and experimental results have demonstrated its obvious superiority compared to other state-of-the-art methods.
Authors: Yushun Tang, Shuoshuo Chen, Zhihe Lu, Xinchao Wang, Zhihai He
Abstract: Transformer-based methods have achieved remarkable success in various machine learning tasks. How to design efficient test-time adaptation methods for transformer models becomes an important research task. In this work, motivated by the dual-subband wavelet lifting scheme developed in multi-scale signal processing which is able to efficiently separate the input signals into principal components and noise components, we introduce a dual-path token lifting for domain shift correction in test time adaptation. Specifically, we introduce an extra token, referred to as \textit{domain shift token}, at each layer of the transformer network. We then perform dual-path lifting with interleaved token prediction and update between the path of domain shift tokens and the path of class tokens at all network layers. The prediction and update networks are learned in an adversarial manner. Specifically, the task of the prediction network is to learn the residual noise of domain shift which should be largely invariant across all classes and all samples in the target domain. In other words, the predicted domain shift noise should be indistinguishable between all sample classes. On the other hand, the task of the update network is to update the class tokens by removing the domain shift from the input image samples so that input samples become more discriminative between different classes in the feature space. To effectively learn the prediction and update networks with two adversarial tasks, both theoretically and practically, we demonstrate that it is necessary to use smooth optimization for the update network but non-smooth optimization for the prediction network. Experimental results on the benchmark datasets demonstrate that our proposed method significantly improves the online fully test-time domain adaptation performance. Code is available at \url{https://github.com/yushuntang/DPAL}.
Authors: Li Guo, Anas M. Tahir, Dong Zhang, Z. Jane Wang, Rabab K. Ward
Abstract: The increasing demand for medical imaging has surpassed the capacity of available radiologists, leading to diagnostic delays and potential misdiagnoses. Artificial intelligence (AI) techniques, particularly in automatic medical report generation (AMRG), offer a promising solution to this dilemma. This review comprehensively examines AMRG methods from 2021 to 2024. It (i) presents solutions to primary challenges in this field, (ii) explores AMRG applications across various imaging modalities, (iii) introduces publicly available datasets, (iv) outlines evaluation metrics, (v) identifies techniques that significantly enhance model performance, and (vi) discusses unresolved issues and potential future research directions. This paper aims to provide a comprehensive understanding of the existing literature and inspire valuable future research.
Authors: Yixuan He, Lin Geng Foo, Ajmal Saeed Mian, Hossein Rahmani, Jun Jiu
Abstract: Language based editing of 3D human avatars to precisely match user requirements is challenging due to the inherent ambiguity and limited expressiveness of natural language. To overcome this, we propose the Avatar Concept Slider (ACS), a 3D avatar editing method that allows precise manipulation of semantic concepts in human avatars towards a specified intermediate point between two extremes of concepts, akin to moving a knob along a slider track. To achieve this, our ACS has three designs. 1) A Concept Sliding Loss based on Linear Discriminant Analysis to pinpoint the concept-specific axis for precise editing. 2) An Attribute Preserving Loss based on Principal Component Analysis for improved preservation of avatar identity during editing. 3) A 3D Gaussian Splatting primitive selection mechanism based on concept-sensitivity, which updates only the primitives that are the most sensitive to our target concept, to improve efficiency. Results demonstrate that our ACS enables fine-grained 3D avatar editing with efficient feedback, without harming the avatar quality or compromising the avatar's identifying attributes.
Authors: Qihang Ge, Wei Sun, Yu Zhang, Yunhao Li, Zhongpeng Ji, Fengyu Sun, Shangling Jui, Xiongkuo Min, Guangtao Zhai
Abstract: The explosive growth of videos on streaming media platforms has underscored the urgent need for effective video quality assessment (VQA) algorithms to monitor and perceptually optimize the quality of streaming videos. However, VQA remains an extremely challenging task due to the diverse video content and the complex spatial and temporal distortions, thus necessitating more advanced methods to address these issues. Nowadays, large multimodal models (LMMs), such as GPT-4V, have exhibited strong capabilities for various visual understanding tasks, motivating us to leverage the powerful multimodal representation ability of LMMs to solve the VQA task. Therefore, we propose the first Large Multi-Modal Video Quality Assessment (LMM-VQA) model, which introduces a novel spatiotemporal visual modeling strategy for quality-aware feature extraction. Specifically, we first reformulate the quality regression problem into a question and answering (Q&A) task and construct Q&A prompts for VQA instruction tuning. Then, we design a spatiotemporal vision encoder to extract spatial and temporal features to represent the quality characteristics of videos, which are subsequently mapped into the language space by the spatiotemporal projector for modality alignment. Finally, the aligned visual tokens and the quality-inquired text tokens are aggregated as inputs for the large language model (LLM) to generate the quality score and level. Extensive experiments demonstrate that LMM-VQA achieves state-of-the-art performance across five VQA benchmarks, exhibiting an average improvement of $5\%$ in generalization ability over existing methods. Furthermore, due to the advanced design of the spatiotemporal encoder and projector, LMM-VQA also performs exceptionally well on general video understanding tasks, further validating its effectiveness. Our code will be released at https://github.com/Sueqk/LMM-VQA.
Authors: Zhuoyue Wang, Yiyi Tao, Danqing Ma
Abstract: In this paper, a color edge detection strategy based on collaborative filtering combined with multiscale gradient fusion is proposed. The block-matching and 3D (BM3D) filter are used to enhance the sparse representation in the transform domain and achieve the effect of denoising, whereas the multiscale gradient fusion makes up for the defect of loss of details in single-scale edge detection and improves the edge detection resolution and quality. First, the RGB images in the dataset are converted to XYZ color space images through mathematical operations. Second, the colored block-matching and 3D (CBM3D) filter are used on the sparse images and to remove noise interference. Then, the vector gradients of the color image and the anisotropic Gaussian directional derivative of the two scale parameters are calculated and averaged pixel-by-pixel to obtain a new edge strength map. Finally, the edge features are enhanced by image normalization and non-maximum suppression technology, and on that basis, the edge contour is obtained by double threshold selection and a new morphological refinement method. Through an experimental analysis of the edge detection dataset, the method proposed has good noise robustness and high edge quality, which is better than the Color Sobel, Color Canny, SE and Color AGDD as shown by the PR curve, AUC, PSNR, MSE, and FOM indicators.
Authors: Zhenggang Tang, Peiye Zhuang, Chaoyang Wang, Aliaksandr Siarohin, Yash Kant, Alexander Schwing, Sergey Tulyakov, Hsin-Ying Lee
Abstract: The task of image-to-multi-view generation refers to generating novel views of an instance from a single image. Recent methods achieve this by extending text-to-image latent diffusion models to multi-view version, which contains an VAE image encoder and a U-Net diffusion model. Specifically, these generation methods usually fix VAE and finetune the U-Net only. However, the significant downscaling of the latent vectors computed from the input images and independent decoding leads to notable pixel-level misalignment across multiple views. To address this, we propose a novel method for pixel-level image-to-multi-view generation. Unlike prior work, we incorporate attention layers across multi-view images in the VAE decoder of a latent video diffusion model. Specifically, we introduce a depth-truncated epipolar attention, enabling the model to focus on spatially adjacent regions while remaining memory efficient. Applying depth-truncated attn is challenging during inference as the ground-truth depth is usually difficult to obtain and pre-trained depth estimation models is hard to provide accurate depth. Thus, to enhance the generalization to inaccurate depth when ground truth depth is missing, we perturb depth inputs during training. During inference, we employ a rapid multi-view to 3D reconstruction approach, NeuS, to obtain coarse depth for the depth-truncated epipolar attention. Our model enables better pixel alignment across multi-view images. Moreover, we demonstrate the efficacy of our approach in improving downstream multi-view to 3D reconstruction tasks.
Authors: Jiajun Fei, Dian Li, Zhidong Deng, Zekun Wang, Gang Liu, Hui Wang
Abstract: Multi-modal large language models (MLLMs) have demonstrated considerable potential across various downstream tasks that require cross-domain knowledge. MLLMs capable of processing videos, known as Video-MLLMs, have attracted broad interest in video-language understanding. However, videos, especially long videos, contain more visual tokens than images, making them difficult for LLMs to process. Existing works either downsample visual features or extend the LLM context size, risking the loss of high-resolution information or slowing down inference speed. To address these limitations, we apply cross-attention layers in the intermediate projector between the visual encoder and the large language model (LLM). As the naive cross-attention mechanism is insensitive to temporal order, we further introduce causal cross-attention masks (CCAMs) within the cross-attention layers. This Video-MLLM, named Video-CCAM, is trained in a straightforward two-stage fashion: feature alignment and visual instruction tuning. We develop several Video-CCAM models based on LLMs of different sizes (4B, 9B, and 14B). Video-CCAM proves to be a robust Video-MLLM and shows outstanding performance from short videos to long ones. Among standard video benchmarks like MVBench and VideoChatGPT-QA, Video-CCAM shows outstanding performances (1st/2nd/3rd in MVBench and TGIF-QA, 2nd/3rd/4th in MSVD-QA, MSRVTT-QA, and ActivityNet-QA). In benchmarks encompassing long videos, Video-CCAM models can be directly adapted to long video understanding and still achieve exceptional scores despite being trained solely with images and 16-frame videos. Using 96 frames (6$\times$ the training number of frames), Video-CCAM models rank 1st/2nd/3rd in VideoVista and 1st/2nd/4th in MLVU among all open-source Video-MLLMs, respectively. The code is publicly available in \url{https://github.com/QQ-MM/Video-CCAM}.
Authors: Joseph Cho, Samuel Schmidgall, Cyril Zakka, Mrudang Mathur, Rohan Shad, William Hiesinger
Abstract: Diffusion-based video generation models have made significant strides, producing outputs with improved visual fidelity, temporal coherence, and user control. These advancements hold great promise for improving surgical education by enabling more realistic, diverse, and interactive simulation environments. In this study, we introduce SurGen, a text-guided diffusion model tailored for surgical video synthesis, producing the highest resolution and longest duration videos among existing surgical video generation models. We validate the visual and temporal quality of the outputs using standard image and video generation metrics. Additionally, we assess their alignment to the corresponding text prompts through a deep learning classifier trained on surgical data. Our results demonstrate the potential of diffusion models to serve as valuable educational tools for surgical trainees.
Authors: Bingcheng Dong, Yuning Ding, Jinrong Zhang, Sifan Zhang, Shenglan Liu
Abstract: Open Set Object Detection has seen rapid development recently, but it continues to pose significant challenges. Language-based methods, grappling with the substantial modal disparity between textual and visual modalities, require extensive computational resources to bridge this gap. Although integrating visual prompts into these frameworks shows promise for enhancing performance, it always comes with constraints related to textual semantics. In contrast, viusal-only methods suffer from the low-quality fusion of multiple visual prompts. In response, we introduce a strong DETR-based model, Visual Intersection Network for Open Set Object Detection (VINO), which constructs a multi-image visual bank to preserve the semantic intersections of each category across all time steps. Our innovative multi-image visual updating mechanism learns to identify the semantic intersections from various visual prompts, enabling the flexible incorporation of new information and continuous optimization of feature representations. Our approach guarantees a more precise alignment between target category semantics and region semantics, while significantly reducing pre-training time and resource demands compared to language-based methods. Furthermore, the integration of a segmentation head illustrates the broad applicability of visual intersection in various visual tasks. VINO, which requires only 7 RTX4090 GPU days to complete one epoch on the Objects365v1 dataset, achieves competitive performance on par with vision-language models on benchmarks such as LVIS and ODinW35.
Authors: Zhenghao Feng, Lu Wen, Binyu Yan, Jiaqi Cui, Yan Wang
Abstract: Semi-supervised learning (SSL) has shown notable potential in relieving the heavy demand of dense prediction tasks on large-scale well-annotated datasets, especially for the challenging multi-organ segmentation (MoS). However, the prevailing class-imbalance problem in MoS, caused by the substantial variations in organ size, exacerbates the learning difficulty of the SSL network. To alleviate this issue, we present a two-phase semi-supervised network (BSR-Net) with balanced subclass regularization for MoS. Concretely, in Phase I, we introduce a class-balanced subclass generation strategy based on balanced clustering to effectively generate multiple balanced subclasses from original biased ones according to their pixel proportions. Then, in Phase II, we design an auxiliary subclass segmentation (SCS) task within the multi-task framework of the main MoS task. The SCS task contributes a balanced subclass regularization to the main MoS task and transfers unbiased knowledge to the MoS network, thus alleviating the influence of the class-imbalance problem. Extensive experiments conducted on two publicly available datasets, i.e., the MICCAI FLARE 2022 dataset and the WORD dataset, verify the superior performance of our method compared with other methods.
Authors: Yuncheng Jiang, Zixun Zhang, Jun Wei, Chun-Mei Feng, Guanbin Li, Xiang Wan, Shuguang Cui, Zhen Li
Abstract: AI-assisted lesion detection models play a crucial role in the early screening of cancer. However, previous image-based models ignore the inter-frame contextual information present in videos. On the other hand, video-based models capture the inter-frame context but are computationally expensive. To mitigate this contradiction, we delve into Video-to-Image knowledge distillation leveraging DEtection TRansformer (V2I-DETR) for the task of medical video lesion detection. V2I-DETR adopts a teacher-student network paradigm. The teacher network aims at extracting temporal contexts from multiple frames and transferring them to the student network, and the student network is an image-based model dedicated to fast prediction in inference. By distilling multi-frame contexts into a single frame, the proposed V2I-DETR combines the advantages of utilizing temporal contexts from video-based models and the inference speed of image-based models. Through extensive experiments, V2I-DETR outperforms previous state-of-the-art methods by a large margin while achieving the real-time inference speed (30 FPS) as the image-based model.
Authors: Shichen Liu, Huaxing Lu
Abstract: This paper employs deep learning methods to investigate the visual similarity of ethnic minority patterns in Southwest China. A customized SResNet-18 network was developed, achieving an accuracy of 98.7% on the test set, outperforming ResNet-18, VGGNet-16, and AlexNet. The extracted feature vectors from SResNet-18 were evaluated using three metrics: cosine similarity, Euclidean distance, and Manhattan distance. The analysis results were visually represented on an ethnic thematic map, highlighting the connections between ethnic patterns and their regional distributions.
Authors: Xiaoyu Yuan, Xiaohua Huang, Zibo Zhang, Yabo Sun
Abstract: The Houma Alliance Book, one of history's earliest calligraphic examples, was unearthed in the 1970s. These artifacts were meticulously organized, reproduced, and copied by the Shanxi Provincial Institute of Cultural Relics. However, because of their ancient origins and severe ink erosion, identifying characters in the Houma Alliance Book is challenging, necessitating the use of digital technology. In this paper, we propose a new ancient handwritten character recognition database for the Houma alliance book, along with a novel benchmark based on deep learning architectures. More specifically, a collection of 26,732 characters samples from the Houma Alliance Book were gathered, encompassing 327 different types of ancient characters through iterative annotation. Furthermore, benchmark algorithms were proposed by combining four deep neural network classifiers with two data augmentation methods. This research provides valuable resources and technical support for further studies on the Houma Alliance Book and other ancient characters. This contributes to our understanding of ancient culture and history, as well as the preservation and inheritance of humanity's cultural heritage.
Authors: Zhongwen Yu, Qiu Guan, Jianmin Yang, Zhiqiang Yang, Qianwei Zhou, Yang Chen, Feng Chen
Abstract: In existing medical Region of Interest (ROI) detection, there lacks an algorithm that can simultaneously satisfy both real-time performance and accuracy, not meeting the growing demand for automatic detection in medicine. Although the basic YOLO framework ensures real-time detection due to its fast speed, it still faces challenges in maintaining precision concurrently. To alleviate the above problems, we propose a novel model named Lightweight Shunt Matching-YOLO (LSM-YOLO), with Lightweight Adaptive Extraction (LAE) and Multipath Shunt Feature Matching (MSFM). Firstly, by using LAE to refine feature extraction, the model can obtain more contextual information and high-resolution details from multiscale feature maps, thereby extracting detailed features of ROI in medical images while reducing the influence of noise. Secondly, MSFM is utilized to further refine the fusion of high-level semantic features and low-level visual features, enabling better fusion between ROI features and neighboring features, thereby improving the detection rate for better diagnostic assistance. Experimental results demonstrate that LSM-YOLO achieves 48.6% AP on a private dataset of pancreatic tumors, 65.1% AP on the BCCD blood cell detection public dataset, and 73.0% AP on the Br35h brain tumor detection public dataset. Our model achieves state-of-the-art performance with minimal parameter cost on the above three datasets. The source codes are at: https://github.com/VincentYuuuuuu/LSM-YOLO.
Authors: Abu Saleh Musa Miah, Md. Al Mehedi Hasan, Md Hadiuzzaman, Muhammad Nazrul Islam, Jungpil Shin
Abstract: Hand gesture-based sign language recognition (SLR) is one of the most advanced applications of machine learning, and computer vision uses hand gestures. Although, in the past few years, many researchers have widely explored and studied how to address BSL problems, specific unaddressed issues remain, such as skeleton and transformer-based BSL recognition. In addition, the lack of evaluation of the BSL model in various concealed environmental conditions can prove the generalized property of the existing model by facing daily life signs. As a consequence, existing BSL recognition systems provide a limited perspective of their generalisation ability as they are tested on datasets containing few BSL alphabets that have a wide disparity in gestures and are easy to differentiate. To overcome these limitations, we propose a spatial-temporal attention-based BSL recognition model considering hand joint skeletons extracted from the sequence of images. The main aim of utilising hand skeleton-based BSL data is to ensure the privacy and low-resolution sequence of images, which need minimum computational cost and low hardware configurations. Our model captures discriminative structural displacements and short-range dependency based on unified joint features projected onto high-dimensional feature space. Specifically, the use of Separable TCN combined with a powerful multi-head spatial-temporal attention architecture generated high-performance accuracy. The extensive experiments with a proposed dataset and two benchmark BSL datasets with a wide range of evaluations, such as intra- and inter-dataset evaluation settings, demonstrated that our proposed models achieve competitive performance with extremely low computational complexity and run faster than existing models.
Authors: Ruohua Shi, Qiufan Pang, Lei Ma, Lingyu Duan, Tiejun Huang, Tingting Jiang
Abstract: Electron microscopy (EM) imaging offers unparalleled resolution for analyzing neural tissues, crucial for uncovering the intricacies of synaptic connections and neural processes fundamental to understanding behavioral mechanisms. Recently, the foundation models have demonstrated impressive performance across numerous natural and medical image segmentation tasks. However, applying these foundation models to EM segmentation faces significant challenges due to domain disparities. This paper presents ShapeMamba-EM, a specialized fine-tuning method for 3D EM segmentation, which employs adapters for long-range dependency modeling and an encoder for local shape description within the original foundation model. This approach effectively addresses the unique volumetric and morphological complexities of EM data. Tested over a wide range of EM images, covering five segmentation tasks and 10 datasets, ShapeMamba-EM outperforms existing methods, establishing a new standard in EM image segmentation and enhancing the understanding of neural tissue architecture.
Authors: Sven Oehri, Nikolas Ebert, Ahmed Abdullah, Didier Stricker, Oliver Wasenm\"uller
Abstract: Recent studies showcase the competitive accuracy of Vision Transformers (ViTs) in relation to Convolutional Neural Networks (CNNs), along with their remarkable robustness. However, ViTs demand a large amount of data to achieve adequate performance, which makes their application to small datasets challenging, falling behind CNNs. To overcome this, we propose GenFormer, a data augmentation strategy utilizing generated images, thereby improving transformer accuracy and robustness on small-scale image classification tasks. In our comprehensive evaluation we propose Tiny ImageNetV2, -R, and -A as new test set variants of Tiny ImageNet by transferring established ImageNet generalization and robustness benchmarks to the small-scale data domain. Similarly, we introduce MedMNIST-C and EuroSAT-C as corrupted test set variants of established fine-grained datasets in the medical and aerial domain. Through a series of experiments conducted on small datasets of various domains, including Tiny ImageNet, CIFAR, EuroSAT and MedMNIST datasets, we demonstrate the synergistic power of our method, in particular when combined with common train and test time augmentations, knowledge distillation, and architectural design choices. Additionally, we prove the effectiveness of our approach under challenging conditions with limited training data, demonstrating significant improvements in both accuracy and robustness, bridging the gap between CNNs and ViTs in the small-scale dataset domain.
Authors: Chaohua Shi, Xuan Wang, Si Shi, Xule Wang, Mingrui Zhu, Nannan Wang, Xinbo Gao
Abstract: Food image composition requires the use of existing dish images and background images to synthesize a natural new image, while diffusion models have made significant advancements in image generation, enabling the construction of end-to-end architectures that yield promising results. However, existing diffusion models face challenges in processing and fusing information from multiple images and lack access to high-quality publicly available datasets, which prevents the application of diffusion models in food image composition. In this paper, we introduce a large-scale, high-quality food image composite dataset, FC22k, which comprises 22,000 foreground, background, and ground truth ternary image pairs. Additionally, we propose a novel food image composition method, Foodfusion, which leverages the capabilities of the pre-trained diffusion models and incorporates a Fusion Module for processing and integrating foreground and background information. This fused information aligns the foreground features with the background structure by merging the global structural information at the cross-attention layer of the denoising UNet. To further enhance the content and structure of the background, we also integrate a Content-Structure Control Module. Extensive experiments demonstrate the effectiveness and scalability of our proposed method.
Authors: Chiara Galdi, Michele Panariello, Massimiliano Todisco, Nicholas Evans
Abstract: We introduce 2D-Malafide, a novel and lightweight adversarial attack designed to deceive face deepfake detection systems. Building upon the concept of 1D convolutional perturbations explored in the speech domain, our method leverages 2D convolutional filters to craft perturbations which significantly degrade the performance of state-of-the-art face deepfake detectors. Unlike traditional additive noise approaches, 2D-Malafide optimises a small number of filter coefficients to generate robust adversarial perturbations which are transferable across different face images. Experiments, conducted using the FaceForensics++ dataset, demonstrate that 2D-Malafide substantially degrades detection performance in both white-box and black-box settings, with larger filter sizes having the greatest impact. Additionally, we report an explainability analysis using GradCAM which illustrates how 2D-Malafide misleads detection systems by altering the image areas used most for classification. Our findings highlight the vulnerability of current deepfake detection systems to convolutional adversarial attacks as well as the need for future work to enhance detection robustness through improved image fidelity constraints.
Authors: Hae Jin Song, Patrycja Krawczuk, Po-Hsuan Huang
Abstract: Geospatial data come from various sources, such as satellites, aircraft, and LiDAR. The variability of the source is not limited to the types of data acquisition techniques, as we have maps from different time periods. To incorporate these data for a coherent analysis, it is essential to first align different "styles" of geospatial data to its matching images that point to the same location on the surface of the Earth. In this paper, we approach the image registration as a two-step process of (1) extracting geospatial contents invariant to visual (and any other non-content-related) information, and (2) matching the data based on such (purely) geospatial contents. We hypothesize that a combination of $\beta$-VAE-like architecture [2] and adversarial training will achieve both the disentanglement of the geographic information and artistic styles and generation of new map tiles by composing the encoded geographic information with any artistic style.
Authors: Lucas M\"oller, Pascal Tilli, Ngoc Thang Vu, Sebastian Pad\'o
Abstract: Dual encoder architectures like CLIP models map two types of inputs into a shared embedding space and learn similarities between them. However, it is not understood how such models compare two inputs. Here, we address this research gap with two contributions. First, we derive a method to attribute predictions of any differentiable dual encoder onto feature-pair interactions between its inputs. Second, we apply our method to CLIP-type models and show that they learn fine-grained correspondences between parts of captions and regions in images. They match objects across input modes and also account for mismatches. However, this visual-linguistic grounding ability heavily varies between object classes, depends on the training data distribution, and largely improves after in-domain training. Using our method we can identify knowledge gaps about specific object classes in individual models and can monitor their improvement upon fine-tuning.
Authors: Ombretta Strafforello, Gonzalo Muradas Odriozola, Fatemeh Behrad, Li-Wei Chen, Anne-Sofie Maerten, Derya Soydaner, Johan Wagemans
Abstract: Assessing the aesthetic quality of artistic images presents unique challenges due to the subjective nature of aesthetics and the complex visual characteristics inherent to artworks. Basic data augmentation techniques commonly applied to natural images in computer vision may not be suitable for art images in aesthetic evaluation tasks, as they can change the composition of the art images. In this paper, we explore the impact of local and global data augmentation techniques on artistic image aesthetic assessment (IAA). We introduce BackFlip, a local data augmentation technique designed specifically for artistic IAA. We evaluate the performance of BackFlip across three artistic image datasets and four neural network architectures, comparing it with the commonly used data augmentation techniques. Then, we analyze the effects of components within the BackFlip pipeline through an ablation study. Our findings demonstrate that local augmentations, such as BackFlip, tend to outperform global augmentations on artistic IAA in most cases, probably because they do not perturb the composition of the art images. These results emphasize the importance of considering both local and global augmentations in future computational aesthetics research.
Authors: Trung Dao, Thuan Hoang Nguyen, Thanh Le, Duc Vu, Khoi Nguyen, Cuong Pham, Anh Tran
Abstract: In this paper, we aim to enhance the performance of SwiftBrush, a prominent one-step text-to-image diffusion model, to be competitive with its multi-step Stable Diffusion counterpart. Initially, we explore the quality-diversity trade-off between SwiftBrush and SD Turbo: the former excels in image diversity, while the latter excels in image quality. This observation motivates our proposed modifications in the training methodology, including better weight initialization and efficient LoRA training. Moreover, our introduction of a novel clamped CLIP loss enhances image-text alignment and results in improved image quality. Remarkably, by combining the weights of models trained with efficient LoRA and full training, we achieve a new state-of-the-art one-step diffusion model, achieving an FID of 8.14 and surpassing all GAN-based and multi-step Stable Diffusion models. The evaluation code is available at: https://github.com/vinairesearch/swiftbrushv2.
Authors: Albert Luginov, Muhammad Shahzad
Abstract: We introduce NimbleD, an efficient self-supervised monocular depth estimation learning framework that incorporates supervision from pseudo-labels generated by a large vision model. This framework does not require camera intrinsics, enabling large-scale pre-training on publicly available videos. Our straightforward yet effective learning strategy significantly enhances the performance of fast and lightweight models without introducing any overhead, allowing them to achieve performance comparable to state-of-the-art self-supervised monocular depth estimation models. This advancement is particularly beneficial for virtual and augmented reality applications requiring low latency inference. The source code, model weights, and acknowledgments are available at https://github.com/xapaxca/nimbled .
Authors: Yiwei Ma, Jiayi Ji, Ke Ye, Weihuang Lin, Zhibin Wang, Yonghan Zheng, Qiang Zhou, Xiaoshuai Sun, Rongrong Ji
Abstract: Significant progress has been made in the field of Instruction-based Image Editing (IIE). However, evaluating these models poses a significant challenge. A crucial requirement in this field is the establishment of a comprehensive evaluation benchmark for accurately assessing editing results and providing valuable insights for its further development. In response to this need, we propose I2EBench, a comprehensive benchmark designed to automatically evaluate the quality of edited images produced by IIE models from multiple dimensions. I2EBench consists of 2,000+ images for editing, along with 4,000+ corresponding original and diverse instructions. It offers three distinctive characteristics: 1) Comprehensive Evaluation Dimensions: I2EBench comprises 16 evaluation dimensions that cover both high-level and low-level aspects, providing a comprehensive assessment of each IIE model. 2) Human Perception Alignment: To ensure the alignment of our benchmark with human perception, we conducted an extensive user study for each evaluation dimension. 3) Valuable Research Insights: By analyzing the advantages and disadvantages of existing IIE models across the 16 dimensions, we offer valuable research insights to guide future development in the field. We will open-source I2EBench, including all instructions, input images, human annotations, edited images from all evaluated methods, and a simple script for evaluating the results from new IIE models. The code, dataset and generated images from all IIE models are provided in github: https://github.com/cocoshe/I2EBench.
Authors: Georg B\"okman, Johan Edstedt, Michael Felsberg, Fredrik Kahl
Abstract: We propose a way to train deep learning based keypoint descriptors that makes them approximately equivariant for locally affine transformations of the image plane. The main idea is to use the representation theory of GL(2) to generalize the recently introduced concept of steerers from rotations to affine transformations. Affine steerers give high control over how keypoint descriptions transform under image transformations. We demonstrate the potential of using this control for image matching. Finally, we propose a way to finetune keypoint descriptors with a set of steerers on upright images and obtain state-of-the-art results on several standard benchmarks. Code will be published at github.com/georg-bn/affine-steerers.
Authors: Jiasong Feng, Lichun Wang, Hongbo Xu, Kai Xu, Baocai Yin
Abstract: Scene Graph Generation (SGG) aims to generate a comprehensive graphical representation that accurately captures the semantic information of a given scenario. However, the SGG model's performance in predicting more fine-grained predicates is hindered by a significant predicate bias. According to existing works, the long-tail distribution of predicates in training data results in the biased scene graph. However, the semantic overlap between predicate categories makes predicate prediction difficult, and there is a significant difference in the sample size of semantically similar predicates, making the predicate prediction more difficult. Therefore, higher requirements are placed on the discriminative ability of the model. In order to address this problem, this paper proposes Ensemble Predicate Decoding (EPD), which employs multiple decoders to attain unbiased scene graph generation. Two auxiliary decoders trained on lower-frequency predicates are used to improve the discriminative ability of the model. Extensive experiments are conducted on the VG, and the experiment results show that EPD enhances the model's representation capability for predicates. In addition, we find that our approach ensures a relatively superior predictive capability for more frequent predicates compared to previous unbiased SGG methods.
Authors: Pengyu Li, Chenhe Liu, Tengfei Li, Xinyu Wang, Shihui Zhang, Dongyang Yu
Abstract: The detection of small objects, particularly traffic signs, is a critical subtask within object detection and autonomous driving. Despite the notable achievements in previous research, two primary challenges persist. Firstly, the main issue is the singleness of feature extraction. Secondly, the detection process fails to effectively integrate with objects of varying sizes or scales. These issues are also prevalent in generic object detection. Motivated by these challenges, in this paper, we propose a novel object detection network named Efficient Multi-scale and Diverse Feature Network (EMDFNet) for traffic sign detection that integrates an Augmented Shortcut Module and an Efficient Hybrid Encoder to address the aforementioned issues simultaneously. Specifically, the Augmented Shortcut Module utilizes multiple branches to integrate various spatial semantic information and channel semantic information, thereby enhancing feature diversity. The Efficient Hybrid Encoder utilizes global feature fusion and local feature interaction based on various features to generate distinctive classification features by integrating feature information in an adaptable manner. Extensive experiments on the Tsinghua-Tencent 100K (TT100K) benchmark and the German Traffic Sign Detection Benchmark (GTSDB) demonstrate that our EMDFNet outperforms other state-of-the-art detectors in performance while retaining the real-time processing capabilities of single-stage models. This substantiates the effectiveness of EMDFNet in detecting small traffic signs.
Authors: Bingchen Yan
Abstract: Few-shot classification involves identifying new categories using a limited number of labeled samples. Current few-shot classification methods based on local descriptors primarily leverage underlying consistent features across visible and invisible classes, facing challenges including redundant neighboring information, noisy representations, and limited interpretability. This paper proposes a Feature Aligning Few-shot Learning Method Using Local Descriptors Weighted Rules (FAFD-LDWR). It innovatively introduces a cross-normalization method into few-shot image classification to preserve the discriminative information of local descriptors as much as possible; and enhances classification performance by aligning key local descriptors of support and query sets to remove background noise. FAFD-LDWR performs excellently on three benchmark datasets , outperforming state-of-the-art methods in both 1-shot and 5-shot settings. The designed visualization experiments also demonstrate FAFD-LDWR's improvement in prediction interpretability.
Authors: Yu Yang, Jianbiao Mei, Yukai Ma, Siliang Du, Wenqing Chen, Yijie Qian, Yuxiang Feng, Yong Liu
Abstract: World models envision potential future states based on various ego actions. They embed extensive knowledge about the driving environment, facilitating safe and scalable autonomous driving. Most existing methods primarily focus on either data generation or the pretraining paradigms of world models. Unlike the aforementioned prior works, we propose Drive-OccWorld, which adapts a vision-centric 4D forecasting world model to end-to-end planning for autonomous driving. Specifically, we first introduce a semantic and motion-conditional normalization in the memory module, which accumulates semantic and dynamic information from historical BEV embeddings. These BEV features are then conveyed to the world decoder for future occupancy and flow forecasting, considering both geometry and spatiotemporal modeling. Additionally, we propose injecting flexible action conditions, such as velocity, steering angle, trajectory, and commands, into the world model to enable controllable generation and facilitate a broader range of downstream applications. Furthermore, we explore integrating the generative capabilities of the 4D world model with end-to-end planning, enabling continuous forecasting of future states and the selection of optimal trajectories using an occupancy-based cost function. Extensive experiments on the nuScenes dataset demonstrate that our method can generate plausible and controllable 4D occupancy, opening new avenues for driving world generation and end-to-end planning.
Authors: Xu He, Xiaoyu Li, Di Kang, Jiangnan Ye, Chaopeng Zhang, Liyang Chen, Xiangjun Gao, Han Zhang, Zhiyong Wu, Haolin Zhuang
Abstract: Existing works in single-image human reconstruction suffer from weak generalizability due to insufficient training data or 3D inconsistencies for a lack of comprehensive multi-view knowledge. In this paper, we introduce MagicMan, a human-specific multi-view diffusion model designed to generate high-quality novel view images from a single reference image. As its core, we leverage a pre-trained 2D diffusion model as the generative prior for generalizability, with the parametric SMPL-X model as the 3D body prior to promote 3D awareness. To tackle the critical challenge of maintaining consistency while achieving dense multi-view generation for improved 3D human reconstruction, we first introduce hybrid multi-view attention to facilitate both efficient and thorough information interchange across different views. Additionally, we present a geometry-aware dual branch to perform concurrent generation in both RGB and normal domains, further enhancing consistency via geometry cues. Last but not least, to address ill-shaped issues arising from inaccurate SMPL-X estimation that conflicts with the reference image, we propose a novel iterative refinement strategy, which progressively optimizes SMPL-X accuracy while enhancing the quality and consistency of the generated multi-views. Extensive experimental results demonstrate that our method significantly outperforms existing approaches in both novel view synthesis and subsequent 3D human reconstruction tasks.
Authors: Anh-Dzung Doan, Vu Minh Hieu Phan, Surabhi Gupta, Markus Wagner, Tat-Jun Chin, Ian Reid
Abstract: Infrared imaging offers resilience against changing lighting conditions by capturing object temperatures. Yet, in few scenarios, its lack of visual details compared to daytime visible images, poses a significant challenge for human and machine interpretation. This paper proposes a novel diffusion method, dubbed Temporally Consistent Patch Diffusion Models (TC-DPM), for infrared-to-visible video translation. Our method, extending the Patch Diffusion Model, consists of two key components. Firstly, we propose a semantic-guided denoising, leveraging the strong representations of foundational models. As such, our method faithfully preserves the semantic structure of generated visible images. Secondly, we propose a novel temporal blending module to guide the denoising trajectory, ensuring the temporal consistency between consecutive frames. Experiment shows that TC-PDM outperforms state-of-the-art methods by 35.3% in FVD for infrared-to-visible video translation and by 6.1% in AP50 for day-to-night object detection. Our code is publicly available at https://github.com/dzungdoan6/tc-pdm
Authors: Leonid Erlygin, Alexey Zaytsev
Abstract: Accurately estimating image quality and model robustness improvement are critical challenges in unconstrained face recognition, which can be addressed through uncertainty estimation via probabilistic face embeddings. Previous research mainly focused on uncertainty estimation in face verification, leaving the open-set face recognition task underexplored. In open-set face recognition, one seeks to classify an image, which could also be unknown. Here, the low variance of probabilistic embedding does not imply a low error probability: an image embedding could be close to several classes in a gallery, thus yielding high uncertainty. We propose a method aware of two sources of ambiguity in the open-set recognition system: (1) the gallery uncertainty caused by overlapping classes and (2) the uncertainty of the face embeddings. To detect both types, we use a Bayesian probabilistic model of embedding distribution, which provides a principled uncertainty estimate. Challenging open-set face recognition datasets, such as IJB-C, serve as a testbed for our method. We also propose a new open-set recognition protocol for whale and dolphin identification. The proposed approach better identifies recognition errors than uncertainty estimation methods based solely on image quality.
Authors: Hao Li, Jiangxin Dong, Jinshan Pan
Abstract: Existing video super-resolution (VSR) methods generally adopt a recurrent propagation network to extract spatio-temporal information from the entire video sequences, exhibiting impressive performance. However, the key components in recurrent-based VSR networks significantly impact model efficiency, e.g., the alignment module occupies a substantial portion of model parameters, while the bidirectional propagation mechanism significantly amplifies the inference time. Consequently, developing a compact and efficient VSR method that can be deployed on resource-constrained devices, e.g., smartphones, remains challenging. To this end, we propose a cascaded temporal updating network (CTUN) for efficient VSR. We first develop an implicit cascaded alignment module to explore spatio-temporal correspondences from adjacent frames. Moreover, we propose a unidirectional propagation updating network to efficiently explore long-range temporal information, which is crucial for high-quality video reconstruction. Specifically, we develop a simple yet effective hidden updater that can leverage future information to update hidden features during forward propagation, significantly reducing inference time while maintaining performance. Finally, we formulate all of these components into an end-to-end trainable VSR network. Extensive experimental results show that our CTUN achieves a favorable trade-off between efficiency and performance compared to existing methods. Notably, compared with BasicVSR, our method obtains better results while employing only about 30% of the parameters and running time. The source code and pre-trained models will be available at https://github.com/House-Leo/CTUN.
Authors: Vishal Chudasama, Hiran Sarkar, Pankaj Wasnik, Vineeth N Balasubramanian, Jayateja Kalla
Abstract: Object detection is a critical field in computer vision focusing on accurately identifying and locating specific objects in images or videos. Traditional methods for object detection rely on large labeled training datasets for each object category, which can be time-consuming and expensive to collect and annotate. To address this issue, researchers have introduced few-shot object detection (FSOD) approaches that merge few-shot learning and object detection principles. These approaches allow models to quickly adapt to new object categories with only a few annotated samples. While traditional FSOD methods have been studied before, this survey paper comprehensively reviews FSOD research with a specific focus on covering different FSOD settings such as standard FSOD, generalized FSOD, incremental FSOD, open-set FSOD, and domain adaptive FSOD. These approaches play a vital role in reducing the reliance on extensive labeled datasets, particularly as the need for efficient machine learning models continues to rise. This survey paper aims to provide a comprehensive understanding of the above-mentioned few-shot settings and explore the methodologies for each FSOD task. It thoroughly compares state-of-the-art methods across different FSOD settings, analyzing them in detail based on their evaluation protocols. Additionally, it offers insights into their applications, challenges, and potential future directions in the evolving field of object detection with limited data.
Authors: Laurenz Reichardt, Luca Uhr, Oliver Wasenm\"uller
Abstract: LiDAR data of urban scenarios poses unique challenges, such as heterogeneous characteristics and inherent class imbalance. Therefore, large-scale datasets are necessary to apply deep learning methods. Instance augmentation has emerged as an efficient method to increase dataset diversity. However, current methods require the time-consuming curation of 3D models or costly manual data annotation. To overcome these limitations, we propose Text3DAug, a novel approach leveraging generative models for instance augmentation. Text3DAug does not depend on labeled data and is the first of its kind to generate instances and annotations from text. This allows for a fully automated pipeline, eliminating the need for manual effort in practical applications. Additionally, Text3DAug is sensor agnostic and can be applied regardless of the LiDAR sensor used. Comprehensive experimental analysis on LiDAR segmentation, detection and novel class discovery demonstrates that Text3DAug is effective in supplementing existing methods or as a standalone method, performing on par or better than established methods, however while overcoming their specific drawbacks. The code is publicly available.
Authors: Chao Chen, Zhizhong Han, Yu-Shen Liu
Abstract: It is challenging to reconstruct 3D point clouds in unseen classes from single 2D images. Instead of object-centered coordinate system, current methods generalized global priors learned in seen classes to reconstruct 3D shapes from unseen classes in viewer-centered coordinate system. However, the reconstruction accuracy and interpretability are still eager to get improved. To resolve this issue, we introduce to learn local pattern modularization for reconstructing 3D shapes in unseen classes, which achieves both good generalization ability and high reconstruction accuracy. Our insight is to learn a local prior which is class-agnostic and easy to generalize in object-centered coordinate system. Specifically, the local prior is learned via a process of learning and customizing local pattern modularization in seen classes. During this process, we first learn a set of patterns in local regions, which is the basis in the object-centered coordinate system to represent an arbitrary region on shapes across different classes. Then, we modularize each region on an initially reconstructed shape using the learned local patterns. Based on that, we customize the local pattern modularization using the input image by refining the reconstruction with more details. Our method enables to reconstruct high fidelity point clouds from unseen classes in object-centered coordinate system without requiring a large number of patterns or any additional information, such as segmentation supervision or camera poses. Our experimental results under widely used benchmarks show that our method achieves the state-of-the-art reconstruction accuracy for shapes from unseen classes. The code is available at https://github.com/chenchao15/Unseen.
Authors: Weide Liu, Camilo Calixto, Simon K. Warfield, Davood Karimi
Abstract: Diffusion-weighted magnetic resonance imaging (dMRI) is the only non-invasive tool for studying white matter tracts and structural connectivity of the brain. These assessments rely heavily on tractography techniques, which reconstruct virtual streamlines representing white matter fibers. Much effort has been devoted to improving tractography methodology for adult brains, while tractography of the fetal brain has been largely neglected. Fetal tractography faces unique difficulties due to low dMRI signal quality, immature and rapidly developing brain structures, and paucity of reference data. This work presents the first machine learning model for fetal tractography. The model input consists of five sources of information: (1) Fiber orientation, inferred from a diffusion tensor fit to the dMRI signal; (2) Directions of recent propagation steps; (3) Global spatial information, encoded as distances to keypoints in the brain cortex; (4) Tissue segmentation information; and (5) Prior information about the expected local fiber orientations supplied with an atlas. In order to mitigate the local tensor estimation error, a large spatial context around the current point in the diffusion tensor image is encoded using convolutional and attention neural network modules. Moreover, the diffusion tensor information at a hypothetical next point is included in the model input. Filtering rules based on anatomically constrained tractography are applied to prune implausible streamlines. We trained the model on manually-refined whole-brain fetal tractograms and validated the trained model on an independent set of 11 test scans with gestational ages between 23 and 36 weeks. Results show that our proposed method achieves superior performance across all evaluated tracts. The new method can significantly advance the capabilities of dMRI for studying normal and abnormal brain development in utero.
Authors: Ghazal Alinezhad Noghre, Shanle Yao, Armin Danesh Pazho, Babak Rahimi Ardabili, Vinit Katariya, Hamed Tabkhi
Abstract: PHEVA, a Privacy-preserving Human-centric Ethical Video Anomaly detection dataset. By removing pixel information and providing only de-identified human annotations, PHEVA safeguards personally identifiable information. The dataset includes seven indoor/outdoor scenes, featuring one novel, context-specific camera, and offers over 5x the pose-annotated frames compared to the largest previous dataset. This study benchmarks state-of-the-art methods on PHEVA using a comprehensive set of metrics, including the 10% Error Rate (10ER), a metric used for anomaly detection for the first time providing insights relevant to real-world deployment. As the first of its kind, PHEVA bridges the gap between conventional training and real-world deployment by introducing continual learning benchmarks, with models outperforming traditional methods in 82.14% of cases. The dataset is publicly available at https://github.com/TeCSAR-UNCC/PHEVA.git.
Authors: Xindi Wu, Dingli Yu, Yangsibo Huang, Olga Russakovsky, Sanjeev Arora
Abstract: Compositionality is a critical capability in Text-to-Image (T2I) models, as it reflects their ability to understand and combine multiple concepts from text descriptions. Existing evaluations of compositional capability rely heavily on human-designed text prompts or fixed templates, limiting their diversity and complexity, and yielding low discriminative power. We propose ConceptMix, a scalable, controllable, and customizable benchmark which automatically evaluates compositional generation ability of T2I models. This is done in two stages. First, ConceptMix generates the text prompts: concretely, using categories of visual concepts (e.g., objects, colors, shapes, spatial relationships), it randomly samples an object and k-tuples of visual concepts, then uses GPT4-o to generate text prompts for image generation based on these sampled concepts. Second, ConceptMix evaluates the images generated in response to these prompts: concretely, it checks how many of the k concepts actually appeared in the image by generating one question per visual concept and using a strong VLM to answer them. Through administering ConceptMix to a diverse set of T2I models (proprietary as well as open ones) using increasing values of k, we show that our ConceptMix has higher discrimination power than earlier benchmarks. Specifically, ConceptMix reveals that the performance of several models, especially open models, drops dramatically with increased k. Importantly, it also provides insight into the lack of prompt diversity in widely-used training datasets. Additionally, we conduct extensive human studies to validate the design of ConceptMix and compare our automatic grading with human judgement. We hope it will guide future T2I model development.
Authors: Xinrui Zhang, Ailong Cai, Lei Li, Bin Yan
Abstract: Metal artifacts in computed tomography (CT) imaging pose significant challenges to accurate clinical diagnosis. The presence of high-density metallic implants results in artifacts that deteriorate image quality, manifesting in the forms of streaking, blurring, or beam hardening effects, etc. Nowadays, various deep learning-based approaches, particularly generative models, have been proposed for metal artifact reduction (MAR). However, these methods have limited perception ability in the diverse morphologies of different metal implants with artifacts, which may generate spurious anatomical structures and exhibit inferior generalization capability. To address the issues, we leverage visual-language model (VLM) to identify these morphological features and introduce them into a dual-domain CLIP-assisted residual optimization perception model (DuDoCROP) for MAR. Specifically, a dual-domain CLIP (DuDoCLIP) is fine-tuned on the image domain and sinogram domain using contrastive learning to extract semantic descriptions from anatomical structures and metal artifacts. Subsequently, a diffusion model is guided by the embeddings of DuDoCLIP, thereby enabling the dual-domain prior generation. Additionally, we design prompt engineering for more precise image-text descriptions that can enhance the model's perception capability. Then, a downstream task is devised for the one-step residual optimization and integration of dual-domain priors, while incorporating raw data fidelity. Ultimately, a new perceptual indicator is proposed to validate the model's perception and generation performance. With the assistance of DuDoCLIP, our DuDoCROP exhibits at least 63.7% higher generalization capability compared to the baseline model. Numerical experiments demonstrate that the proposed method can generate more realistic image structures and outperform other SOTA approaches both qualitatively and quantitatively.
Authors: Henry Marichal, Gregory Randall
Abstract: This work presents the INBD network proposed by Gillert et al. in CVPR-2023 and studies its application for delineating tree rings in RGB images of Pinus taeda cross sections captured by a smartphone (UruDendro dataset), which are images with different characteristics from the ones used to train the method. The INBD network operates in two stages: first, it segments the background, pith, and ring boundaries. In the second stage, the image is transformed into polar coordinates, and ring boundaries are iteratively segmented from the pith to the bark. Both stages are based on the U-Net architecture. The method achieves an F-Score of 77.5, a mAR of 0.540, and an ARAND of 0.205 on the evaluation set. The code for the experiments is available at https://github.com/hmarichal93/mlbrief_inbd.
Authors: Omiros Pantazis, Peggy Bevan, Holly Pringle, Guilherme Braga Ferreira, Daniel J. Ingram, Emily Madsen, Liam Thomas, Dol Raj Thanet, Thakur Silwal, Santosh Rayamajhi, Gabriel Brostow, Oisin Mac Aodha, Kate E. Jones
Abstract: Large wildlife image collections from camera traps are crucial for biodiversity monitoring, offering insights into species richness, occupancy, and activity patterns. However, manual processing of these data is time-consuming, hindering analytical processes. To address this, deep neural networks have been widely adopted to automate image analysis. Despite their growing use, the impact of model training decisions on downstream ecological metrics remains unclear. Here, we analyse camera trap data from an African savannah and an Asian sub-tropical dry forest to compare key ecological metrics derived from expert-generated species identifications with those generated from deep neural networks. We assess the impact of model architecture, training data noise, and dataset size on ecological metrics, including species richness, occupancy, and activity patterns. Our results show that while model architecture has minimal impact, large amounts of noise and reduced dataset size significantly affect these metrics. Nonetheless, estimated ecological metrics are resilient to considerable noise, tolerating up to 10% error in species labels and a 50% reduction in training set size without changing significantly. We also highlight that conventional metrics like classification error may not always be representative of a model's ability to accurately measure ecological metrics. We conclude that ecological metrics derived from deep neural network predictions closely match those calculated from expert labels and remain robust to variations in the factors explored. However, training decisions for deep neural networks can impact downstream ecological analysis. Therefore, practitioners should prioritize creating large, clean training sets and evaluate deep neural network solutions based on their ability to measure the ecological metrics of interest.
Authors: Francesco Di Salvo, Sebastian Doerrich, Ines Rieger, Christian Ledig
Abstract: The performance of deep neural networks scales with dataset size and label quality, rendering the efficient mitigation of low-quality data annotations crucial for building robust and cost-effective systems. Existing strategies to address label noise exhibit severe limitations due to computational complexity and application dependency. In this work, we propose WANN, a Weighted Adaptive Nearest Neighbor approach that builds on self-supervised feature representations obtained from foundation models. To guide the weighted voting scheme, we introduce a reliability score, which measures the likelihood of a data label being correct. WANN outperforms reference methods, including a linear layer trained with robust loss functions, on diverse datasets of varying size and under various noise types and severities. WANN also exhibits superior generalization on imbalanced data compared to both Adaptive-NNs (ANN) and fixed k-NNs. Furthermore, the proposed weighting scheme enhances supervised dimensionality reduction under noisy labels. This yields a significant boost in classification performance with 10x and 100x smaller image embeddings, minimizing latency and storage requirements. Our approach, emphasizing efficiency and explainability, emerges as a simple, robust solution to overcome the inherent limitations of deep neural network training. The code is available at https://github.com/francescodisalvo05/wann-noisy-labels .
URLs: https://github.com/francescodisalvo05/wann-noisy-labels
Authors: Sarah Rastegar, Mohammadreza Salehi, Yuki M. Asano, Hazel Doughty, Cees G. M. Snoek
Abstract: In this paper, we address Generalized Category Discovery, aiming to simultaneously uncover novel categories and accurately classify known ones. Traditional methods, which lean heavily on self-supervision and contrastive learning, often fall short when distinguishing between fine-grained categories. To address this, we introduce a novel concept called `self-expertise', which enhances the model's ability to recognize subtle differences and uncover unknown categories. Our approach combines unsupervised and supervised self-expertise strategies to refine the model's discernment and generalization. Initially, hierarchical pseudo-labeling is used to provide `soft supervision', improving the effectiveness of self-expertise. Our supervised technique differs from traditional methods by utilizing more abstract positive and negative samples, aiding in the formation of clusters that can generalize to novel categories. Meanwhile, our unsupervised strategy encourages the model to sharpen its category distinctions by considering within-category examples as `hard' negatives. Supported by theoretical insights, our empirical results showcase that our method outperforms existing state-of-the-art techniques in Generalized Category Discovery across several fine-grained datasets. Our code is available at: https://github.com/SarahRastegar/SelEx.
Authors: Vishal Batchu, Alex Wilson, Betty Peng, Carl Elkin, Umangi Jain, Christopher Van Arsdale, Ross Goroshin, Varun Gulshan
Abstract: The transition to renewable energy, particularly solar, is key to mitigating climate change. Google's Solar API aids this transition by estimating solar potential from aerial imagery, but its impact is constrained by geographical coverage. This paper proposes expanding the API's reach using satellite imagery, enabling global solar potential assessment. We tackle challenges involved in building a Digital Surface Model (DSM) and roof instance segmentation from lower resolution and single oblique views using deep learning models. Our models, trained on aligned satellite and aerial datasets, produce 25cm DSMs and roof segments. With ~1m DSM MAE on buildings, ~5deg roof pitch error and ~56% IOU on roof segmentation, they significantly enhance the Solar API's potential to promote solar adoption.
Authors: Trung Dinh Quoc Dang, Huy Hoang Nguyen, Aleksei Tiulpin
Abstract: Mamba, a State Space Model (SSM), has recently shown competitive performance to Convolutional Neural Networks (CNNs) and Transformers in Natural Language Processing and general sequence modeling. Various attempts have been made to adapt Mamba to Computer Vision tasks, including medical image segmentation (MIS). Vision Mamba (VM)-based networks are particularly attractive due to their ability to achieve global receptive fields, similar to Vision Transformers, while also maintaining linear complexity in the number of tokens. However, the existing VM models still struggle to maintain both spatially local and global dependencies of tokens in high dimensional arrays due to their sequential nature. Employing multiple and/or complicated scanning strategies is computationally costly, which hinders applications of SSMs to high-dimensional 2D and 3D images that are common in MIS problems. In this work, we propose Local-Global Vision Mamba, LoG-VMamba, that explicitly enforces spatially adjacent tokens to remain nearby on the channel axis, and retains the global context in a compressed form. Our method allows the SSMs to access the local and global contexts even before reaching the last token while requiring only a simple scanning strategy. Our segmentation models are computationally efficient and substantially outperform both CNN and Transformers-based baselines on a diverse set of 2D and 3D MIS tasks. The implementation of LoG-VMamba is available at \url{https://github.com/Oulu-IMEDS/LoG-VMamba}.
Authors: Reuma Arav, Dennis Wittich, Franz Rottensteiner
Abstract: In recent years, three-dimensional point clouds are used increasingly to document natural environments. Each dataset contains a diverse set of objects, at varying shapes and sizes, distributed throughout the data and intricately intertwined with the topography. Therefore, regions of interest are difficult to find and consequent analyses become a challenge. Inspired from visual perception principles, we propose to differentiate objects of interest from the cluttered environment by evaluating how much they stand out from their surroundings, i.e., their geometric salience. Previous saliency detection approaches suggested mostly handcrafted attributes for the task. However, such methods fail when the data are too noisy or have high levels of texture. Here we propose a learning-based mechanism that accommodates noise and textured surfaces. We assume that within the natural environment any change from the prevalent surface would suggest a salient object. Thus, we first learn the underlying surface and then search for anomalies within it. Initially, a deep neural network is trained to reconstruct the surface. Regions where the reconstructed part deviates significantly from the original point cloud yield a substantial reconstruction error, signifying an anomaly, i.e., saliency. We demonstrate the effectiveness of the proposed approach by searching for salient features in various natural scenarios, which were acquired by different acquisition platforms. We show the strong correlation between the reconstruction error and salient objects.
Authors: Meng Zheng, Benjamin Planche, Zhongpai Gao, Terrence Chen, Richard J. Radke, Ziyan Wu
Abstract: Conventional 3D medical image segmentation methods typically require learning heavy 3D networks (e.g., 3D-UNet), as well as large amounts of in-domain data with accurate pixel/voxel-level labels to avoid overfitting. These solutions are thus extremely time- and labor-expensive, but also may easily fail to generalize to unseen objects during training. To alleviate this issue, we present MSFSeg, a novel few-shot 3D segmentation framework with a lightweight multi-surrogate fusion (MSF). MSFSeg is able to automatically segment unseen 3D objects/organs (during training) provided with one or a few annotated 2D slices or 3D sequence segments, via learning dense query-support organ/lesion anatomy correlations across patient populations. Our proposed MSF module mines comprehensive and diversified morphology correlations between unlabeled and the few labeled slices/sequences through multiple designated surrogates, making it able to generate accurate cross-domain 3D segmentation masks given annotated slices or sequences. We demonstrate the effectiveness of our proposed framework by showing superior performance on conventional few-shot segmentation benchmarks compared to prior art, and remarkable cross-domain cross-volume segmentation performance on proprietary 3D segmentation datasets for challenging entities, i.e., tubular structures, with only limited 2D or 3D labels.
Authors: Carina I. Hausladen, Manuel Knott, Colin F. Camerer, Pietro Perona
Abstract: We explore social perception of human faces in CLIP, a widely used open-source vision-language model. To this end, we compare the similarity in CLIP embeddings between different textual prompts and a set of face images. Our textual prompts are constructed from well-validated social psychology terms denoting social perception. The face images are synthetic and are systematically and independently varied along six dimensions: the legally protected attributes of age, gender, and race, as well as facial expression, lighting, and pose. Independently and systematically manipulating face attributes allows us to study the effect of each on social perception and avoids confounds that can occur in wild-collected data due to uncontrolled systematic correlations between attributes. Thus, our findings are experimental rather than observational. Our main findings are three. First, while CLIP is trained on the widest variety of images and texts, it is able to make fine-grained human-like social judgments on face images. Second, age, gender, and race do systematically impact CLIP's social perception of faces, suggesting an undesirable bias in CLIP vis-a-vis legally protected attributes. Most strikingly, we find a strong pattern of bias concerning the faces of Black women, where CLIP produces extreme values of social perception across different ages and facial expressions. Third, facial expression impacts social perception more than age and lighting as much as age. The last finding predicts that studies that do not control for unprotected visual attributes may reach the wrong conclusions on bias. Our novel method of investigation, which is founded on the social psychology literature and on the experiments involving the manipulation of individual attributes, yields sharper and more reliable observations than previous observational methods and may be applied to study biases in any vision-language model.
Authors: Mahrukh Awan, Asmar Nadeem, Muhammad Junaid Awan, Armin Mustafa, Syed Sameed Husain
Abstract: Exploiting both audio and visual modalities for video classification is a challenging task, as the existing methods require large model architectures, leading to high computational complexity and resource requirements. Smaller architectures, on the other hand, struggle to achieve optimal performance. In this paper, we propose Attend-Fusion, an audio-visual (AV) fusion approach that introduces a compact model architecture specifically designed to capture intricate audio-visual relationships in video data. Through extensive experiments on the challenging YouTube-8M dataset, we demonstrate that Attend-Fusion achieves an F1 score of 75.64\% with only 72M parameters, which is comparable to the performance of larger baseline models such as Fully-Connected Late Fusion (75.96\% F1 score, 341M parameters). Attend-Fusion achieves similar performance to the larger baseline model while reducing the model size by nearly 80\%, highlighting its efficiency in terms of model complexity. Our work demonstrates that the Attend-Fusion model effectively combines audio and visual information for video classification, achieving competitive performance with significantly reduced model size. This approach opens new possibilities for deploying high-performance video understanding systems in resource-constrained environments across various applications.
Authors: Axel Klawonn, Martin Lanser, Janine Weber
Abstract: Deep convolutional neural networks (CNNs) have been shown to be very successful in a wide range of image processing applications. However, due to their increasing number of model parameters and an increasing availability of large amounts of training data, parallelization strategies to efficiently train complex CNNs are necessary. In previous work by the authors, a novel model parallel CNN architecture was proposed which is loosely inspired by domain decomposition. In particular, the novel network architecture is based on a decomposition of the input data into smaller subimages. For each of these subimages, local CNNs with a proportionally smaller number of parameters are trained in parallel and the resulting local classifications are then aggregated in a second step by a dense feedforward neural network (DNN). In the present work, we compare the resulting CNN-DNN architecture to less costly alternatives to combine the local classifications into a final, global decision. Additionally, we investigate the performance of the CNN-DNN trained as one coherent model as well as using a transfer learning strategy, where the parameters of the pre-trained local CNNs are used as initial values for a subsequently trained global coherent CNN-DNN model.
Authors: Domen Tabernik, Jon Muhovi\v{c}, Matej Urbas, Danijel Sko\v{c}aj
Abstract: Object grasping is a fundamental challenge in robotics and computer vision, critical for advancing robotic manipulation capabilities. Deformable objects, like fabrics and cloths, pose additional challenges due to their non-rigid nature. In this work, we introduce CeDiRNet-3DoF, a deep-learning model for grasp point detection, with a particular focus on cloth objects. CeDiRNet-3DoF employs center direction regression alongside a localization network, attaining first place in the perception task of ICRA 2023's Cloth Manipulation Challenge. Recognizing the lack of standardized benchmarks in the literature that hinder effective method comparison, we present the ViCoS Towel Dataset. This extensive benchmark dataset comprises 8,000 real and 12,000 synthetic images, serving as a robust resource for training and evaluating contemporary data-driven deep-learning approaches. Extensive evaluation revealed CeDiRNet-3DoF's robustness in real-world performance, outperforming state-of-the-art methods, including the latest transformer-based models. Our work bridges a crucial gap, offering a robust solution and benchmark for cloth grasping in computer vision and robotics. Code and dataset are available at: https://github.com/vicoslab/CeDiRNet-3DoF
Authors: Domen Tabernik, Jon Muhovi\v{c}, Danijel Sko\v{c}aj
Abstract: Object counting and localization problems are commonly addressed with point supervised learning, which allows the use of less labor-intensive point annotations. However, learning based on point annotations poses challenges due to the high imbalance between the sets of annotated and unannotated pixels, which is often treated with Gaussian smoothing of point annotations and focal loss. However, these approaches still focus on the pixels in the immediate vicinity of the point annotations and exploit the rest of the data only indirectly. In this work, we propose a novel approach termed CeDiRNet for point-supervised learning that uses a dense regression of directions pointing towards the nearest object centers, i.e. center-directions. This provides greater support for each center point arising from many surrounding pixels pointing towards the object center. We propose a formulation of center-directions that allows the problem to be split into the domain-specific dense regression of center-directions and the final localization task based on a small, lightweight, and domain-agnostic localization network that can be trained with synthetic data completely independent of the target domain. We demonstrate the performance of the proposed method on six different datasets for object counting and localization, and show that it outperforms the existing state-of-the-art methods. The code is accessible on GitHub at https://github.com/vicoslab/CeDiRNet.git.
Authors: Qirui Chen, Shangzhe Di, Weidi Xie
Abstract: This paper considers the problem of Multi-Hop Video Question Answering (MH-VidQA) in long-form egocentric videos. This task not only requires to answer visual questions, but also to localize multiple relevant time intervals within the video as visual evidences. We develop an automated pipeline to create multi-hop question-answering pairs with associated temporal evidence, enabling to construct a large-scale dataset for instruction-tuning. To monitor the progress of this new task, we further curate a high-quality benchmark, MultiHop-EgoQA, with careful manual verification and refinement. Experimental results reveal that existing multi-modal systems exhibit inadequate multi-hop grounding and reasoning abilities, resulting in unsatisfactory performance. We then propose a novel architecture, termed as Grounding Scattered Evidence with Large Language Model (GeLM), that enhances multi-modal large language models (MLLMs) by incorporating a grounding module to retrieve temporal evidence from videos using flexible grounding tokens. Trained on our visual instruction data, GeLM demonstrates improved multi-hop grounding and reasoning capabilities, setting a new baseline for this challenging task. Furthermore, when trained on third-person view videos, the same architecture also achieves state-of-the-art performance on the single-hop VidQA benchmark, ActivityNet-RTL, demonstrating its effectiveness.
Authors: Karsten Roth, Vishaal Udandarao, Sebastian Dziadzio, Ameya Prabhu, Mehdi Cherti, Oriol Vinyals, Olivier H\'enaff, Samuel Albanie, Matthias Bethge, Zeynep Akata
Abstract: Multimodal foundation models serve numerous applications at the intersection of vision and language. Still, despite being pretrained on extensive data, they become outdated over time. To keep models updated, research into continual pretraining mainly explores scenarios with either (1) infrequent, indiscriminate updates on large-scale new data, or (2) frequent, sample-level updates. However, practical model deployment often operates in the gap between these two limit cases, as real-world applications often demand adaptation to specific subdomains, tasks or concepts -- spread over the entire, varying life cycle of a model. In this work, we complement current perspectives on continual pretraining through a research test bed as well as provide comprehensive guidance for effective continual model updates in such scenarios. We first introduce FoMo-in-Flux, a continual multimodal pretraining benchmark with realistic compute constraints and practical deployment requirements, constructed over 63 datasets with diverse visual and semantic coverage. Using FoMo-in-Flux, we explore the complex landscape of practical continual pretraining through multiple perspectives: (1) A data-centric investigation of data mixtures and stream orderings that emulate real-world deployment situations, (2) a method-centric investigation ranging from simple fine-tuning and traditional continual learning strategies to parameter-efficient updates and model merging, (3) meta learning rate schedules and mechanistic design choices, and (4) the influence of model and compute scaling. Together, our insights provide a practitioner's guide to continual multimodal pretraining for real-world deployment. Our benchmark and code is here: https://github.com/ExplainableML/fomo_in_flux.
Authors: Hetvi Waghela, Jaydip Sen, Sneha Rakshit
Abstract: Adversarial attacks, particularly the Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) pose significant threats to the robustness of deep learning models in image classification. This paper explores and refines defense mechanisms against these attacks to enhance the resilience of neural networks. We employ a combination of adversarial training and innovative preprocessing techniques, aiming to mitigate the impact of adversarial perturbations. Our methodology involves modifying input data before classification and investigating different model architectures and training strategies. Through rigorous evaluation of benchmark datasets, we demonstrate the effectiveness of our approach in defending against FGSM and PGD attacks. Our results show substantial improvements in model robustness compared to baseline methods, highlighting the potential of our defense strategies in real-world applications. This study contributes to the ongoing efforts to develop secure and reliable machine learning systems, offering practical insights and paving the way for future research in adversarial defense. By bridging theoretical advancements and practical implementation, we aim to enhance the trustworthiness of AI applications in safety-critical domains.
Authors: Chengyu Wu, Yatao Zhang, Yaqi Wang, Qifeng Wang, Shuai Wang
Abstract: Survival prediction for esophageal squamous cell cancer (ESCC) is crucial for doctors to assess a patient's condition and tailor treatment plans. The application and development of multi-modal deep learning in this field have attracted attention in recent years. However, the prognostically relevant features between cross-modalities have not been further explored in previous studies, which could hinder the performance of the model. Furthermore, the inherent semantic gap between different modal feature representations is also ignored. In this work, we propose a novel autoencoder-based deep learning model to predict the overall survival of the ESCC. Two novel modules were designed for multi-modal prognosis-related feature reinforcement and modeling ability enhancement. In addition, a novel joint loss was proposed to make the multi-modal feature representations more aligned. Comparison and ablation experiments demonstrated that our model can achieve satisfactory results in terms of discriminative ability, risk stratification, and the effectiveness of the proposed modules.
Authors: Xinmei Xu
Abstract: Pneumonia disease is one of the leading causes of death among children and adults worldwide. In the last ten years, computer-aided pneumonia detection methods have been developed to improve the efficiency and accuracy of the diagnosis process. Among those methods, the effects of deep learning approaches surpassed that of other traditional machine learning methods. This review paper searched and examined existing mainstream deep-learning approaches in the detection of pneumonia regions. This paper focuses on key aspects of the collected research, including their datasets, data processing techniques, general workflow, outcomes, advantages, and limitations. This paper also discusses current challenges in the field and proposes future work that can be done to enhance research procedures and the overall performance of deep learning models in detecting, classifying, and localizing infected regions. This review aims to offer an insightful summary and analysis of current research, facilitating the development of deep learning approaches in addressing treatable diseases.
Authors: Nicholas S. DiBrita, Daniel Leeds, Yuqian Huo, Jason Ludmir, Tirthak Patel
Abstract: Quantum computing has shown theoretical promise of speedup in several machine learning tasks, including generative tasks using generative adversarial networks (GANs). While quantum computers have been implemented with different types of technologies, recently, analog Rydberg atom quantum computers have been demonstrated to have desirable properties such as reconfigurable qubit (quantum bit) positions and multi-qubit operations. To leverage the properties of this technology, we propose ReCon, the first work to implement quantum GANs on analog Rydberg atom quantum computers. Our evaluation using simulations and real-computer executions shows 33% better quality (measured using Frechet Inception Distance (FID)) in generated images than the state-of-the-art technique implemented on superconducting-qubit technology.
Authors: Tomoumi Takase, Ryo Karakida
Abstract: While data augmentation (DA) is generally applied to input data, several studies have reported that applying DA to hidden layers in neural networks, i.e., feature augmentation, can improve performance. However, in previous studies, the layers to which DA is applied have not been carefully considered, often being applied randomly and uniformly or only to a specific layer, leaving room for arbitrariness. Thus, in this study, we investigated the trends of suitable layers for applying DA in various experimental configurations, e.g., training from scratch, transfer learning, various dataset settings, and different models. In addition, to adjust the suitable layers for DA automatically, we propose the adaptive layer selection (AdaLASE) method, which updates the ratio to perform DA for each layer based on the gradient descent method during training. The experimental results obtained on several image classification datasets indicate that the proposed AdaLASE method altered the ratio as expected and achieved high overall test accuracy.
Authors: Tianxiang Huang, Jing Shi, Ge Jin, Juncheng Li, Jun Wang, Jun Du, Jun Shi
Abstract: The B-mode ultrasound based computer-aided diagnosis (CAD) has demonstrated its effectiveness for diagnosis of Developmental Dysplasia of the Hip (DDH) in infants. However, due to effect of speckle noise in ultrasound im-ages, it is still a challenge task to accurately detect hip landmarks. In this work, we propose a novel hip landmark detection model by integrating the Topological GCN (TGCN) with an Improved Conformer (TGCN-ICF) into a unified frame-work to improve detection performance. The TGCN-ICF includes two subnet-works: an Improved Conformer (ICF) subnetwork to generate heatmaps and a TGCN subnetwork to additionally refine landmark detection. This TGCN can effectively improve detection accuracy with the guidance of class labels. Moreo-ver, a Mutual Modulation Fusion (MMF) module is developed for deeply ex-changing and fusing the features extracted from the U-Net and Transformer branches in ICF. The experimental results on the real DDH dataset demonstrate that the proposed TGCN-ICF outperforms all the compared algorithms.
Authors: Garrett Tanzer
Abstract: Sign language translation has historically been peripheral to mainstream machine translation research. In order to help converge the fields, we introduce FLEURS-ASL, an extension of the multiway parallel benchmarks FLORES (for text) and FLEURS (for speech) to support their first sign language (as video), American Sign Language, translated by 5 Certified Deaf Interpreters. FLEURS-ASL can be used to evaluate a variety of tasks -- primarily sentence- and discourse-level translation -- between ASL and 200 other languages as text, or 102 languages as speech. We provide baselines for tasks from ASL to English text using a unified modeling approach that incorporates timestamp tokens and previous text tokens in a 34-second context window, trained on random video clips from YouTube-ASL. This model meets or exceeds the performance of phrase-level baselines while supporting a multitude of new tasks. We also use FLEURS-ASL to show that multimodal frontier models have virtually no understanding of ASL, underscoring the importance of including sign languages in standard evaluation suites.
Authors: Yifei He, Yuzheng Hu, Yong Lin, Tong Zhang, Han Zhao
Abstract: Model merging offers an effective strategy to combine the strengths of multiple finetuned models into a unified model that preserves the specialized capabilities of each. Existing methods merge models in a global manner, performing arithmetic operations across all model parameters. However, such global merging often leads to task interference, degrading the performance of the merged model. In this work, we introduce Localize-and-Stitch, a novel approach that merges models in a localized way. Our algorithm works in two steps: i) Localization: identify tiny ($1\%$ of the total parameters) localized regions in the finetuned models containing essential skills for the downstream tasks, and ii) Stitching: reintegrate only these essential regions back into the pretrained model for task synergy. We demonstrate that our approach effectively locates sparse regions responsible for finetuned performance, and the localized regions could be treated as compact and interpretable representations of the finetuned models (tasks). Empirically, we evaluate our method on various vision and language benchmarks, showing that it outperforms existing model merging methods under different data availability scenarios. Beyond strong empirical performance, our algorithm also facilitates model compression and preserves pretrained knowledge, enabling flexible and continual skill composition from multiple finetuned models with minimal storage and computational overhead. Our code is available at https://github.com/yifei-he/Localize-and-Stitch.
Authors: Meiyi Wei, Liu Xie, Ying Sun, Gang Chen
Abstract: Recent advancements in local Implicit Neural Representation (INR) demonstrate its exceptional capability in handling images at various resolutions. However, frequency discrepancies between high-resolution (HR) and ground-truth images, especially at larger scales, result in significant artifacts and blurring in HR images. This paper introduces Frequency Consistency for Implicit Neural Representation (FreqINR), an innovative Arbitrary-scale Super-resolution method aimed at enhancing detailed textures by ensuring spectral consistency throughout both training and inference. During training, we employ Adaptive Discrete Cosine Transform Frequency Loss (ADFL) to minimize the frequency gap between HR and ground-truth images, utilizing 2-Dimensional DCT bases and focusing dynamically on challenging frequencies. During inference, we extend the receptive field to preserve spectral coherence between low-resolution (LR) and ground-truth images, which is crucial for the model to generate high-frequency details from LR counterparts. Experimental results show that FreqINR, as a lightweight approach, achieves state-of-the-art performance compared to existing Arbitrary-scale Super-resolution methods and offers notable improvements in computational efficiency. The code for our method will be made publicly available.
Authors: Mohammad Mohammadi, Sreejita Ghosh
Abstract: Classification of sets of inputs (e.g., images and texts) is an active area of research within both computer vision (CV) and natural language processing (NLP). A common way to represent a set of vectors is to model them as linear subspaces. In this contribution, we present a prototype-based approach for learning on the manifold formed from such linear subspaces, the Grassmann manifold. Our proposed method learns a set of subspace prototypes capturing the representative characteristics of classes and a set of relevance factors automating the selection of the dimensionality of the subspaces. This leads to a transparent classifier model which presents the computed impact of each input vector on its decision. Through experiments on benchmark image and text datasets, we have demonstrated the efficiency of our proposed classifier, compared to the transformer-based models in terms of not only performance and explainability but also computational resource requirements.
Authors: Zheyu Zhang, Xinzhao Liu, Zheng Chen, Yueyi Zhang, Huanjing Yue, Yunwei Ou, Xiaoyan Sun
Abstract: Multi-modal Magnetic Resonance Imaging (MRI) is imperative for accurate brain tumor segmentation, offering indispensable complementary information. Nonetheless, the absence of modalities poses significant challenges in achieving precise segmentation. Recognizing the shared anatomical structures between mono-modal and multi-modal representations, it is noteworthy that mono-modal images typically exhibit limited features in specific regions and tissues. In response to this, we present Anatomical Consistency Distillation and Inconsistency Synthesis (ACDIS), a novel framework designed to transfer anatomical structures from multi-modal to mono-modal representations and synthesize modality-specific features. ACDIS consists of two main components: Anatomical Consistency Distillation (ACD) and Modality Feature Synthesis Block (MFSB). ACD incorporates the Anatomical Feature Enhancement Block (AFEB), meticulously mining anatomical information. Simultaneously, Anatomical Consistency ConsTraints (ACCT) are employed to facilitate the consistent knowledge transfer, i.e., the richness of information and the similarity in anatomical structure, ensuring precise alignment of structural features across mono-modality and multi-modality. Complementarily, MFSB produces modality-specific features to rectify anatomical inconsistencies, thereby compensating for missing information in the segmented features. Through validation on the BraTS2018 and BraTS2020 datasets, ACDIS substantiates its efficacy in the segmentation of brain tumors with missing MRI modalities.
Authors: Ruiqing Sun, Delong Yang, Yiyan Su, Shaohui Zhang, Qun Hao
Abstract: Fourier Ptychographic Microscopy (FPM) is a computational imaging technique that enables high-resolution imaging over a large field of view. However, its application in the biomedical field has been limited due to the long image reconstruction time and poor noise robustness. In this paper, we propose a fast and robust FPM reconstruction method based on physical neural networks with batch update stochastic gradient descent (SGD) optimization strategy, capable of achieving attractive results with low single-to-noise ratio and correcting multiple system parameters simultaneously. Our method leverages a random batch optimization approach, breaks away from the fixed sequential iterative order and gives greater attention to high-frequency information. The proposed method has better convergence performance even for low signal-to-noise ratio data sets, such as low exposure time dark-field images. As a result, it can greatly increase the image recording and result reconstruction speed without any additional hardware modifications. By utilizing advanced deep learning optimizers and perform parallel computational scheme, our method enhances GPU computational efficiency, significantly reducing reconstruction costs. Experimental results demonstrate that our method achieves near real-time digital refocusing of a 1024 x 1024 pixels region of interest on consumer-grade GPUs. This approach significantly improves temporal resolution (by reducing the exposure time of dark-field images), noise resistance, and reconstruction speed, and therefore can efficiently promote the practical application of FPM in clinical diagnostics, digital pathology, and biomedical research, etc. In addition, we believe our algorithm scheme can help researchers quickly validate and implement FPM-related ideas. We invite requests for the full code via email.
Authors: Yujia Lin, Aiwei Lian, Minyu Liao, Yipeng Liu
Abstract: Previous research has established that breast cancer is a prevalent cancer type, with Invasive Ductal Carcinoma (IDC) being the most common subtype. The incidence of this dangerous cancer continues to rise, making accurate and rapid diagnosis, particularly in the early stages, critically important. While modern Computer-Aided Diagnosis (CAD) systems can address most cases, medical professionals still face challenges in using them in the field without powerful computing resources. In this paper, we propose a novel CNN model called BCDNet, which effectively detects IDC in histopathological images with an accuracy of up to 89.5\% and reduces training time effectively.
Authors: Ardhendu Sekhar, Vrinda Goel, Garima Jain, Abhijeet Patil, Ravi Kant Gupta, Amit Sethi
Abstract: The current standard for detecting human epidermal growth factor receptor 2 (HER2) status in breast cancer patients relies on HER2 amplification, identified through fluorescence in situ hybridization (FISH) or immunohistochemistry (IHC). However, hematoxylin and eosin (H\&E) tumor stains are more widely available, and accurately predicting HER2 status using H\&E could reduce costs and expedite treatment selection. Deep Learning algorithms for H&E have shown effectiveness in predicting various cancer features and clinical outcomes, including moderate success in HER2 status prediction. In this work, we employed a customized weak supervision classification technique combined with MoCo-v2 contrastive learning to predict HER2 status. We trained our pipeline on 182 publicly available H&E Whole Slide Images (WSIs) from The Cancer Genome Atlas (TCGA), for which annotations by the pathology team at Yale School of Medicine are publicly available. Our pipeline achieved an Area Under the Curve (AUC) of 0.85 across four different test folds. Additionally, we tested our model on 44 H&E slides from the TCGA-BRCA dataset, which had an HER2 score of 2+ and included corresponding HER2 status and FISH test results. These cases are considered equivocal for IHC, requiring an expensive FISH test on their IHC slides for disambiguation. Our pipeline demonstrated an AUC of 0.81 on these challenging H&E slides. Reducing the need for FISH test can have significant implications in cancer treatment equity for underserved populations.
Authors: Ran An, Yinghui Zhang, Xi Chen, Lemeng Li, Ke Chen, Hongwei Li
Abstract: Low-dose computed tomography (LDCT) offers significant advantages in reducing the potential harm to human bodies. However, reducing the X-ray dose in CT scanning often leads to severe noise and artifacts in the reconstructed images, which might adversely affect diagnosis. By utilizing the expectation maximization (EM) algorithm, statistical priors could be combined with artificial priors to improve LDCT reconstruction quality. However, conventional EM-based regularization methods adopt an alternating solving strategy, i.e. full reconstruction followed by image-regularization, resulting in over-smoothing and slow convergence. In this paper, we propose to integrate TV regularization into the ``M''-step of the EM algorithm, thus achieving effective and efficient regularization. Besides, by employing the Chambolle-Pock (CP) algorithm and the ordered subset (OS) strategy, we propose the OSEM-CP algorithm for LDCT reconstruction, in which both reconstruction and regularization are conducted view-by-view. Furthermore, by unrolling OSEM-CP, we propose an end-to-end reconstruction neural network (NN), named OSEM-CPNN, with remarkable performance and efficiency that achieves high-quality reconstructions in just one full-view iteration. Experiments on different models and datasets demonstrate our methods' outstanding performance compared to traditional and state-of-the-art deep-learning methods.
Authors: Suyash Vardhan Mathur, Jainit Sushil Bafna, Kunal Kartik, Harshita Khandelwal, Manish Shrivastava, Vivek Gupta, Mohit Bansal, Dan Roth
Abstract: Existing datasets for tabular question answering typically focus exclusively on text within cells. However, real-world data is inherently multimodal, often blending images such as symbols, faces, icons, patterns, and charts with textual content in tables. With the evolution of AI models capable of multimodal reasoning, it is pertinent to assess their efficacy in handling such structured data. This study investigates whether current AI models can perform knowledge-aware reasoning on multimodal structured data. We explore their ability to reason on tables that integrate both images and text, introducing MMTabQA, a new dataset designed for this purpose. Our experiments highlight substantial challenges for current AI models in effectively integrating and interpreting multiple text and image inputs, understanding visual context, and comparing visual content across images. These findings establish our dataset as a robust benchmark for advancing AI's comprehension and capabilities in analyzing multimodal structured data.
Authors: Lei Li, Hannah Smith, Yilin Lyu, Julia Camps, Blanca Rodriguez, Abhirup Banerjee, Vicente Grau
Abstract: Cardiac digital twins (CDTs) offer personalized \textit{in-silico} cardiac representations for the inference of multi-scale properties tied to cardiac mechanisms. The creation of CDTs requires precise information about the electrode position on the torso, especially for the personalized electrocardiogram (ECG) calibration. However, current studies commonly rely on additional acquisition of torso imaging and manual/semi-automatic methods for ECG electrode localization. In this study, we propose a novel and efficient topology-informed model to fully automatically extract personalized ECG electrode locations from 2D clinically standard cardiac MRIs. Specifically, we obtain the sparse torso contours from the cardiac MRIs and then localize the electrodes from the contours. Cardiac MRIs aim at imaging of the heart instead of the torso, leading to incomplete torso geometry within the imaging. To tackle the missing topology, we incorporate the electrodes as a subset of the keypoints, which can be explicitly aligned with the 3D torso topology. The experimental results demonstrate that the proposed model outperforms the time-consuming conventional method in terms of accuracy (Euclidean distance: $1.24 \pm 0.293$ cm vs. $1.48 \pm 0.362$ cm) and efficiency ($2$~s vs. $30$-$35$~min). We further demonstrate the effectiveness of using the detected electrodes for \textit{in-silico} ECG simulation, highlighting their potential for creating accurate and efficient CDT models. The code will be released publicly after the manuscript is accepted for publication.
Authors: Qiuli Wang, Yongxu Liu, Li Ma, Xianqi Wang, Wei Chen, Xiaohong Yao
Abstract: Histological Tertiary Lymphoid Structures (TLSs) are increasingly recognized for their correlation with the efficacy of immunotherapy in various solid tumors. Traditionally, the identification and characterization of TLSs rely on immunohistochemistry (IHC) staining techniques, utilizing markers such as CD20 for B cells. Despite the specificity of IHC, Hematoxylin-Eosin (H&E) staining offers a more accessible and cost-effective choice. Capitalizing on the prevalence of H&E staining slides, we introduce a novel Mask-Guided Adversarial Transfer Learning method designed for virtual pathological staining. This method adeptly captures the nuanced color variations across diverse tissue types under various staining conditions, such as nucleus, red blood cells, positive reaction regions, without explicit label information, and adeptly synthesizes realistic IHC-like virtual staining patches, even replicating the positive reaction. Further, we propose the Virtual IHC Pathology Analysis Network (VIPA-Net), an integrated framework encompassing a Mask-Guided Transfer Module and an H&E-Based Virtual Staining TLS Detection Module. VIPA-Net synergistically harnesses both H\&E staining slides and the synthesized virtual IHC patches to enhance the detection of TLSs within H&E Whole Slide Images (WSIs). We evaluate the network with a comprehensive dataset comprising 1019 annotated slides from The Cancer Genome Atlas (TCGA). Experimental results compellingly illustrate that the VIPA-Net substantially elevates TLS detection accuracy, effectively circumventing the need for actual CD20 staining across the public dataset.
Authors: Chunran Zheng, Wei Xu, Zuhao Zou, Tong Hua, Chongjian Yuan, Dongjiao He, Bingyang Zhou, Zheng Liu, Jiarong Lin, Fangcheng Zhu, Yunfan Ren, Rong Wang, Fanle Meng, Fu Zhang
Abstract: This paper proposes FAST-LIVO2: a fast, direct LiDAR-inertial-visual odometry framework to achieve accurate and robust state estimation in SLAM tasks and provide great potential in real-time, onboard robotic applications. FAST-LIVO2 fuses the IMU, LiDAR and image measurements efficiently through an ESIKF. To address the dimension mismatch between the heterogeneous LiDAR and image measurements, we use a sequential update strategy in the Kalman filter. To enhance the efficiency, we use direct methods for both the visual and LiDAR fusion, where the LiDAR module registers raw points without extracting edge or plane features and the visual module minimizes direct photometric errors without extracting ORB or FAST corner features. The fusion of both visual and LiDAR measurements is based on a single unified voxel map where the LiDAR module constructs the geometric structure for registering new LiDAR scans and the visual module attaches image patches to the LiDAR points. To enhance the accuracy of image alignment, we use plane priors from the LiDAR points in the voxel map (and even refine the plane prior) and update the reference patch dynamically after new images are aligned. Furthermore, to enhance the robustness of image alignment, FAST-LIVO2 employs an on-demanding raycast operation and estimates the image exposure time in real time. Lastly, we detail three applications of FAST-LIVO2: UAV onboard navigation demonstrating the system's computation efficiency for real-time onboard navigation, airborne mapping showcasing the system's mapping accuracy, and 3D model rendering (mesh-based and NeRF-based) underscoring the suitability of our reconstructed dense map for subsequent rendering tasks. We open source our code, dataset and application on GitHub to benefit the robotics community.
Authors: Bharath Rajiv Nair
Abstract: This paper explores the paradigm of Collaborative Perception (CP), where multiple robots and sensors in the environment share and integrate sensor data to construct a comprehensive representation of the surroundings. By aggregating data from various sensors and utilizing advanced algorithms, the collaborative perception framework improves task efficiency, coverage, and safety. Two case studies are presented to showcase the benefits of collaborative perception in multi-robot systems. The first case study illustrates the benefits and advantages of using CP for the task of household cleaning with a team of cleaning robots. The second case study performs a comparative analysis of the performance of CP versus Standalone Perception (SP) for Autonomous Mobile Robots operating in a warehouse environment. The case studies validate the effectiveness of CP in enhancing multi-robot coordination, task completion, and overall system performance and its potential to impact operations in other applications as well. Future investigations will focus on optimizing the framework and validating its performance through empirical testing.
Authors: Md Awsafur Rahman, Zaber Ibn Abdul Hakim, Najibul Haque Sarker, Bishmoy Paul, Shaikh Anowarul Fattah
Abstract: The recent surge in AI-generated songs presents exciting possibilities and challenges. While these tools democratize music creation, they also necessitate the ability to distinguish between human-composed and AI-generated songs for safeguarding artistic integrity and content curation. Existing research and datasets in fake song detection only focus on singing voice deepfake detection (SVDD), where the vocals are AI-generated but the instrumental music is sourced from real songs. However, this approach is inadequate for contemporary end-to-end AI-generated songs where all components (vocals, lyrics, music, and style) could be AI-generated. Additionally, existing datasets lack lyrics-music diversity, long-duration songs, and open fake songs. To address these gaps, we introduce SONICS, a novel dataset for end-to-end Synthetic Song Detection (SSD), comprising over 97k songs with over 49k synthetic songs from popular platforms like Suno and Udio. Furthermore, we highlight the importance of modeling long-range temporal dependencies in songs for effective authenticity detection, an aspect overlooked in existing methods. To capture these patterns, we propose a novel model, SpecTTTra, that is up to 3 times faster and 6 times more memory efficient compared to popular CNN and Transformer-based models while maintaining competitive performance. Finally, we offer both AI-based and Human evaluation benchmarks, addressing another deficiency in current research.
Authors: Chang Gao, Jianfei Chen, Kang Zhao, Jiaqi Wang, Liping Jing
Abstract: Fully quantized training (FQT) accelerates the training of deep neural networks by quantizing the activations, weights, and gradients into lower precision. To explore the ultimate limit of FQT (the lowest achievable precision), we make a first attempt to 1-bit FQT. We provide a theoretical analysis of FQT based on Adam and SGD, revealing that the gradient variance influences the convergence of FQT. Building on these theoretical results, we introduce an Activation Gradient Pruning (AGP) strategy. The strategy leverages the heterogeneity of gradients by pruning less informative gradients and enhancing the numerical precision of remaining gradients to mitigate gradient variance. Additionally, we propose Sample Channel joint Quantization (SCQ), which utilizes different quantization strategies in the computation of weight gradients and activation gradients to ensure that the method is friendly to low-bitwidth hardware. Finally, we present a framework to deploy our algorithm. For fine-tuning VGGNet-16 and ResNet-18 on multiple datasets, our algorithm achieves an average accuracy improvement of approximately 6%, compared to per-sample quantization. Moreover, our training speedup can reach a maximum of 5.13x compared to full precision training.
Authors: Langrui Zhou, Guang Li
Abstract: The current mainstream multi-modal medical image-to-image translation methods face a contradiction. Supervised methods with outstanding performance rely on pixel-wise aligned training data to constrain the model optimization. However, obtaining pixel-wise aligned multi-modal medical image datasets is challenging. Unsupervised methods can be trained without paired data, but their reliability cannot be guaranteed. At present, there is no ideal multi-modal medical image-to-image translation method that can generate reliable translation results without the need for pixel-wise aligned data. This work aims to develop a novel medical image-to-image translation model that is independent of pixel-wise aligned data (MITIA), enabling reliable multi-modal medical image-to-image translation under the condition of misaligned training data. The proposed MITIA model utilizes a prior extraction network composed of a multi-modal medical image registration module and a multi-modal misalignment error detection module to extract pixel-level prior information from training data with misalignment errors to the largest extent. The extracted prior information is then used to construct a regularization term to constrain the optimization of the unsupervised cycle-consistent GAN model, restricting its solution space and thereby improving the performance and reliability of the generator. We trained the MITIA model using six datasets containing different misalignment errors and two well-aligned datasets. Subsequently, we compared the proposed method with six other state-of-the-art image-to-image translation methods. The results of both quantitative analysis and qualitative visual inspection indicate that MITIA achieves superior performance compared to the competing state-of-the-art methods, both on misaligned data and aligned data.
Authors: Michael Kirchhof
Abstract: Uncertainty quantification is a key pillar of trustworthy machine learning. It enables safe reactions under unsafe inputs, like predicting only when the machine learning model detects sufficient evidence, discarding anomalous data, or emitting warnings when an error is likely to be inbound. This is particularly crucial in safety-critical areas like medical image classification or self-driving cars. Despite the plethora of proposed uncertainty quantification methods achieving increasingly higher scores on performance benchmarks, uncertainty estimates are often shied away from in practice. Many machine learning projects start from pretrained latent representations that come without uncertainty estimates. Uncertainties would need to be trained by practitioners on their own, which is notoriously difficult and resource-intense. This thesis makes uncertainty estimates easily accessible by adding them to the latent representation vectors of pretrained computer vision models. Besides proposing approaches rooted in probability and decision theory, such as Monte-Carlo InfoNCE (MCInfoNCE) and loss prediction, we delve into both theoretical and empirical questions. We show that these unobservable uncertainties about unobservable latent representations are indeed provably correct. We also provide an uncertainty-aware representation learning (URL) benchmark to compare these unobservables against observable ground-truths. Finally, we compile our findings to pretrain lightweight representation uncertainties on large-scale computer vision models that transfer to unseen datasets in a zero-shot manner. Our findings do not only advance the current theoretical understanding of uncertainties over latent variables, but also facilitate the access to uncertainty quantification for future researchers inside and outside the field, enabling straightforward but trustworthy machine learning.
Authors: Monica Millunzi, Lorenzo Bonicelli, Angelo Porrello, Jacopo Credi, Petter N. Kolm, Simone Calderara
Abstract: Forgetting presents a significant challenge during incremental training, making it particularly demanding for contemporary AI systems to assimilate new knowledge in streaming data environments. To address this issue, most approaches in Continual Learning (CL) rely on the replay of a restricted buffer of past data. However, the presence of noise in real-world scenarios, where human annotation is constrained by time limitations or where data is automatically gathered from the web, frequently renders these strategies vulnerable. In this study, we address the problem of CL under Noisy Labels (CLN) by introducing Alternate Experience Replay (AER), which takes advantage of forgetting to maintain a clear distinction between clean, complex, and noisy samples in the memory buffer. The idea is that complex or mislabeled examples, which hardly fit the previously learned data distribution, are most likely to be forgotten. To grasp the benefits of such a separation, we equip AER with Asymmetric Balanced Sampling (ABS): a new sample selection strategy that prioritizes purity on the current task while retaining relevant samples from the past. Through extensive computational comparisons, we demonstrate the effectiveness of our approach in terms of both accuracy and purity of the obtained buffer, resulting in a remarkable average gain of 4.71% points in accuracy with respect to existing loss-based purification strategies. Code is available at https://github.com/aimagelab/mammoth.
Authors: Hai Nguyen, Andrea Baisero, David Klee, Dian Wang, Robert Platt, Christopher Amato
Abstract: Incorporating inductive biases is a promising approach for tackling challenging robot learning domains with sample-efficient solutions. This paper identifies partially observable domains where symmetries can be a useful inductive bias for efficient learning. Specifically, by encoding the equivariance regarding specific group symmetries into the neural networks, our actor-critic reinforcement learning agents can reuse solutions in the past for related scenarios. Consequently, our equivariant agents outperform non-equivariant approaches significantly in terms of sample efficiency and final performance, demonstrated through experiments on a range of robotic tasks in simulation and real hardware.
Authors: Dongyue Li, Kailai Chen, Predrag Radivojac, Hongyang R. Zhang
Abstract: Data augmentation is widely used for training a neural network given little labeled data. A common practice of augmentation training is applying a composition of multiple transformations sequentially to the data. Existing augmentation methods such as RandAugment randomly sample from a list of pre-selected transformations, while methods such as AutoAugment apply advanced search to optimize over an augmentation set of size $k^d$, which is the number of transformation sequences of length $d$, given a list of $k$ transformations. In this paper, we design efficient algorithms whose running time complexity is much faster than the worst-case complexity of $O(k^d)$, provably. We propose a new algorithm to search for a binary tree-structured composition of $k$ transformations, where each tree node corresponds to one transformation. The binary tree generalizes sequential augmentations, such as the SimCLR augmentation scheme for contrastive learning. Using a top-down, recursive search procedure, our algorithm achieves a runtime complexity of $O(2^d k)$, which is much faster than $O(k^d)$ as $k$ increases above $2$. We apply our algorithm to tackle data distributions with heterogeneous subpopulations by searching for one tree in each subpopulation and then learning a weighted combination, resulting in a forest of trees. We validate our proposed algorithms on numerous graph and image datasets, including a multi-label graph classification dataset we collected. The dataset exhibits significant variations in the sizes of graphs and their average degrees, making it ideal for studying data augmentation. We show that our approach can reduce the computation cost by 43% over existing search methods while improving performance by 4.3%. The tree structures can be used to interpret the relative importance of each transformation, such as identifying the important transformations on small vs. large graphs.
Authors: Xiaoman Zhang, Juli\'an N. Acosta, Hong-Yu Zhou, Pranav Rajpurkar
Abstract: Recent advancements in artificial intelligence have significantly improved the automatic generation of radiology reports. However, existing evaluation methods fail to reveal the models' understanding of radiological images and their capacity to achieve human-level granularity in descriptions. To bridge this gap, we introduce a system, named ReXKG, which extracts structured information from processed reports to construct a comprehensive radiology knowledge graph. We then propose three metrics to evaluate the similarity of nodes (ReXKG-NSC), distribution of edges (ReXKG-AMS), and coverage of subgraphs (ReXKG-SCS) across various knowledge graphs. We conduct an in-depth comparative analysis of AI-generated and human-written radiology reports, assessing the performance of both specialist and generalist models. Our study provides a deeper understanding of the capabilities and limitations of current AI models in radiology report generation, offering valuable insights for improving model performance and clinical applicability.
Authors: Shubham Bharti, Shiyun Cheng, Jihyun Rho, Martina Rao, Xiaojin Zhu
Abstract: We introduce CHARTOM, a visual theory-of-mind benchmark for multimodal large language models. CHARTOM consists of specially designed data visualizing charts. Given a chart, a language model needs to not only correctly comprehend the chart (the FACT question) but also judge if the chart will be misleading to a human reader (the MIND question). Both questions have significant societal benefits. We detail the construction of the CHARTOM benchmark including its calibration on human performance.
Authors: Illia Oleksiienko, Paraskevi Nousi, Nikolaos Passalis, Anastasios Tefas, Alexandros Iosifidis
Abstract: In this paper, we propose a novel voxel-based 3D single object tracking (3D SOT) method called Voxel Pseudo Image Tracking (VPIT). VPIT is the first method that uses voxel pseudo images for 3D SOT. The input point cloud is structured by pillar-based voxelization, and the resulting pseudo image is used as an input to a 2D-like Siamese SOT method. The pseudo image is created in the Bird's-eye View (BEV) coordinates, and therefore the objects in it have constant size. Thus, only the object rotation can change in the new coordinate system and not the object scale. For this reason, we replace multi-scale search with a multi-rotation search, where differently rotated search regions are compared against a single target representation to predict both position and rotation of the object. Experiments on KITTI Tracking dataset show that VPIT is the fastest 3D SOT method and maintains competitive Success and Precision values. Application of a SOT method in a real-world scenario meets with limitations such as lower computational capabilities of embedded devices and a latency-unforgiving environment, where the method is forced to skip certain data frames if the inference speed is not high enough. We implement a real-time evaluation protocol and show that other methods lose most of their performance on embedded devices, while VPIT maintains its ability to track the object.
Authors: Matthias Rottmann, Marco Reese
Abstract: In this work, we for the first time present a method for detecting label errors in image datasets with semantic segmentation, i.e., pixel-wise class labels. Annotation acquisition for semantic segmentation datasets is time-consuming and requires plenty of human labor. In particular, review processes are time consuming and label errors can easily be overlooked by humans. The consequences are biased benchmarks and in extreme cases also performance degradation of deep neural networks (DNNs) trained on such datasets. DNNs for semantic segmentation yield pixel-wise predictions, which makes detection of label errors via uncertainty quantification a complex task. Uncertainty is particularly pronounced at the transitions between connected components of the prediction. By lifting the consideration of uncertainty to the level of predicted components, we enable the usage of DNNs together with component-level uncertainty quantification for the detection of label errors. We present a principled approach to benchmarking the task of label error detection by dropping labels from the Cityscapes dataset as well from a dataset extracted from the CARLA driving simulator, where in the latter case we have the labels under control. Our experiments show that our approach is able to detect the vast majority of label errors while controlling the number of false label error detections. Furthermore, we apply our method to semantic segmentation datasets frequently used by the computer vision community and present a collection of label errors along with sample statistics.
Authors: Shizheng Zhou, Juntao Jiang, Xiaohan Hong, Yan Hong, Pengcheng Fu
Abstract: Marine microalgae are widespread in the ocean and play a crucial role in the ecosystem. Automatic identification and location of marine microalgae in microscopy images would help establish marine ecological environment monitoring and water quality evaluation system. We proposed a new dataset for the detection of marine microalgae and a range of detection methods, the dataset including images of different genus of algae and the same genus in different states. We set the number of unbalanced classes in the data set and added images of mixed water samples in the test set to simulate the actual situation in the field. Then we trained, validated and tested the, TOOD, YOLOv5, YOLOv8 and variants of RCNN algorithms on this dataset. The results showed both one-stage and two-stage object detection models can achieve high mean average precision, which proves the ability of computer vision in multi-object detection of microalgae, and provides basic data and models for real-time detection of microalgal cells.
Authors: Saed Moradi, Alireza Memarmoghadam, Payman Moallem, Mohamad Farzan Sabahi
Abstract: Infrared small target detection (IRSTD) poses a significant challenge in the field of computer vision. While substantial efforts have been made over the past two decades to improve the detection capabilities of IRSTD algorithms, there has been a lack of extensive investigation into the evaluation metrics used for assessing their performance. In this paper, we employ a systematic approach to address this issue by first evaluating the effectiveness of existing metrics and then proposing new metrics to overcome the limitations of conventional ones. To achieve this, we carefully analyze the necessary conditions for successful detection and identify the shortcomings of current evaluation metrics, including both pre-thresholding and post-thresholding metrics. We then introduce new metrics that are designed to align with the requirements of real-world systems. Furthermore, we utilize these newly proposed metrics to compare and evaluate the performance of five widely recognized small infrared target detection algorithms. The results demonstrate that the new metrics provide consistent and meaningful quantitative assessments, aligning with qualitative observations.
Authors: Chengliang Liu, Jie Wen, Yong Xu, Bob Zhang, Liqiang Nie, Min Zhang
Abstract: As a cross-topic of multi-view learning and multi-label classification, multi-view multi-label classification has gradually gained traction in recent years. The application of multi-view contrastive learning has further facilitated this process, however, the existing multi-view contrastive learning methods crudely separate the so-called negative pair, which largely results in the separation of samples belonging to the same category or similar ones. Besides, plenty of multi-view multi-label learning methods ignore the possible absence of views and labels. To address these issues, in this paper, we propose an incomplete multi-view partial multi-label classification network named RANK. In this network, a label-driven multi-view contrastive learning strategy is proposed to leverage supervised information to preserve the structure within view and perform consistent alignment across views. Furthermore, we break through the view-level weights inherent in existing methods and propose a quality-aware sub-network to dynamically assign quality scores to each view of each sample. The label correlation information is fully utilized in the final multi-label cross-entropy classification loss, effectively improving the discriminative power. Last but not least, our model is not only able to handle complete multi-view multi-label datasets, but also works on datasets with missing instances and labels. Extensive experiments confirm that our RANK outperforms existing state-of-the-art methods.
Authors: Jielu Zhang, Zhongliang Zhou, Gengchen Mai, Mengxuan Hu, Zihan Guan, Sheng Li, Lan Mu
Abstract: Remote sensing imagery has attracted significant attention in recent years due to its instrumental role in global environmental monitoring, land usage monitoring, and more. As image databases grow each year, performing automatic segmentation with deep learning models has gradually become the standard approach for processing the data. Despite the improved performance of current models, certain limitations remain unresolved. Firstly, training deep learning models for segmentation requires per-pixel annotations. Given the large size of datasets, only a small portion is fully annotated and ready for training. Additionally, the high intra-dataset variance in remote sensing data limits the transfer learning ability of such models. Although recently proposed generic segmentation models like SAM have shown promising results in zero-shot instance-level segmentation, adapting them to semantic segmentation is a non-trivial task. To tackle these challenges, we propose a novel method named Text2Seg for remote sensing semantic segmentation. Text2Seg overcomes the dependency on extensive annotations by employing an automatic prompt generation process using different visual foundation models (VFMs), which are trained to understand semantic information in various ways. This approach not only reduces the need for fully annotated datasets but also enhances the model's ability to generalize across diverse datasets. Evaluations on four widely adopted remote sensing datasets demonstrate that Text2Seg significantly improves zero-shot prediction performance compared to the vanilla SAM model, with relative improvements ranging from 31% to 225%. Our code is available at https://github.com/Douglas2Code/Text2Seg.
Authors: Zhu Liu, Jinyuan Liu, Guanyao Wu, Zihang Chen, Xin Fan, Risheng Liu
Abstract: In recent years, learning-based methods have achieved significant advancements in multi-exposure image fusion. However, two major stumbling blocks hinder the development, including pixel misalignment and inefficient inference. Reliance on aligned image pairs in existing methods causes susceptibility to artifacts due to device motion. Additionally, existing techniques often rely on handcrafted architectures with huge network engineering, resulting in redundant parameters, adversely impacting inference efficiency and flexibility. To mitigate these limitations, this study introduces an architecture search-based paradigm incorporating self-alignment and detail repletion modules for robust multi-exposure image fusion. Specifically, targeting the extreme discrepancy of exposure, we propose the self-alignment module, leveraging scene relighting to constrain the illumination degree for following alignment and feature extraction. Detail repletion is proposed to enhance the texture details of scenes. Additionally, incorporating a hardware-sensitive constraint, we present the fusion-oriented architecture search to explore compact and efficient networks for fusion. The proposed method outperforms various competitive schemes, achieving a noteworthy 3.19\% improvement in PSNR for general scenarios and an impressive 23.5\% enhancement in misaligned scenarios. Moreover, it significantly reduces inference time by 69.1\%. The code will be available at https://github.com/LiuZhu-CV/CRMEF.
Authors: Thanh-Dat Truong, Khoa Luu
Abstract: Understanding action recognition in egocentric videos has emerged as a vital research topic with numerous practical applications. With the limitation in the scale of egocentric data collection, learning robust deep learning-based action recognition models remains difficult. Transferring knowledge learned from the large-scale exocentric data to the egocentric data is challenging due to the difference in videos across views. Our work introduces a novel cross-view learning approach to action recognition (CVAR) that effectively transfers knowledge from the exocentric to the selfish view. First, we present a novel geometric-based constraint into the self-attention mechanism in Transformer based on analyzing the camera positions between two views. Then, we propose a new cross-view self-attention loss learned on unpaired cross-view data to enforce the self-attention mechanism learning to transfer knowledge across views. Finally, to further improve the performance of our cross-view learning approach, we present the metrics to measure the correlations in videos and attention maps effectively. Experimental results on standard egocentric action recognition benchmarks, i.e., Charades-Ego, EPIC-Kitchens-55, and EPIC-Kitchens-100, have shown our approach's effectiveness and state-of-the-art performance.
Authors: Zhi-Hao Lin, Bohan Liu, Yi-Ting Chen, Kuan-Sheng Chen, David Forsyth, Jia-Bin Huang, Anand Bhattad, Shenlong Wang
Abstract: We present UrbanIR (Urban Scene Inverse Rendering), a new inverse graphics model that enables realistic, free-viewpoint renderings of scenes under various lighting conditions with a single video. It accurately infers shape, albedo, visibility, and sun and sky illumination from wide-baseline videos, such as those from car-mounted cameras, differing from NeRF's dense view settings. In this context, standard methods often yield subpar geometry and material estimates, such as inaccurate roof representations and numerous 'floaters'. UrbanIR addresses these issues with novel losses that reduce errors in inverse graphics inference and rendering artifacts. Its techniques allow for precise shadow volume estimation in the original scene. The model's outputs support controllable editing, enabling photorealistic free-viewpoint renderings of night simulations, relit scenes, and inserted objects, marking a significant improvement over existing state-of-the-art methods.
Authors: Mohamed Elghandouri, Khalil Ezzinbi, Mouad Klai, Olivier Monga
Abstract: Partial Differential Equations (PDEs) play a crucial role as tools for modeling and comprehending intricate natural processes, notably within the domain of biology. This research explores the domain of microbial activity within the complex matrix of 3D soil structures, providing valuable understanding into both the existence and uniqueness of solutions and the asymptotic behavior of the corresponding PDE model. Our investigation results in the discovery of a global attractor, a fundamental feature with significant implications for long-term system behavior. To enhance the clarity of our findings, numerical simulations are employed to visually illustrate the attributes of this global attractor.
Authors: Taekyung Kim, Sanghyuk Chun, Byeongho Heo, Dongyoon Han
Abstract: Masked image modeling (MIM) has become a leading self-supervised learning strategy. MIMs such as Masked Autoencoder (MAE) learn strong representations by randomly masking input tokens for the encoder to process, with the decoder reconstructing the masked tokens to the input. However, MIM pre-trained encoders often exhibit a limited attention span, attributed to MIM's sole focus on regressing masked tokens only, which may impede the encoder's broader context learning. To tackle the limitation, we improve MIM by explicitly incorporating unmasked tokens into the training process. Specifically, our method enables the encoder to learn from broader context supervision, allowing unmasked tokens to experience broader contexts while the decoder reconstructs masked tokens. Thus, the encoded unmasked tokens are equipped with extensive contextual information, empowering masked tokens to leverage the enhanced unmasked tokens for MIM. As a result, our simple remedy trains more discriminative representations revealed by achieving 84.2% top-1 accuracy with ViT-B on ImageNet-1K with 0.6%p gain. We attribute the success to the enhanced pre-training method, as evidenced by the singular value spectrum and attention analyses. Finally, our models achieve significant performance gains at the downstream semantic segmentation and fine-grained visual classification tasks; and on diverse robust evaluation metrics. Code is available at https://github.com/naver-ai/lut
Authors: Zhang Li, Biao Yang, Qiang Liu, Zhiyin Ma, Shuo Zhang, Jingxu Yang, Yabo Sun, Yuliang Liu, Xiang Bai
Abstract: Large Multimodal Models (LMMs) have shown promise in vision-language tasks but struggle with high-resolution input and detailed scene understanding. Addressing these challenges, we introduce Monkey to enhance LMM capabilities. Firstly, Monkey processes input images by dividing them into uniform patches, each matching the size (e.g., 448x448) used in the original training of the well-trained vision encoder. Equipped with individual adapter for each patch, Monkey can handle higher resolutions up to 1344x896 pixels, enabling the detailed capture of complex visual information. Secondly, it employs a multi-level description generation method, enriching the context for scene-object associations. This two-part strategy ensures more effective learning from generated data: the higher resolution allows for a more detailed capture of visuals, which in turn enhances the effectiveness of comprehensive descriptions. Extensive ablative results validate the effectiveness of our designs. Additionally, experiments on 18 datasets further demonstrate that Monkey surpasses existing LMMs in many tasks like Image Captioning and various Visual Question Answering formats. Specially, in qualitative tests focused on dense text question answering, Monkey has exhibited encouraging results compared with GPT4V. Code is available at https://github.com/Yuliang-Liu/Monkey.
Authors: Jun-En Ding, Chien-Chin Hsu, Feng Liu
Abstract: Parkinson's Disease (PD) affects millions globally, impacting movement. Prior research utilized deep learning for PD prediction, primarily focusing on medical images, neglecting the data's underlying manifold structure. This work proposes a multimodal approach encompassing both image and non-image features, leveraging contrastive cross-view graph fusion for PD classification. We introduce a novel multimodal co-attention module, integrating embeddings from separate graph views derived from low-dimensional representations of images and clinical features. This enables more robust and structured feature extraction for improved multi-view data analysis. Additionally, a simplified contrastive loss-based fusion method is devised to enhance cross-view fusion learning. Our graph-view multimodal approach achieves an accuracy of 0.91 and an area under the receiver operating characteristic curve (AUC) of 0.93 in five-fold cross-validation. It also demonstrates superior predictive capabilities on non-image data compared to solely machine learning-based methods.
Authors: Qi Liu, Xinchen Liu, Kun Liu, Xiaoyan Gu, Wu Liu
Abstract: Multi-modal human action segmentation is a critical and challenging task with a wide range of applications. Nowadays, the majority of approaches concentrate on the fusion of dense signals (i.e., RGB, optical flow, and depth maps). However, the potential contributions of sparse IoT sensor signals, which can be crucial for achieving accurate recognition, have not been fully explored. To make up for this, we introduce a Sparse signalguided Transformer (SigFormer) to combine both dense and sparse signals. We employ mask attention to fuse localized features by constraining cross-attention within the regions where sparse signals are valid. However, since sparse signals are discrete, they lack sufficient information about the temporal action boundaries. Therefore, in SigFormer, we propose to emphasize the boundary information at two stages to alleviate this problem. In the first feature extraction stage, we introduce an intermediate bottleneck module to jointly learn both category and boundary features of each dense modality through the inner loss functions. After the fusion of dense modalities and sparse signals, we then devise a two-branch architecture that explicitly models the interrelationship between action category and temporal boundary. Experimental results demonstrate that SigFormer outperforms the state-of-the-art approaches on a multi-modal action segmentation dataset from real industrial environments, reaching an outstanding F1 score of 0.958. The codes and pre-trained models have been available at https://github.com/LIUQI-creat/SigFormer.
Authors: Yang Ai, Qiang Bai, Jindong Li, Xi Yang
Abstract: Point cloud registration (PCR) is an essential task in 3D vision. Existing methods achieve increasingly higher accuracy. However, a large proportion of non-overlapping points in point cloud registration consume a lot of computational resources while negatively affecting registration accuracy. To overcome this challenge, we introduce a novel Efficient Point Cloud Registration via Dynamic Iteration framework, DIPR, that makes the neural network interactively focus on overlapping points based on sparser input points. We design global and local registration stages to achieve efficient course-tofine processing. Beyond basic matching modules, we propose the Refined Nodes to narrow down the scope of overlapping points by using adopted density-based clustering to significantly reduce the computation amount. And our SC Classifier serves as an early-exit mechanism to terminate the registration process in time according to matching accuracy. Extensive experiments on multiple datasets show that our proposed approach achieves superior registration accuracy while significantly reducing computational time and GPU memory consumption compared to state-of-the-art methods.
Authors: Lei Zhang, Fangxun Shu, Tianyang Liu, Sucheng Ren, Hao Jiang, Cihang Xie
Abstract: The increasing availability of image-text pairs has largely fueled the rapid advancement in vision-language foundation models. However, the vast scale of these datasets inevitably introduces significant variability in data quality, which can adversely affect the model performance. This highlights the critical role of data filtering, not only to enhance training efficiency but also to improve overall data quality. Existing methods typically rely on metrics such as CLIP Score and BLIP Score, which are derived from pre-trained models. However, these models are often trained on uncurated, noisy datasets, which can perpetuate errors and misalignments in the filtered dataset. We present a novel algorithm that incorporates human knowledge on image-text alignment to guide filtering vast corpus of web-crawled image-text datasets into a compact and high-quality form. To systemically capture human preferences on image-text alignments, we collect a diverse image-text dataset where each image is associated with multiple captions from various sources, and establish a comprehensive set of both subjective and objective criteria for critically guiding the alignment assessment from labelers. Additionally, we train a reward model on these human-preference annotations to internalize the nuanced human understanding of image-text alignment. The resulting reward model thus can act as a human-like referee to filter image-text pairs. Extensive experiments demonstrate that we can maintain, sometimes even improve, model performance while compressing the image-text datasets up to ~90%. An impressive example is that, by aggressively reducing the total training sample from 130M to only 15.5M, our BLIP-B/16 models consistently show an average improvement of 2.9% on retrieval tasks and 11.5% on captioning tasks compared to full-size-dataset counterparts.
Authors: Qian Wang, Yaoyao Liu, Hefei Ling, Yingwei Li, Qihao Liu, Ping Li, Jiazhong Chen, Alan Yuille, Ning Yu
Abstract: In response to the rapidly evolving nature of adversarial attacks against visual classifiers on a monthly basis, numerous defenses have been proposed to generalize against as many known attacks as possible. However, designing a defense method that generalizes to all types of attacks is not realistic because the environment in which defense systems operate is dynamic and comprises various unique attacks that emerge as time goes on. A well-matched approach to the dynamic environment lies in a defense system that continuously collects adversarial data online to quickly improve itself. Therefore, we put forward a practical defense deployment against a challenging threat model and propose, for the first time, the Continual Adversarial Defense (CAD) framework that adapts to attack sequences under four principles: (1) continual adaptation to new attacks without catastrophic forgetting, (2) few-shot adaptation, (3) memory-efficient adaptation, and (4) high accuracy on both clean and adversarial data. We explore and integrate cutting-edge continual learning, few-shot learning, and ensemble learning techniques to qualify the principles. Extensive experiments validate the effectiveness of our approach against multiple stages of modern adversarial attacks and demonstrate significant improvements over numerous baseline methods. In particular, CAD is capable of quickly adapting with minimal budget and a low cost of defense failure while maintaining good performance against previous attacks. Our research sheds light on a brand-new paradigm for continual defense adaptation against dynamic and evolving attacks.
Authors: Vardaan Pahuja, Weidi Luo, Yu Gu, Cheng-Hao Tu, Hong-You Chen, Tanya Berger-Wolf, Charles Stewart, Song Gao, Wei-Lun Chao, Yu Su
Abstract: Camera traps are important tools in animal ecology for biodiversity monitoring and conservation. However, their practical application is limited by issues such as poor generalization to new and unseen locations. Images are typically associated with diverse forms of context, which may exist in different modalities. In this work, we exploit the structured context linked to camera trap images to boost out-of-distribution generalization for species classification tasks in camera traps. For instance, a picture of a wild animal could be linked to details about the time and place it was captured, as well as structured biological knowledge about the animal species. While often overlooked by existing studies, incorporating such context offers several potential benefits for better image understanding, such as addressing data scarcity and enhancing generalization. However, effectively incorporating such heterogeneous context into the visual domain is a challenging problem. To address this, we propose a novel framework that transforms species classification as link prediction in a multimodal knowledge graph (KG). This framework enables the seamless integration of diverse multimodal contexts for visual recognition. We apply this framework for out-of-distribution species classification on the iWildCam2020-WILDS and Snapshot Mountain Zebra datasets and achieve competitive performance with state-of-the-art approaches. Furthermore, our framework enhances sample efficiency for recognizing under-represented species.
Authors: Gyutae Hwang, Sang Jun Lee
Abstract: Human health can be critically affected by cardiovascular diseases, such as hypertension, arrhythmias, and stroke. Heart rate and blood pressure are important biometric information for the monitoring of cardiovascular system and early diagnosis of cardiovascular diseases. Existing methods for estimating the heart rate are based on electrocardiography and photoplethyomography, which require contacting the sensor to the skin surface. Moreover, catheter and cuff-based methods for measuring blood pressure cause inconvenience and have limited applicability. Therefore, in this thesis, we propose a vision-based method for estimating the heart rate and blood pressure. This thesis proposes a 2-stage deep learning framework consisting of a dual remote photoplethysmography network (DRP-Net) and bounded blood pressure network (BBP-Net). In the first stage, DRP-Net infers remote photoplethysmography (rPPG) signals for the acral and facial regions, and these phase-shifted rPPG signals are utilized to estimate the heart rate. In the second stage, BBP-Net integrates temporal features and analyzes phase discrepancy between the acral and facial rPPG signals to estimate SBP and DBP values. To improve the accuracy of estimating the heart rate, we employed a data augmentation method based on a frame interpolation model. Moreover, we designed BBP-Net to infer blood pressure within a predefined range by incorporating a scaled sigmoid function. Our method resulted in estimating the heart rate with the mean absolute error (MAE) of 1.78 BPM, reducing the MAE by 34.31 % compared to the recent method, on the MMSE-HR dataset. The MAE for estimating the systolic blood pressure (SBP) and diastolic blood pressure (DBP) were 10.19 mmHg and 7.09 mmHg. On the V4V dataset, the MAE for the heart rate, SBP, and DBP were 3.83 BPM, 13.64 mmHg, and 9.4 mmHg, respectively.
Authors: Ziyun Li, Christoph Meinel, Haojin Yang
Abstract: Generalized Class Discovery (GCD) plays a pivotal role in discerning both known and unknown categories from unlabeled datasets by harnessing the insights derived from a labeled set comprising recognized classes. A significant limitation in prevailing GCD methods is their presumption of an equitably distributed category occurrence in unlabeled data. Contrary to this assumption, visual classes in natural environments typically exhibit a long-tailed distribution, with known or prevalent categories surfacing more frequently than their rarer counterparts. Our research endeavors to bridge this disconnect by focusing on the long-tailed Generalized Category Discovery (Long-tailed GCD) paradigm, which echoes the innate imbalances of real-world unlabeled datasets. In response to the unique challenges posed by Long-tailed GCD, we present a robust methodology anchored in two strategic regularizations: (i) a reweighting mechanism that bolsters the prominence of less-represented, tail-end categories, and (ii) a class prior constraint that aligns with the anticipated class distribution. Comprehensive experiments reveal that our proposed method surpasses previous state-of-the-art GCD methods by achieving an improvement of approximately 6 - 9% on ImageNet100 and competitive performance on CIFAR100.
Authors: Prarthana Bhattacharyya, Chengjie Huang, Krzysztof Czarnecki
Abstract: This paper addresses motion forecasting in multi-agent environments, pivotal for ensuring safety of autonomous vehicles. Traditional as well as recent data-driven marginal trajectory prediction methods struggle to properly learn non-linear agent-to-agent interactions. We present SSL-Interactions that proposes pretext tasks to enhance interaction modeling for trajectory prediction. We introduce four interaction-aware pretext tasks to encapsulate various aspects of agent interactions: range gap prediction, closest distance prediction, direction of movement prediction, and type of interaction prediction. We further propose an approach to curate interaction-heavy scenarios from datasets. This curated data has two advantages: it provides a stronger learning signal to the interaction model, and facilitates generation of pseudo-labels for interaction-centric pretext tasks. We also propose three new metrics specifically designed to evaluate predictions in interactive scenes. Our empirical evaluations indicate SSL-Interactions outperforms state-of-the-art motion forecasting methods quantitatively with up to 8% improvement, and qualitatively, for interaction-heavy scenarios.
Authors: Yifan Zhang, Siyu Ren, Junhui Hou, Jinjian Wu, Yixuan Yuan, Guangming Shi
Abstract: This paper introduces a novel self-supervised learning framework for enhancing 3D perception in autonomous driving scenes. Specifically, our approach, namely NCLR, focuses on 2D-3D neural calibration, a novel pretext task that estimates the rigid pose aligning camera and LiDAR coordinate systems. First, we propose the learnable transformation alignment to bridge the domain gap between image and point cloud data, converting features into a unified representation space for effective comparison and matching. Second, we identify the overlapping area between the image and point cloud with the fused features. Third, we establish dense 2D-3D correspondences to estimate the rigid pose. The framework not only learns fine-grained matching from points to pixels but also achieves alignment of the image and point cloud at a holistic level, understanding their relative pose. We demonstrate the efficacy of NCLR by applying the pre-trained backbone to downstream tasks, such as LiDAR-based 3D semantic segmentation, object detection, and panoptic segmentation. Comprehensive experiments on various datasets illustrate the superiority of NCLR over existing self-supervised methods. The results confirm that joint learning from different modalities significantly enhances the network's understanding abilities and effectiveness of learned representation. The code is publicly available at https://github.com/Eaphan/NCLR.
Authors: Fei Teng, Jiaming Zhang, Jiawei Liu, Kunyu Peng, Xina Cheng, Zhiyong Li, Kailun Yang
Abstract: Leveraging rich information is crucial for dense prediction tasks. Light field (LF) cameras are instrumental in this regard, as they allow data to be sampled from various perspectives. This capability provides valuable spatial, depth, and angular information, enhancing scene-parsing tasks. However, we have identified two overlooked issues for the LF salient object detection (SOD) task. (1): Previous approaches predominantly employ a customized two-stream design to discover the spatial and depth features within light field images. The network struggles to learn the implicit angular information between different images due to a lack of intra-network data connectivity. (2): Little research has been directed towards the data augmentation strategy for LF SOD. Research on inter-network data connectivity is scant. In this study, we propose an efficient paradigm (LF Tracy) to address those issues. This comprises a single-pipeline encoder paired with a highly efficient information aggregation (IA) module (around 8M parameters) to establish an intra-network connection. Then, a simple yet effective data augmentation strategy called MixLD is designed to bridge the inter-network connections. Owing to this innovative paradigm, our model surpasses the existing state-of-the-art method through extensive experiments. Especially, LF Tracy demonstrates a 23% improvement over previous results on the latest large-scale PKU dataset. The source code is publicly available at: https://github.com/FeiBryantkit/LF-Tracy.
Authors: Letian Huang, Jiayang Bai, Jie Guo, Yuanqi Li, Yanwen Guo
Abstract: 3D Gaussian Splatting has garnered extensive attention and application in real-time neural rendering. Concurrently, concerns have been raised about the limitations of this technology in aspects such as point cloud storage, performance, and robustness in sparse viewpoints, leading to various improvements. However, there has been a notable lack of attention to the fundamental problem of projection errors introduced by the local affine approximation inherent in the splatting itself, and the consequential impact of these errors on the quality of photo-realistic rendering. This paper addresses the projection error function of 3D Gaussian Splatting, commencing with the residual error from the first-order Taylor expansion of the projection function. The analysis establishes a correlation between the error and the Gaussian mean position. Subsequently, leveraging function optimization theory, this paper analyzes the function's minima to provide an optimal projection strategy for Gaussian Splatting referred to Optimal Gaussian Splatting, which can accommodate a variety of camera models. Experimental validation further confirms that this projection methodology reduces artifacts, resulting in a more convincingly realistic rendering.
Authors: Farnoosh Arefi, Amir M. Mansourian, Shohreh Kasaei
Abstract: Deep spectral methods reframe the image decomposition process as a graph partitioning task by extracting features using self-supervised learning and utilizing the Laplacian of the affinity matrix to obtain eigensegments. However, instance segmentation has received less attention compared to other tasks within the context of deep spectral methods. This paper addresses the fact that not all channels of the feature map extracted from a self-supervised backbone contain sufficient information for instance segmentation purposes. In fact, Some channels are noisy and hinder the accuracy of the task. To overcome this issue, this paper proposes two channel reduction modules: Noise Channel Reduction (NCR) and Deviation-based Channel Reduction (DCR). The NCR retains channels with lower entropy, as they are less likely to be noisy, while DCR prunes channels with low standard deviation, as they lack sufficient information for effective instance segmentation. Furthermore, the paper demonstrates that the dot product, commonly used in deep spectral methods, is not suitable for instance segmentation due to its sensitivity to feature map values, potentially leading to incorrect instance segments. A new similarity metric called Bray-Curtis over Chebyshev (BoC) is proposed to address this issue. It takes into account the distribution of features in addition to their values, providing a more robust similarity measure for instance segmentation. Quantitative and qualitative results on the Youtube-VIS2019 dataset highlight the improvements achieved by the proposed channel reduction methods and the use of BoC instead of the conventional dot product for creating the affinity matrix. These improvements are observed in terms of mean Intersection over Union and extracted instance segments, demonstrating enhanced instance segmentation performance. The code is available on: https://github.com/farnooshar/SpecUnIIS
Authors: Yitao Zhu, Sheng Wang, Mengjie Xu, Zixu Zhuang, Zhixin Wang, Kaidong Wang, Han Zhang, Qian Wang
Abstract: Multiple cameras can provide comprehensive multi-view video coverage of a person. Fusing this multi-view data is crucial for tasks like behavioral analysis, although it traditionally requires camera calibration, a process that is often complex. Moreover, previous studies have overlooked the challenges posed by self-occlusion under multiple views and the continuity of human body shape estimation. In this study, we introduce a method to reconstruct the 3D human body from multiple uncalibrated camera views. Initially, we utilize a pre-trained human body encoder to process each camera view individually, enabling the reconstruction of human body models and parameters for each view along with predicted camera positions. Rather than merely averaging the models across views, we develop a neural network trained to assign weights to individual views for all human body joints, based on the estimated distribution of joint distances from each camera. Additionally, we focus on the mesh surface of the human body for dynamic fusion, allowing for the seamless integration of facial expressions and body shape into a unified human body model. Our method has shown excellent performance in reconstructing the human body on two public datasets, advancing beyond previous work from the SMPL model to the SMPL-X model. This extension incorporates more complex hand poses and facial expressions, enhancing the detail and accuracy of the reconstructions. Crucially, it supports the flexible ad-hoc deployment of any number of cameras, offering significant potential for various applications. Our code is available at https://github.com/AbsterZhu/MUC.
Authors: Amir M. Mansourian, Arya Jalali, Rozhan Ahmadi, Shohreh Kasaei
Abstract: In contrast to existing complex methodologies commonly employed for distilling knowledge from a teacher to a student, this paper showcases the efficacy of a simple yet powerful method for utilizing refined feature maps to transfer attention. The proposed method has proven to be effective in distilling rich information, outperforming existing methods in semantic segmentation as a dense prediction task. The proposed Attention-guided Feature Distillation (AttnFD) method, employs the Convolutional Block Attention Module (CBAM), which refines feature maps by taking into account both channel-specific and spatial information content. Simply using the Mean Squared Error (MSE) loss function between the refined feature maps of the teacher and the student, AttnFD demonstrates outstanding performance in semantic segmentation, achieving state-of-the-art results in terms of improving the mean Intersection over Union (mIoU) of the student network on the PascalVoc 2012, Cityscapes, COCO, and CamVid datasets.
Authors: Sarthak Kumar Maharana, Baoming Zhang, Yunhui Guo
Abstract: Real-world vision models in dynamic environments face rapid shifts in domain distributions, leading to decreased recognition performance. Using unlabeled test data, continual test-time adaptation (CTTA) directly adjusts a pre-trained source discriminative model to these changing domains. A highly effective CTTA method involves applying layer-wise adaptive learning rates for selectively adapting pre-trained layers. However, it suffers from the poor estimation of domain shift and the inaccuracies arising from the pseudo-labels. This work aims to overcome these limitations by identifying layers for adaptation via quantifying model prediction uncertainty without relying on pseudo-labels. We utilize the magnitude of gradients as a metric, calculated by backpropagating the KL divergence between the softmax output and a uniform distribution, to select layers for further adaptation. Subsequently, for the parameters exclusively belonging to these selected layers, with the remaining ones frozen, we evaluate their sensitivity to approximate the domain shift and adjust their learning rates accordingly. We conduct extensive image classification experiments on CIFAR-10C, CIFAR-100C, and ImageNet-C, demonstrating the superior efficacy of our method compared to prior approaches.
Authors: Justin Kay, Timm Haucke, Suzanne Stathatos, Siqi Deng, Erik Young, Pietro Perona, Sara Beery, Grant Van Horn
Abstract: Object detectors often perform poorly on data that differs from their training set. Domain adaptive object detection (DAOD) methods have recently demonstrated strong results on addressing this challenge. Unfortunately, we identify systemic benchmarking pitfalls that call past results into question and hamper further progress: (a) Overestimation of performance due to underpowered baselines, (b) Inconsistent implementation practices preventing transparent comparisons of methods, and (c) Lack of generality due to outdated backbones and lack of diversity in benchmarks. We address these problems by introducing: (1) A unified benchmarking and implementation framework, Align and Distill (ALDI), enabling comparison of DAOD methods and supporting future development, (2) A fair and modern training and evaluation protocol for DAOD that addresses benchmarking pitfalls, (3) A new DAOD benchmark dataset, CFC-DAOD, enabling evaluation on diverse real-world data, and (4) A new method, ALDI++, that achieves state-of-the-art results by a large margin. ALDI++ outperforms the previous state-of-the-art by +3.5 AP50 on Cityscapes to Foggy Cityscapes, +5.7 AP50 on Sim10k to Cityscapes (where ours is the only method to outperform a fair baseline), and +0.6 AP50 on CFC Kenai to Channel. Our framework, dataset, and state-of-the-art method offer a critical reset for DAOD and provide a strong foundation for future research. Code and data are available: https://github.com/justinkay/aldi and https://github.com/visipedia/caltech-fish-counting.
URLs: https://github.com/justinkay/aldi, https://github.com/visipedia/caltech-fish-counting.
Authors: Ziming Wang, Ziling Wang, Huaning Li, Lang Qin, Runhao Jiang, De Ma, Huajin Tang
Abstract: Event cameras, with their high dynamic range and temporal resolution, are ideally suited for object detection, especially under scenarios with motion blur and challenging lighting conditions. However, while most existing approaches prioritize optimizing spatiotemporal representations with advanced detection backbones and early aggregation functions, the crucial issue of adaptive event sampling remains largely unaddressed. Spiking Neural Networks (SNNs), which operate on an event-driven paradigm through sparse spike communication, emerge as a natural fit for addressing this challenge. In this study, we discover that the neural dynamics of spiking neurons align closely with the behavior of an ideal temporal event sampler. Motivated by this insight, we propose a novel adaptive sampling module that leverages recurrent convolutional SNNs enhanced with temporal memory, facilitating a fully end-to-end learnable framework for event-based detection. Additionally, we introduce Residual Potential Dropout (RPD) and Spike-Aware Training (SAT) to regulate potential distribution and address performance degradation encountered in spike-based sampling modules. Empirical evaluation on neuromorphic detection datasets demonstrates that our approach outperforms existing state-of-the-art spike-based methods with significantly fewer parameters and time steps. For instance, our method yields a 4.4\% mAP improvement on the Gen1 dataset, while requiring 38\% fewer parameters and only three time steps. Moreover, the applicability and effectiveness of our adaptive sampling methodology extend beyond SNNs, as demonstrated through further validation on conventional non-spiking models. Code is available at https://github.com/Windere/EAS-SNN.
Authors: Yao Wei, Martin Renqiang Min, George Vosselman, Li Erran Li, Michael Ying Yang
Abstract: Compositional 3D scene synthesis has diverse applications across a spectrum of industries such as robotics, films, and video games, as it closely mirrors the complexity of real-world multi-object environments. Conventional works typically employ shape retrieval based frameworks which naturally suffer from limited shape diversity. Recent progresses have been made in object shape generation with generative models such as diffusion models, which increases the shape fidelity. However, these approaches separately treat 3D shape generation and layout generation. The synthesized scenes are usually hampered by layout collision, which suggests that the scene-level fidelity is still under-explored. In this paper, we aim at generating realistic and reasonable 3D indoor scenes from scene graph. To enrich the priors of the given scene graph inputs, large language model is utilized to aggregate the global-wise features with local node-wise and edge-wise features. With a unified graph encoder, graph features are extracted to guide joint layout-shape generation. Additional regularization is introduced to explicitly constrain the produced 3D layouts. Benchmarked on the SG-FRONT dataset, our method achieves better 3D scene synthesis, especially in terms of scene-level fidelity. The source code will be released after publication.
Authors: Sayanton V. Dibbo, Adam Breuer, Juston Moore, Michael Teti
Abstract: Recent model inversion attack algorithms permit adversaries to reconstruct a neural network's private and potentially sensitive training data by repeatedly querying the network. In this work, we develop a novel network architecture that leverages sparse-coding layers to obtain superior robustness to this class of attacks. Three decades of computer science research has studied sparse coding in the context of image denoising, object recognition, and adversarial misclassification settings, but to the best of our knowledge, its connection to state-of-the-art privacy vulnerabilities remains unstudied. In this work, we hypothesize that sparse coding architectures suggest an advantageous means to defend against model inversion attacks because they allow us to control the amount of irrelevant private information encoded by a network in a manner that is known to have little effect on classification accuracy. Specifically, compared to networks trained with a variety of state-of-the-art defenses, our sparse-coding architectures maintain comparable or higher classification accuracy while degrading state-of-the-art training data reconstructions by factors of 1.1 to 18.3 across a variety of reconstruction quality metrics (PSNR, SSIM, FID). This performance advantage holds across 5 datasets ranging from CelebA faces to medical images and CIFAR-10, and across various state-of-the-art SGD-based and GAN-based inversion attacks, including Plug-&-Play attacks. We provide a cluster-ready PyTorch codebase to promote research and standardize defense evaluations.
Authors: Young Seok Jeon, Hongfei Yang, Huazhu Fu, Mengling Feng
Abstract: Imposing key anatomical features, such as the number of organs, their shapes and relative positions, is crucial for building a robust multi-organ segmentation model. Current attempts to incorporate anatomical features include broadening the effective receptive field (ERF) size with data-intensive modules, or introducing anatomical constraints that scales poorly to multi-organ segmentation. We introduce a novel architecture called the Anatomy-Informed Cascaded Segmentation Network (AIC-Net). AIC-Net incorporates a learnable input termed "Anatomical Prior", which can be adapted to patient-specific anatomy using a differentiable spatial deformation. The deformed prior later guides decoder layers towards more anatomy-informed predictions. We repeat this process at a local patch level to enhance the representation of intricate objects, resulting in a cascaded network structure. AIC-Net is a general method that enhances any existing segmentation models to be more anatomy-aware. We have validated the performance of AIC-Net, with various backbones, on two multi-organ segmentation tasks: abdominal organs and vertebrae. For each respective task, our benchmarks demonstrate improved dice score and Hausdorff distance.
Authors: Jie Zhu, Jirong Zha, Ding Li, Leye Wang
Abstract: Self-supervised learning shows promise in harnessing extensive unlabeled data, but it also confronts significant privacy concerns, especially in vision. In this paper, we aim to perform membership inference on visual self-supervised models in a more realistic setting: self-supervised training method and details are unknown for an adversary when attacking as he usually faces a black-box system in practice. In this setting, considering that self-supervised model could be trained by completely different self-supervised paradigms, e.g., masked image modeling and contrastive learning, with complex training details, we propose a unified membership inference method called PartCrop. It is motivated by the shared part-aware capability among models and stronger part response on the training data. Specifically, PartCrop crops parts of objects in an image to query responses with the image in representation space. We conduct extensive attacks on self-supervised models with different training protocols and structures using three widely used image datasets. The results verify the effectiveness and generalization of PartCrop. Moreover, to defend against PartCrop, we evaluate two common approaches, i.e., early stop and differential privacy, and propose a tailored method called shrinking crop scale range. The defense experiments indicate that all of them are effective. Our code is available at https://github.com/JiePKU/PartCrop.
Authors: Yi-Xin Huang, Hou-I Liu, Hong-Han Shuai, Wen-Huang Cheng
Abstract: Despite previous DETR-like methods having performed successfully in generic object detection, tiny object detection is still a challenging task for them since the positional information of object queries is not customized for detecting tiny objects, whose scale is extraordinarily smaller than general objects. Also, DETR-like methods using a fixed number of queries make them unsuitable for aerial datasets, which only contain tiny objects, and the numbers of instances are imbalanced between different images. Thus, we present a simple yet effective model, named DQ-DETR, which consists of three different components: categorical counting module, counting-guided feature enhancement, and dynamic query selection to solve the above-mentioned problems. DQ-DETR uses the prediction and density maps from the categorical counting module to dynamically adjust the number of object queries and improve the positional information of queries. Our model DQ-DETR outperforms previous CNN-based and DETR-like methods, achieving state-of-the-art mAP 30.2% on the AI-TOD-V2 dataset, which mostly consists of tiny objects.
Authors: Zhiyuan Yang, Ryan Rad
Abstract: Solar Photovoltaic (PV) technology is increasingly recognized as a pivotal solution in the global pursuit of clean and renewable energy. This technology addresses the urgent need for sustainable energy alternatives by converting solar power into electricity without greenhouse gas emissions. It not only curtails global carbon emissions but also reduces reliance on finite, non-renewable energy sources. In this context, monitoring solar panel farms becomes essential for understanding and facilitating the worldwide shift toward clean energy. This study contributes to this effort by developing the first comprehensive global dataset of multispectral satellite imagery of solar panel farms. This dataset is intended to form the basis for training robust machine learning models, which can accurately map and analyze the expansion and distribution of solar panel farms globally. The insights gained from this endeavor will be instrumental in guiding informed decision-making for a sustainable energy future. https://github.com/yzyly1992/GloSoFarID
Authors: Bonan Ding, Jin Xie, Jing Nie, Jiale Cao, Xuelong Li, Yanwei Pang
Abstract: Due to its cost-effectiveness and widespread availability, monocular 3D object detection, which relies solely on a single camera during inference, holds significant importance across various applications, including autonomous driving and robotics. Nevertheless, directly predicting the coordinates of objects in 3D space from monocular images poses challenges. Therefore, an effective solution involves transforming monocular images into LiDAR-like representations and employing a LiDAR-based 3D object detector to predict the 3D coordinates of objects. The key step in this method is accurately converting the monocular image into a reliable point cloud form. In this paper, we present VFMM3D, an innovative framework that leverages the capabilities of Vision Foundation Models (VFMs) to accurately transform single-view images into LiDAR point cloud representations. VFMM3D utilizes the Segment Anything Model (SAM) and Depth Anything Model (DAM) to generate high-quality pseudo-LiDAR data enriched with rich foreground information. Specifically, the Depth Anything Model (DAM) is employed to generate dense depth maps. Subsequently, the Segment Anything Model (SAM) is utilized to differentiate foreground and background regions by predicting instance masks. These predicted instance masks and depth maps are then combined and projected into 3D space to generate pseudo-LiDAR points. Finally, any object detectors based on point clouds can be utilized to predict the 3D coordinates of objects. Comprehensive experiments are conducted on two challenging 3D object detection datasets, KITTI and Waymo. Our VFMM3D establishes a new state-of-the-art performance on both datasets. Additionally, experimental results demonstrate the generality of VFMM3D, showcasing its seamless integration into various LiDAR-based 3D object detectors.
Authors: Haniyeh Ehsani Oskouie, Mohammad-Shahram Moin, Shohreh Kasaei
Abstract: Deep neural networks have made significant advancements in accurately estimating scene flow using point clouds, which is vital for many applications like video analysis, action recognition, and navigation. The robustness of these techniques, however, remains a concern, particularly in the face of adversarial attacks that have been proven to deceive state-of-the-art deep neural networks in many domains. Surprisingly, the robustness of scene flow networks against such attacks has not been thoroughly investigated. To address this problem, the proposed approach aims to bridge this gap by introducing adversarial white-box attacks specifically tailored for scene flow networks. Experimental results show that the generated adversarial examples obtain up to 33.7 relative degradation in average end-point error on the KITTI and FlyingThings3D datasets. The study also reveals the significant impact that attacks targeting point clouds in only one dimension or color channel have on average end-point error. Analyzing the success and failure of these attacks on the scene flow networks and their 2D optical flow network variants shows a higher vulnerability for the optical flow networks. Code is available at https://github.com/aheldis/Attack-on-Scene-Flow-using-Point-Clouds.git.
URLs: https://github.com/aheldis/Attack-on-Scene-Flow-using-Point-Clouds.git.
Authors: Wen Liang, Peipei Ran, Mengchao Bai, Xiao Liu, P. Bilha Githinji, Wei Zhao, Peiwu Qin
Abstract: Salient object detection (SOD) aims at finding the most salient objects in images and outputs pixel-level binary masks. Transformer-based methods achieve promising performance due to their global semantic understanding, crucial for identifying salient objects. However, these models tend to be large and require numerous training parameters. To better harness the potential of transformers for SOD, we propose a novel parameter-efficient fine-tuning method aimed at reducing the number of training parameters while enhancing the salient object detection capability. Our model, termed EXternal Prompt features Enhanced adapteR Tuning (ExPert), features an encoder-decoder structure with adapters and injectors interspersed between the layers of a frozen transformer encoder. The adapter modules adapt the pretrained backbone to SOD while the injector modules incorporate external prompt features to enhance the awareness of salient objects. Comprehensive experiments demonstrate the superiority of our method. Surpassing former state-of-the-art (SOTA) models across five SOD datasets, ExPert achieves 0.215 mean absolute error (MAE) in the ECSSD dataset with 80.2M trained parameters, 21% better than SelfReformer and 47% better than EGNet.
Authors: Peter Kulits, Haiwen Feng, Weiyang Liu, Victoria Abrevaya, Michael J. Black
Abstract: Inverse graphics -- the task of inverting an image into physical variables that, when rendered, enable reproduction of the observed scene -- is a fundamental challenge in computer vision and graphics. Successfully disentangling an image into its constituent elements, such as the shape, color, and material properties of the objects of the 3D scene that produced it, requires a comprehensive understanding of the environment. This complexity limits the ability of existing carefully engineered approaches to generalize across domains. Inspired by the zero-shot ability of large language models (LLMs) to generalize to novel contexts, we investigate the possibility of leveraging the broad world knowledge encoded in such models to solve inverse-graphics problems. To this end, we propose the Inverse-Graphics Large Language Model (IG-LLM), an inverse-graphics framework centered around an LLM, that autoregressively decodes a visual embedding into a structured, compositional 3D-scene representation. We incorporate a frozen pre-trained visual encoder and a continuous numeric head to enable end-to-end training. Through our investigation, we demonstrate the potential of LLMs to facilitate inverse graphics through next-token prediction, without the application of image-space supervision. Our analysis enables new possibilities for precise spatial reasoning about images that exploit the visual knowledge of LLMs. We release our code and data at https://ig-llm.is.tue.mpg.de/ to ensure the reproducibility of our investigation and to facilitate future research.
Authors: Nithish Muthuchamy Selvaraj, Xiaobao Guo, Adams Wai-Kin Kong, Alex Kot
Abstract: Concept Bottleneck Models (CBM) map images to human-interpretable concepts before making class predictions. Recent approaches automate CBM construction by prompting Large Language Models (LLMs) to generate text concepts and employing Vision Language Models (VLMs) to score these concepts for CBM training. However, it is desired to build CBMs with concepts defined by human experts rather than LLM-generated ones to make them more trustworthy. In this work, we closely examine the faithfulness of VLM concept scores for such expert-defined concepts in domains like fine-grained bird species and animal classification. Our investigations reveal that VLMs like CLIP often struggle to correctly associate a concept with the corresponding visual input, despite achieving a high classification performance. This misalignment renders the resulting models difficult to interpret and less reliable. To address this issue, we propose a novel Contrastive Semi-Supervised (CSS) learning method that leverages a few labeled concept samples to activate truthful visual concepts and improve concept alignment in the CLIP model. Extensive experiments on three benchmark datasets demonstrate that our method significantly enhances both concept (+29.95) and classification (+3.84) accuracies yet requires only a fraction of human-annotated concept labels. To further improve the classification performance, we introduce a class-level intervention procedure for fine-grained classification problems that identifies the confounding classes and intervenes in their concept space to reduce errors.
Authors: Siyu Li, Jiacheng Lin, Hao Shi, Jiaming Zhang, Song Wang, You Yao, Zhiyong Li, Kailun Yang
Abstract: Temporal information plays a pivotal role in Bird's-Eye-View (BEV) driving scene understanding, which can alleviate the visual information sparsity. However, the indiscriminate temporal fusion method will cause the barrier of feature redundancy when constructing vectorized High-Definition (HD) maps. In this paper, we revisit the temporal fusion of vectorized HD maps, focusing on temporal instance consistency and temporal map consistency learning. To improve the representation of instances in single-frame maps, we introduce a novel method, DTCLMapper. This approach uses a dual-stream temporal consistency learning module that combines instance embedding with geometry maps. In the instance embedding component, our approach integrates temporal Instance Consistency Learning (ICL), ensuring consistency from vector points and instance features aggregated from points. A vectorized points pre-selection module is employed to enhance the regression efficiency of vector points from each instance. Then aggregated instance features obtained from the vectorized points preselection module are grounded in contrastive learning to realize temporal consistency, where positive and negative samples are selected based on position and semantic information. The geometry mapping component introduces Map Consistency Learning (MCL) designed with self-supervised learning. The MCL enhances the generalization capability of our consistent learning approach by concentrating on the global location and distribution constraints of the instances. Extensive experiments on well-recognized benchmarks indicate that the proposed DTCLMapper achieves state-of-the-art performance in vectorized mapping tasks, reaching 61.9% and 65.1% mAP scores on the nuScenes and Argoverse datasets, respectively. The source code is available at https://github.com/lynn-yu/DTCLMapper.
Authors: Junhui Li, Xingsong Hou
Abstract: Decoding remote sensing images to achieve high perceptual quality, particularly at low bitrates, remains a significant challenge. To address this problem, we propose the invertible neural network-based remote sensing image compression (INN-RSIC) method. Specifically, we capture compression distortion from an existing image compression algorithm and encode it as a set of Gaussian-distributed latent variables via INN. This ensures that the compression distortion in the decoded image becomes independent of the ground truth. Therefore, by leveraging the inverse mapping of INN, we can input the decoded image along with a set of randomly resampled Gaussian distributed variables into the inverse network, effectively generating enhanced images with better perception quality. To effectively learn compression distortion, channel expansion, Haar transformation, and invertible blocks are employed to construct the INN. Additionally, we introduce a quantization module (QM) to mitigate the impact of format conversion, thus enhancing the framework's generalization and improving the perceptual quality of enhanced images. Extensive experiments demonstrate that our INN-RSIC significantly outperforms the existing state-of-the-art traditional and deep learning-based image compression methods in terms of perception quality.
Authors: Jiachen Sun, Changsheng Wang, Jiongxiao Wang, Yiwei Zhang, Chaowei Xiao
Abstract: Large language models have become increasingly prominent, also signaling a shift towards multimodality as the next frontier in artificial intelligence, where their embeddings are harnessed as prompts to generate textual content. Vision-language models (VLMs) stand at the forefront of this advancement, offering innovative ways to combine visual and textual data for enhanced understanding and interaction. However, this integration also enlarges the attack surface. Patch-based adversarial attack is considered the most realistic threat model in physical vision applications, as demonstrated in many existing literature. In this paper, we propose to address patched visual prompt injection, where adversaries exploit adversarial patches to generate target content in VLMs. Our investigation reveals that patched adversarial prompts exhibit sensitivity to pixel-wise randomization, a trait that remains robust even against adaptive attacks designed to counteract such defenses. Leveraging this insight, we introduce SmoothVLM, a defense mechanism rooted in smoothing techniques, specifically tailored to protect VLMs from the threat of patched visual prompt injectors. Our framework significantly lowers the attack success rate to a range between 0% and 5.0% on two leading VLMs, while achieving around 67.3% to 95.0% context recovery of the benign images, demonstrating a balance between security and usability.
Authors: Yujie Lu, Xiujun Li, Tsu-Jui Fu, Miguel Eckstein, William Yang Wang
Abstract: The rapid progress in Multimodal Large Language Models (MLLMs) has significantly advanced their ability to process and understand complex visual and textual information. However, the integration of multiple images and extensive textual contexts remains a challenge due to the inherent limitation of the models' capacity to handle long input sequences efficiently. In this paper, we introduce SEEKER, a multimodal large language model designed to tackle this issue. SEEKER aims to optimize the compact encoding of long text by compressing the text sequence into the visual pixel space via images, enabling the model to handle long text within a fixed token-length budget efficiently. Our empirical experiments on six long-context multimodal tasks demonstrate that SEEKER can leverage fewer image tokens to convey the same amount of textual information compared with the OCR-based approach, and is more efficient in understanding long-form multimodal input and generating long-form textual output, outperforming all existing proprietary and open-source MLLMs by large margins.
Authors: Khanh-Binh Nguyen, Chae Jung Park
Abstract: Self-supervised learning (SSL) is gaining attention for its ability to learn effective representations with large amounts of unlabeled data. Lightweight models can be distilled from larger self-supervised pre-trained models using contrastive and consistency constraints. Still, the different sizes of the projection heads make it challenging for students to mimic the teacher's embedding accurately. We propose \textsc{Retro}, which reuses the teacher's projection head for students, and our experimental results demonstrate significant improvements over the state-of-the-art on all lightweight models. For instance, when training EfficientNet-B0 using ResNet-50/101/152 as teachers, our approach improves the linear result on ImageNet to $66.9\%$, $69.3\%$, and $69.8\%$, respectively, with significantly fewer parameters.
Authors: Qian Wang, Chen Li, Yuchen Luo, Hefei Ling, Ping Li, Jiazhong Chen, Shijuan Huang, Ning Yu
Abstract: As a defense strategy against adversarial attacks, adversarial detection aims to identify and filter out adversarial data from the data flow based on discrepancies in distribution and noise patterns between natural and adversarial data. Although previous detection methods achieve high performance in detecting gradient-based adversarial attacks, new attacks based on generative models with imbalanced and anisotropic noise patterns evade detection. Even worse, existing techniques either necessitate access to attack data before deploying a defense or incur a significant time cost for inference, rendering them impractical for defending against newly emerging attacks that are unseen by defenders. In this paper, we explore the proximity relationship between adversarial noise distributions and demonstrate the existence of an open covering for them. By learning to distinguish this open covering from the distribution of natural data, we can develop a detector with strong generalization capabilities against all types of adversarial attacks. Based on this insight, we heuristically propose Perturbation Forgery, which includes noise distribution perturbation, sparse mask generation, and pseudo-adversarial data production, to train an adversarial detector capable of detecting unseen gradient-based, generative-model-based, and physical adversarial attacks, while remaining agnostic to any specific models. Comprehensive experiments conducted on multiple general and facial datasets, with a wide spectrum of attacks, validate the strong generalization of our method.
Authors: Kangye Ji, Fei Cheng, Zeqing Wang, Bohu Huang
Abstract: Sample selection is the most straightforward technique to combat label noise, aiming to distinguish mislabeled samples during training and avoid the degradation of the robustness of the model. In the workflow, $\textit{selecting possibly clean data}$ and $\textit{model update}$ are iterative. However, their interplay and intrinsic characteristics hinder the robustness and efficiency of learning with noisy labels: 1) The model chooses clean data with selection bias, leading to the accumulated error in the model update. 2) Most selection strategies leverage partner networks or supplementary information to mitigate label corruption, albeit with increased computation resources and lower throughput speed. Therefore, we employ only one network with the jump manner update to decouple the interplay and mine more semantic information from the loss for a more precise selection. Specifically, the selection of clean data for each model update is based on one of the prior models, excluding the last iteration. The strategy of model update exhibits a jump behavior in the form. Moreover, we map the outputs of the network and labels into the same semantic feature space, respectively. In this space, a detailed and simple loss distribution is generated to distinguish clean samples more effectively. Our proposed approach achieves almost up to $2.53\times$ speedup, $0.46\times$ peak memory footprint, and superior robustness over state-of-the-art works with various noise settings.
Authors: Jiahang Zhang, Lilang Lin, Shuai Yang, Jiaying Liu
Abstract: Self-supervised learning (SSL), which aims to learn meaningful prior representations from unlabeled data, has been proven effective for skeleton-based action understanding. Different from the image domain, skeleton data possesses sparser spatial structures and diverse representation forms, with the absence of background clues and the additional temporal dimension, presenting new challenges for spatial-temporal motion pretext task design. Recently, many endeavors have been made for skeleton-based SSL, achieving remarkable progress. However, a systematic and thorough review is still lacking. In this paper, we conduct, for the first time, a comprehensive survey on self-supervised skeleton-based action representation learning. Following the taxonomy of context-based, generative learning, and contrastive learning approaches, we make a thorough review and benchmark of existing works and shed light on the future possible directions. Remarkably, our investigation demonstrates that most SSL works rely on the single paradigm, learning representations of a single level, and are evaluated on the action recognition task solely, which leaves the generalization power of skeleton SSL models under-explored. To this end, a novel and effective SSL method for skeleton is further proposed, which integrates versatile representation learning objectives of different granularity, substantially boosting the generalization capacity for multiple skeleton downstream tasks. Extensive experiments under three large-scale datasets demonstrate our method achieves superior generalization performance on various downstream tasks, including recognition, retrieval, detection, and few-shot learning.
Authors: Chao Wang, Krzysztof Wolski, Bernhard Kerbl, Ana Serrano, Mojtaba Bemana, Hans-Peter Seidel, Karol Myszkowski, Thomas Leimk\"uhler
Abstract: Radiance field methods represent the state of the art in reconstructing complex scenes from multi-view photos. However, these reconstructions often suffer from one or both of the following limitations: First, they typically represent scenes in low dynamic range (LDR), which restricts their use to evenly lit environments and hinders immersive viewing experiences. Secondly, their reliance on a pinhole camera model, assuming all scene elements are in focus in the input images, presents practical challenges and complicates refocusing during novel-view synthesis. Addressing these limitations, we present a lightweight method based on 3D Gaussian Splatting that utilizes multi-view LDR images of a scene with varying exposure times, apertures, and focus distances as input to reconstruct a high-dynamic-range (HDR) radiance field. By incorporating analytical convolutions of Gaussians based on a thin-lens camera model as well as a tonemapping module, our reconstructions enable the rendering of HDR content with flexible refocusing capabilities. We demonstrate that our combined treatment of HDR and depth of field facilitates real-time cinematic rendering, outperforming the state of the art.
Authors: Yunsong Wang, Na Zhao, Gim Hee Lee
Abstract: The use of synthetic data in indoor 3D object detection offers the potential of greatly reducing the manual labor involved in 3D annotations and training effective zero-shot detectors. However, the complicated domain shifts across syn-to-real indoor datasets remains underexplored. In this paper, we propose a novel Object-wise Hierarchical Domain Alignment (OHDA) framework for syn-to-real unsupervised domain adaptation in indoor 3D object detection. Our approach includes an object-aware augmentation strategy to effectively diversify the source domain data, and we introduce a two-branch adaptation framework consisting of an adversarial training branch and a pseudo labeling branch, in order to simultaneously reach holistic-level and class-level domain alignment. The pseudo labeling is further refined through two proposed schemes specifically designed for indoor UDA. Our adaptation results from synthetic dataset 3D-FRONT to real-world datasets ScanNetV2 and SUN RGB-D demonstrate remarkable mAP25 improvements of 9.7% and 9.1% over Source-Only baselines, respectively, and consistently outperform the methods adapted from 2D and 3D outdoor scenarios. The code will be publicly available upon paper acceptance.
Authors: Yijia Guo, Liwen Hu, Lei Ma, Tiejun Huang
Abstract: 3D Gaussian Splatting (3DGS) demonstrates unparalleled superior performance in 3D scene reconstruction. However, 3DGS heavily relies on the sharp images. Fulfilling this requirement can be challenging in real-world scenarios especially when the camera moves fast, which severely limits the application of 3DGS. To address these challenges, we proposed Spike Gausian Splatting (SpikeGS), the first framework that integrates the spike streams into 3DGS pipeline to reconstruct 3D scenes via a fast-moving bio-inspired camera. With accumulation rasterization, interval supervision, and a specially designed pipeline, SpikeGS extracts detailed geometry and texture from high temporal resolution but texture lacking spike stream, reconstructs 3D scenes captured in 1 second. Extensive experiments on multiple synthetic and real-world datasets demonstrate the superiority of SpikeGS compared with existing spike-based and deblur 3D scene reconstruction methods. Codes and data will be released soon.
Authors: Prarthana Bhattacharyya, Joshua Mitton, Ryan Page, Owen Morgan, Ben Menzies, Gabriel Homewood, Kemi Jacobs, Paolo Baesso, David Trickett, Chris Mair, Taru Muhonen, Rory Clark, Louis Berridge, Richard Vigars, Iain Wallace
Abstract: This paper introduces Helios, the first extremely low-power, real-time, event-based hand gesture recognition system designed for all-day on smart eyewear. As augmented reality (AR) evolves, current smart glasses like the Meta Ray-Bans prioritize visual and wearable comfort at the expense of functionality. Existing human-machine interfaces (HMIs) in these devices, such as capacitive touch and voice controls, present limitations in ergonomics, privacy and power consumption. Helios addresses these challenges by leveraging natural hand interactions for a more intuitive and comfortable user experience. Our system utilizes a extremely low-power and compact 3mmx4mm/20mW event camera to perform natural hand-based gesture recognition for always-on smart eyewear. The camera's output is processed by a convolutional neural network (CNN) running on a NXP Nano UltraLite compute platform, consuming less than 350mW. Helios can recognize seven classes of gestures, including subtle microgestures like swipes and pinches, with 91% accuracy. We also demonstrate real-time performance across 20 users at a remarkably low latency of 60ms. Our user testing results align with the positive feedback we received during our recent successful demo at AWE-USA-2024.
Authors: Zirui Zhou, Junhao Liang, Zizhao Peng, Chao Fan, Fengwei An, Shiqi Yu
Abstract: Scoliosis presents significant diagnostic challenges, particularly in adolescents, where early detection is crucial for effective treatment. Traditional diagnostic and follow-up methods, which rely on physical examinations and radiography, face limitations due to the need for clinical expertise and the risk of radiation exposure, thus restricting their use for widespread early screening. In response, we introduce a novel video-based, non-invasive method for scoliosis classification using gait analysis, effectively circumventing these limitations. This study presents Scoliosis1K, the first large-scale dataset specifically designed for video-based scoliosis classification, encompassing over one thousand adolescents. Leveraging this dataset, we developed ScoNet, an initial model that faced challenges in handling the complexities of real-world data. This led to the development of ScoNet-MT, an enhanced model incorporating multi-task learning, which demonstrates promising diagnostic accuracy for practical applications. Our findings demonstrate that gait can serve as a non-invasive biomarker for scoliosis, revolutionizing screening practices through deep learning and setting a precedent for non-invasive diagnostic methodologies. The dataset and code are publicly available at https://zhouzi180.github.io/Scoliosis1K/.
Authors: Li Li, Hubert P. H. Shum, Toby P. Breckon
Abstract: 3D point clouds play a pivotal role in outdoor scene perception, especially in the context of autonomous driving. Recent advancements in 3D LiDAR segmentation often focus intensely on the spatial positioning and distribution of points for accurate segmentation. However, these methods, while robust in variable conditions, encounter challenges due to sole reliance on coordinates and point intensity, leading to poor isometric invariance and suboptimal segmentation. To tackle this challenge, our work introduces Range-Aware Pointwise Distance Distribution (RAPiD) features and the associated RAPiD-Seg architecture. Our RAPiD features exhibit rigid transformation invariance and effectively adapt to variations in point density, with a design focus on capturing the localized geometry of neighboring structures. They utilize inherent LiDAR isotropic radiation and semantic categorization for enhanced local representation and computational efficiency, while incorporating a 4D distance metric that integrates geometric and surface material reflectivity for improved semantic segmentation. To effectively embed high-dimensional RAPiD features, we propose a double-nested autoencoder structure with a novel class-aware embedding objective to encode high-dimensional features into manageable voxel-wise embeddings. Additionally, we propose RAPiD-Seg which incorporates a channel-wise attention fusion and two effective RAPiD-Seg variants, further optimizing the embedding for enhanced performance and generalization. Our method outperforms contemporary LiDAR segmentation work in terms of mIoU on SemanticKITTI (76.1) and nuScenes (83.6) datasets.
Authors: Xu Zheng, Yuanhuiyi Lyu, Jiazhou Zhou, Lin Wang
Abstract: Fusing an arbitrary number of modalities is vital for achieving robust multi-modal fusion of semantic segmentation yet remains less explored to date. Recent endeavors regard RGB modality as the center and the others as the auxiliary, yielding an asymmetric architecture with two branches. However, the RGB modality may struggle in certain circumstances, e.g., nighttime, while others, e.g., event data, own their merits; thus, it is imperative for the fusion model to discern robust and fragile modalities, and incorporate the most robust and fragile ones to learn a resilient multi-modal framework. To this end, we propose a novel method, named MAGIC, that can be flexibly paired with various backbones, ranging from compact to high-performance models. Our method comprises two key plug-and-play modules. Firstly, we introduce a multi-modal aggregation module to efficiently process features from multi-modal batches and extract complementary scene information. On top, a unified arbitrary-modal selection module is proposed to utilize the aggregated features as the benchmark to rank the multi-modal features based on the similarity scores. This way, our method can eliminate the dependence on RGB modality and better overcome sensor failures while ensuring the segmentation performance. Under the commonly considered multi-modal setting, our method achieves state-of-the-art performance while reducing the model parameters by 60%. Moreover, our method is superior in the novel modality-agnostic setting, where it outperforms prior arts by a large margin of +19.41% mIoU
Authors: Zeyu Wang, Jingyu Lin, Yifei Qian, Yi Huang, Shicen Tian, Bosong Chai, Juncan Deng, Qu Yang, Lan Du, Cunjian Chen, Yufei Guo, Kejie Huang
Abstract: Diffusion models have made significant strides in language-driven and layout-driven image generation. However, most diffusion models are limited to visible RGB image generation. In fact, human perception of the world is enriched by diverse viewpoints, such as chromatic contrast, thermal illumination, and depth information. In this paper, we introduce a novel diffusion model for general layout-guided cross-modal generation, called DiffX. Notably, our DiffX presents a simple yet effective cross-modal generative modeling pipeline, which conducts diffusion and denoising processes in the modality-shared latent space. Moreover, we introduce the Joint-Modality Embedder (JME) to enhance the interaction between layout and text conditions by incorporating a gated attention mechanism. To facilitate the user-instructed training, we construct the cross-modal image datasets with detailed text captions by the Large-Multimodal Model (LMM) and our human-in-the-loop refinement. Through extensive experiments, our DiffX demonstrates robustness in cross-modal ''RGB+X'' image generation on FLIR, MFNet, and COME15K datasets, guided by various layout conditions. It also shows the potential for the adaptive generation of ''RGB+X+Y(+Z)'' images or more diverse modalities on COME15K and MCXFace datasets. Our code and constructed cross-modal image datasets are available at https://github.com/zeyuwang-zju/DiffX.
Authors: Carolin Flosdorf, Justin Engelker, Igor Keller, Nicolas Mohr
Abstract: Skin cancer detection still represents a major challenge in healthcare. Common detection methods can be lengthy and require human assistance which falls short in many countries. Previous research demonstrates how convolutional neural networks (CNNs) can help effectively through both automation and an accuracy that is comparable to the human level. However, despite the progress in previous decades, the precision is still limited, leading to substantial misclassifications that have a serious impact on people's health. Hence, we employ a Vision Transformer (ViT) that has been developed in recent years based on the idea of a self-attention mechanism, specifically two configurations of a pre-trained ViT. We generally find superior metrics for classifying skin lesions after comparing them to base models such as decision tree classifier and k-nearest neighbor (KNN) classifier, as well as to CNNs and less complex ViTs. In particular, we attach greater importance to the performance of melanoma, which is the most lethal type of skin cancer. The ViT-L32 model achieves an accuracy of 91.57% and a melanoma recall of 58.54%, while ViT-L16 achieves an accuracy of 92.79% and a melanoma recall of 56.10%. This offers a potential tool for faster and more accurate diagnoses and an overall improvement for the healthcare sector.
Authors: Wenxuan Wang, Quan Sun, Fan Zhang, Yepeng Tang, Jing Liu, Xinlong Wang
Abstract: Contrastive Language-Image Pre-training (CLIP), which excels at abstracting open-world representations across domains and modalities, has become a foundation for a variety of vision and multimodal tasks. However, recent studies reveal that CLIP has severe visual shortcomings, such as which can hardly distinguish orientation, quantity, color, structure, etc. These visual shortcomings also limit the perception capabilities of multimodal large language models (MLLMs) built on CLIP. The main reason could be that the image-text pairs used to train CLIP are inherently biased, due to the lack of the distinctiveness of the text and the diversity of images. In this work, we present a simple post-training approach for CLIP models, which largely overcomes its visual shortcomings via a self-supervised diffusion process. We introduce DIVA, which uses the DIffusion model as a Visual Assistant for CLIP. Specifically, DIVA leverages generative feedback from text-to-image diffusion models to optimize CLIP representations, with only images (without corresponding text). We demonstrate that DIVA improves CLIP's performance on the challenging MMVP-VLM benchmark which assesses fine-grained visual abilities to a large extent (e.g., 3-7%), and enhances the performance of MLLMs and vision models on multimodal understanding and segmentation tasks. Extensive evaluation on 29 image classification and retrieval benchmarks confirms that our framework preserves CLIP's strong zero-shot capabilities. The code is available at https://github.com/baaivision/DIVA.
Authors: Yuxuan Bian, Ailing Zeng, Xuan Ju, Xian Liu, Zhaoyang Zhang, Wei Liu, Qiang Xu
Abstract: Whole-body multimodal motion generation, controlled by text, speech, or music, has numerous applications including video generation and character animation. However, employing a unified model to achieve various generation tasks with different condition modalities presents two main challenges: motion distribution drifts across different tasks (e.g., co-speech gestures and text-driven daily actions) and the complex optimization of mixed conditions with varying granularities (e.g., text and audio). Additionally, inconsistent motion formats across different tasks and datasets hinder effective training toward multimodal motion generation. In this paper, we propose MotionCraft, a unified diffusion transformer that crafts whole-body motion with plug-and-play multimodal control. Our framework employs a coarse-to-fine training strategy, starting with the first stage of text-to-motion semantic pre-training, followed by the second stage of multimodal low-level control adaptation to handle conditions of varying granularities. To effectively learn and transfer motion knowledge across different distributions, we design MC-Attn for parallel modeling of static and dynamic human topology graphs. To overcome the motion format inconsistency of existing benchmarks, we introduce MC-Bench, the first available multimodal whole-body motion generation benchmark based on the unified SMPL-X format. Extensive experiments show that MotionCraft achieves state-of-the-art performance on various standard motion generation tasks.
Authors: Muhammad Ahmad, Muhammad Hassaan Farooq Butt, Muhammad Usama, Hamad Ahmed Altuwaijri, Manuel Mazzara, Salvatore Distefano
Abstract: Spatial-Spectral Mamba (SSM) improves computational efficiency and captures long-range dependencies, addressing Transformer limitations. However, traditional Mamba models overlook rich spectral information in HSIs and struggle with high dimensionality and sequential data. To address these issues, we propose the SSM with multi-head self-attention and token enhancement (MHSSMamba). This model integrates spectral and spatial information by enhancing spectral tokens and using multi-head attention to capture complex relationships between spectral bands and spatial locations. It also manages long-range dependencies and the sequential nature of HSI data, preserving contextual information across spectral bands. MHSSMamba achieved remarkable classification accuracies of 97.62\% on Pavia University, 96.92\% on the University of Houston, 96.85\% on Salinas, and 99.49\% on Wuhan-longKou datasets. The source code is available at \href{https://github.com/MHassaanButt/MHA\_SS\_Mamba}{GitHub}.
Authors: Xin-Yi Yu, Jun-Xin Yu, Li-Bo Zhou, Yan Wei, Lin-Lin Ou
Abstract: We present InstantStyleGaussian, an innovative 3D style transfer method based on the 3D Gaussian Splatting (3DGS) scene representation. By inputting a target-style image, it quickly generates new 3D GS scenes. Our method operates on pre-reconstructed GS scenes, combining diffusion models with an improved iterative dataset update strategy. It utilizes diffusion models to generate target style images, adds these new images to the training dataset, and uses this dataset to iteratively update and optimize the GS scenes, significantly accelerating the style editing process while ensuring the quality of the generated scenes. Extensive experimental results demonstrate that our method ensures high-quality stylized scenes while offering significant advantages in style transfer speed and consistency.
Authors: Ragib Amin Nihal, Benjamin Yen, Katsutoshi Itoyama, Kazuhiro Nakadai
Abstract: Unmanned aerial vehicles (UAVs) have revolutionized search and rescue (SAR) operations, but the lack of specialized human detection datasets for training machine learning models poses a significant challenge.To address this gap, this paper introduces the Combination to Application (C2A) dataset, synthesized by overlaying human poses onto UAV-captured disaster scenes. Through extensive experimentation with state-of-the-art detection models, we demonstrate that models fine-tuned on the C2A dataset exhibit substantial performance improvements compared to those pre-trained on generic aerial datasets. Furthermore, we highlight the importance of combining the C2A dataset with general human datasets to achieve optimal performance and generalization across various scenarios. This points out the crucial need for a tailored dataset to enhance the effectiveness of SAR operations. Our contributions also include developing dataset creation pipeline and integrating diverse human poses and disaster scenes information to assess the severity of disaster scenarios. Our findings advocate for future developments, to ensure that SAR operations benefit from the most realistic and effective AI-assisted interventions possible.
Authors: Yina He, Lei Peng, Yongcun Zhang, Juanjuan Weng, Zhiming Luo, Shaozi Li
Abstract: Current out-of-distribution (OOD) detection methods typically assume balanced in-distribution (ID) data, while most real-world data follow a long-tailed distribution. Previous approaches to long-tailed OOD detection often involve balancing the ID data by reducing the semantics of head classes. However, this reduction can severely affect the classification accuracy of ID data. The main challenge of this task lies in the severe lack of features for tail classes, leading to confusion with OOD data. To tackle this issue, we introduce a novel Prioritizing Attention to Tail (PATT) method using augmentation instead of reduction. Our main intuition involves using a mixture of von Mises-Fisher (vMF) distributions to model the ID data and a temperature scaling module to boost the confidence of ID data. This enables us to generate infinite contrastive pairs, implicitly enhancing the semantics of ID classes while promoting differentiation between ID and OOD data. To further strengthen the detection of OOD data without compromising the classification performance of ID data, we propose feature calibration during the inference phase. By extracting an attention weight from the training set that prioritizes the tail classes and reduces the confidence in OOD data, we improve the OOD detection capability. Extensive experiments verified that our method outperforms the current state-of-the-art methods on various benchmarks.
Authors: Juepeng Zheng, Yibin Wen, Jinxiao Zhang, Runmin Dong, Haohuan Fu
Abstract: In this paper, we firstly tackle a more realistic Domain Adaptation (DA) setting: Source-Free Blending-Target Domain Adaptation (SF-BTDA), where we can not access to source domain data while facing mixed multiple target domains without any domain labels in prior. Compared to existing DA scenarios, SF-BTDA generally faces the co-existence of different label shifts in different targets, along with noisy target pseudo labels generated from the source model. In this paper, we propose a new method called Evidential Contrastive Alignment (ECA) to decouple the blending target domain and alleviate the effect from noisy target pseudo labels. First, to improve the quality of pseudo target labels, we propose a calibrated evidential learning module to iteratively improve both the accuracy and certainty of the resulting model and adaptively generate high-quality pseudo target labels. Second, we design a graph contrastive learning with the domain distance matrix and confidence-uncertainty criterion, to minimize the distribution gap of samples of a same class in the blended target domains, which alleviates the co-existence of different label shifts in blended targets. We conduct a new benchmark based on three standard DA datasets and ECA outperforms other methods with considerable gains and achieves comparable results compared with those that have domain labels or source data in prior.
Authors: Qiushuo Cheng, Catherine Morgan, Arindam Sikdar, Alessandro Masullo, Alan Whone, Majid Mirmehdi
Abstract: People with Parkinson's Disease (PD) often experience progressively worsening gait, including changes in how they turn around, as the disease progresses. Existing clinical rating tools are not capable of capturing hour-by-hour variations of PD symptoms, as they are confined to brief assessments within clinic settings. Measuring gait turning angles continuously and passively is a component step towards using gait characteristics as sensitive indicators of disease progression in PD. This paper presents a deep learning-based approach to automatically quantify turning angles by extracting 3D skeletons from videos and calculating the rotation of hip and knee joints. We utilise state-of-the-art human pose estimation models, Fastpose and Strided Transformer, on a total of 1386 turning video clips from 24 subjects (12 people with PD and 12 healthy control volunteers), trimmed from a PD dataset of unscripted free-living videos in a home-like setting (Turn-REMAP). We also curate a turning video dataset, Turn-H3.6M, from the public Human3.6M human pose benchmark with 3D ground truth, to further validate our method. Previous gait research has primarily taken place in clinics or laboratories evaluating scripted gait outcomes, but this work focuses on free-living home settings where complexities exist, such as baggy clothing and poor lighting. Due to difficulties in obtaining accurate ground truth data in a free-living setting, we quantise the angle into the nearest bin $45^\circ$ based on the manual labelling of expert clinicians. Our method achieves a turning calculation accuracy of 41.6%, a Mean Absolute Error (MAE) of 34.7{\deg}, and a weighted precision WPrec of 68.3% for Turn-REMAP. This is the first work to explore the use of single monocular camera data to quantify turns by PD patients in a home setting.
Authors: Tianyi Xu, Yiji Zhou, Xiaotao Hu, Kai Zhang, Anran Zhang, Xingye Qiu, Jun Xu
Abstract: Arbitrary-scale super-resolution (ASSR) aims to learn a single model for image super-resolution at arbitrary magnifying scales. Existing ASSR networks typically comprise an off-the-shelf scale-agnostic feature extractor and an arbitrary scale upsampler. These feature extractors often use fixed network architectures to address different ASSR inference tasks, each of which is characterized by an input image and an upsampling scale. However, this overlooks the difficulty variance of super-resolution on different inference scenarios, where simple images or small SR scales could be resolved with less computational effort than difficult images or large SR scales. To tackle this difficulty variability, in this paper, we propose a Task-Aware Dynamic Transformer (TADT) as an input-adaptive feature extractor for efficient image ASSR. Our TADT consists of a multi-scale feature extraction backbone built upon groups of Multi-Scale Transformer Blocks (MSTBs) and a Task-Aware Routing Controller (TARC). The TARC predicts the inference paths within feature extraction backbone, specifically selecting MSTBs based on the input images and SR scales. The prediction of inference path is guided by a new loss function to trade-off the SR accuracy and efficiency. Experiments demonstrate that, when working with three popular arbitrary-scale upsamplers, our TADT achieves state-of-the-art ASSR performance when compared with mainstream feature extractors, but with relatively fewer computational costs. The code will be publicly released.
Authors: Feiyu Pan, Hao Fang, Runmin Cong, Wei Zhang, Xiankai Lu
Abstract: Video Object Segmentation (VOS) task aims to segmenting a particular object instance throughout the entire video sequence given only the object mask of the first frame. Recently, Segment Anything Model 2 (SAM 2) is proposed, which is a foundation model towards solving promptable visual segmentation in images and videos. SAM 2 builds a data engine, which improves model and data via user interaction, to collect the largest video segmentation dataset to date. SAM 2 is a simple transformer architecture with streaming memory for real-time video processing, which trained on the date provides strong performance across a wide range of tasks. In this work, we evaluate the zero-shot performance of SAM 2 on the more challenging VOS datasets MOSE and LVOS. Without fine-tuning on the training set, SAM 2 achieved 75.79 J&F on the test set and ranked 4th place for 6th LSVOS Challenge VOS Track.
Authors: Hao Fang, Feiyu Pan, Xiankai Lu, Wei Zhang, Runmin Cong
Abstract: Referring video object segmentation (RVOS) relies on natural language expressions to segment target objects in video. In this year, LSVOS Challenge RVOS Track replaced the origin YouTube-RVOS benchmark with MeViS. MeViS focuses on referring the target object in a video through its motion descriptions instead of static attributes, posing a greater challenge to RVOS task. In this work, we integrate strengths of that leading RVOS and VOS models to build up a simple and effective pipeline for RVOS. Firstly, We finetune the state-of-the-art RVOS model to obtain mask sequences that are correlated with language descriptions. Secondly, based on a reliable and high-quality key frames, we leverage VOS model to enhance the quality and temporal consistency of the mask results. Finally, we further improve the performance of the RVOS model using semi-supervised learning. Our solution achieved 62.57 J&F on the MeViS test set and ranked 1st place for 6th LSVOS Challenge RVOS Track.
Authors: Yihao Huang, Le Liang, Tianlin Li, Xiaojun Jia, Run Wang, Weikai Miao, Geguang Pu, Yang Liu
Abstract: In recent years, Text-to-Image (T2I) models have garnered significant attention due to their remarkable advancements. However, security concerns have emerged due to their potential to generate inappropriate or Not-Safe-For-Work (NSFW) images. In this paper, inspired by the observation that texts with different semantics can lead to similar human perceptions, we propose an LLM-driven perception-guided jailbreak method, termed PGJ. It is a black-box jailbreak method that requires no specific T2I model (model-free) and generates highly natural attack prompts. Specifically, we propose identifying a safe phrase that is similar in human perception yet inconsistent in text semantics with the target unsafe word and using it as a substitution. The experiments conducted on six open-source models and commercial online services with thousands of prompts have verified the effectiveness of PGJ.
Authors: Javid Akhavan, Chaitanya Krishna Vallabh, Xiayun Zhao, Souran Manoochehri
Abstract: In metal Additive Manufacturing (AM), monitoring the temperature of the Melt Pool (MP) is crucial for ensuring part quality, process stability, defect prevention, and overall process optimization. Traditional methods, are slow to converge and require extensive manual effort to translate data into actionable insights, rendering them impractical for real-time monitoring and control. To address this challenge, we propose an Artificial Intelligence (AI)-based solution aimed at reducing manual data processing reliance and improving the efficiency of transitioning from data to insight. In our study, we utilize a dataset comprising dual-wavelength real-time process monitoring data and corresponding temperature maps. We introduce a deep learning model called the "Binocular model," which exploits dual input observations to perform a precise analysis of MP temperature in Laser Powder Bed Fusion (L-PBF). Through advanced deep learning techniques, we seamlessly convert raw data into temperature maps, significantly streamlining the process and enabling batch processing at a rate of up to 750 frames per second, approximately 1000 times faster than conventional methods. Our Binocular model achieves high accuracy in temperature estimation, evidenced by a 0.95 R-squared score, while simultaneously enhancing processing efficiency by a factor of $\sim1000x$ times. This model directly addresses the challenge of real-time MP temperature monitoring and offers insights into the encountered constraints and the benefits of our Deep Learning-based approach. By combining efficiency and precision, our work contributes to the advancement of temperature monitoring in L-PBF, thus driving progress in the field of metal AM.
Authors: Enze Zhu, Zhan Chen, Dingkai Wang, Hanru Shi, Xiaoxuan Liu, Lei Wang
Abstract: Semantic segmentation of high-resolution remote sensing images is vital in downstream applications such as land-cover mapping, urban planning and disaster assessment.Existing Transformer-based methods suffer from the constraint between accuracy and efficiency, while the recently proposed Mamba is renowned for being efficient. Therefore, to overcome the dilemma, we propose UNetMamba, a UNet-like semantic segmentation model based on Mamba. It incorporates a mamba segmentation decoder (MSD) that can efficiently decode the complex information within high-resolution images, and a local supervision module (LSM), which is train-only but can significantly enhance the perception of local contents. Extensive experiments demonstrate that UNetMamba outperforms the state-of-the-art methods with mIoU increased by 0.87% on LoveDA and 0.36% on ISPRS Vaihingen, while achieving high efficiency through the lightweight design, less memory footprint and reduced computational cost. The source code is available at https://github.com/EnzeZhu2001/UNetMamba.
Authors: Ziming Liu, Jingcai Guo, Song Guo, Xiaocheng Lu
Abstract: This paper investigates a challenging problem of zero-shot learning in the multi-label scenario (MLZSL), wherein the model is trained to recognize multiple unseen classes within a sample (e.g., an image) based on seen classes and auxiliary knowledge, e.g., semantic information. Existing methods usually resort to analyzing the relationship of various seen classes residing in a sample from the dimension of spatial or semantic characteristics and transferring the learned model to unseen ones. However, they neglect the integrity of local and global features. Although the use of the attention structure will accurately locate local features, especially objects, it will significantly lose its integrity, and the relationship between classes will also be affected. Rough processing of global features will also directly affect comprehensiveness. This neglect will make the model lose its grasp of the main components of the image. Relying only on the local existence of seen classes during the inference stage introduces unavoidable bias. In this paper, we propose a novel and comprehensive visual-semantic framework for MLZSL, dubbed Epsilon, to fully make use of such properties and enable a more accurate and robust visual-semantic projection. In terms of spatial information, we achieve effective refinement by group aggregating image features into several semantic prompts. It can aggregate semantic information rather than class information, preserving the correlation between semantics. In terms of global semantics, we use global forward propagation to collect as much information as possible to ensure that semantics are not omitted. Experiments on large-scale MLZSL benchmark datasets NUS-Wide and Open-Images-v4 demonstrate that the proposed Epsilon outperforms other state-of-the-art methods with large margins.
Authors: Zhiqiang Wu, Licheng Sun, Yingjie Liu, Jian Yang, Hanlin Dong, Shing-Ho J. Lin, Xuan Tang, Jinpeng Mi, Bo Jin, Xian Wei
Abstract: Group Equivariant Convolution (GConv) can effectively handle rotational symmetry data. They assume uniform and strict rotational symmetry across all features, as the transformations under the specific group. However, real-world data rarely conforms to strict rotational symmetry commonly referred to as Rotational Symmetry-Breaking in the system or dataset, making GConv unable to adapt effectively to this phenomenon. Motivated by this, we propose a simple but highly effective method to address this problem, which utilizes a set of learnable biases called the $G$-Biases under the group order to break strict group constraints and achieve \textbf{R}elaxed \textbf{R}otational \textbf{E}quivarant \textbf{Conv}olution (RREConv). We conduct extensive experiments to validate Relaxed Rotational Equivariance on rotational symmetry groups $\mathcal{C}_n$ (e.g. $\mathcal{C}_2$, $\mathcal{C}_4$, and $\mathcal{C}_6$ groups). Further experiments demonstrate that our proposed RREConv-based methods achieve excellent performance, compared to existing GConv-based methods in classification and detection tasks on natural image datasets.
Authors: Jinheng Xie, Weijia Mao, Zechen Bai, David Junhao Zhang, Weihao Wang, Kevin Qinghong Lin, Yuchao Gu, Zhijie Chen, Zhenheng Yang, Mike Zheng Shou
Abstract: We present a unified transformer, i.e., Show-o, that unifies multimodal understanding and generation. Unlike fully autoregressive models, Show-o unifies autoregressive and (discrete) diffusion modeling to adaptively handle inputs and outputs of various and mixed modalities. The unified model flexibly supports a wide range of vision-language tasks including visual question-answering, text-to-image generation, text-guided inpainting/extrapolation, and mixed-modality generation. Across various benchmarks, it demonstrates comparable or superior performance to existing individual models with an equivalent or larger number of parameters tailored for understanding or generation. This significantly highlights its potential as a next-generation foundation model. Code and models are released at https://github.com/showlab/Show-o.
Authors: Rawal Khirodkar, Timur Bagautdinov, Julieta Martinez, Su Zhaoen, Austin James, Peter Selednik, Stuart Anderson, Shunsuke Saito
Abstract: We present Sapiens, a family of models for four fundamental human-centric vision tasks - 2D pose estimation, body-part segmentation, depth estimation, and surface normal prediction. Our models natively support 1K high-resolution inference and are extremely easy to adapt for individual tasks by simply fine-tuning models pretrained on over 300 million in-the-wild human images. We observe that, given the same computational budget, self-supervised pretraining on a curated dataset of human images significantly boosts the performance for a diverse set of human-centric tasks. The resulting models exhibit remarkable generalization to in-the-wild data, even when labeled data is scarce or entirely synthetic. Our simple model design also brings scalability - model performance across tasks improves as we scale the number of parameters from 0.3 to 2 billion. Sapiens consistently surpasses existing baselines across various human-centric benchmarks. We achieve significant improvements over the prior state-of-the-art on Humans-5K (pose) by 7.6 mAP, Humans-2K (part-seg) by 17.1 mIoU, Hi4D (depth) by 22.4% relative RMSE, and THuman2 (normal) by 53.5% relative angular error. Project page: $\href{https://about.meta.com/realitylabs/codecavatars/sapiens}{\text{this https URL}}$.
URLs: https://about.meta.com/realitylabs/codecavatars/sapiens
Authors: Weilong Chen, Wenxuan Xu, Haoran Chen, Xinran Zhang, Zhijin Qin, Yanru Zhang, Zhu Han
Abstract: Underwater communication is essential for environmental monitoring, marine biology research, and underwater exploration. Traditional underwater communication faces limitations like low bandwidth, high latency, and susceptibility to noise, while semantic communication (SC) offers a promising solution by focusing on the exchange of semantics rather than symbols or bits. However, SC encounters challenges in underwater environments, including semantic information mismatch and difficulties in accurately identifying and transmitting critical information that aligns with the diverse requirements of underwater applications. To address these challenges, we propose a novel Semantic Communication (SC) framework based on Large Language Models (LLMs). Our framework leverages visual LLMs to perform semantic compression and prioritization of underwater image data according to the query from users. By identifying and encoding key semantic elements within the images, the system selectively transmits high-priority information while applying higher compression rates to less critical regions. On the receiver side, an LLM-based recovery mechanism, along with Global Vision ControlNet and Key Region ControlNet networks, aids in reconstructing the images, thereby enhancing communication efficiency and robustness. Our framework reduces the overall data size to 0.8\% of the original. Experimental results demonstrate that our method significantly outperforms existing approaches, ensuring high-quality, semantically accurate image reconstruction.
Authors: Jiaxin Wei, Stefan Leutenegger
Abstract: Traditional volumetric fusion algorithms preserve the spatial structure of 3D scenes, which is beneficial for many tasks in computer vision and robotics. However, they often lack realism in terms of visualization. Emerging 3D Gaussian splatting bridges this gap, but existing Gaussian-based reconstruction methods often suffer from artifacts and inconsistencies with the underlying 3D structure, and struggle with real-time optimization, unable to provide users with immediate feedback in high quality. One of the bottlenecks arises from the massive amount of Gaussian parameters that need to be updated during optimization. Instead of using 3D Gaussian as a standalone map representation, we incorporate it into a volumetric mapping system to take advantage of geometric information and propose to use a quadtree data structure on images to drastically reduce the number of splats initialized. In this way, we simultaneously generate a compact 3D Gaussian map with fewer artifacts and a volumetric map on the fly. Our method, GSFusion, significantly enhances computational efficiency without sacrificing rendering quality, as demonstrated on both synthetic and real datasets. Code will be available at https://github.com/goldoak/GSFusion.
Authors: Bonan Li, Zicheng Zhang, Xingyi Yang, Xinchao Wang
Abstract: Generating dense multiview images from text prompts is crucial for creating high-fidelity 3D assets. Nevertheless, existing methods struggle with space-view correspondences, resulting in sparse and low-quality outputs. In this paper, we introduce CoSER, a novel consistent dense Multiview Text-to-Image Generator for Text-to-3D, achieving both efficiency and quality by meticulously learning neighbor-view coherence and further alleviating ambiguity through the swift traversal of all views. For achieving neighbor-view consistency, each viewpoint densely interacts with adjacent viewpoints to perceive the global spatial structure, and aggregates information along motion paths explicitly defined by physical principles to refine details. To further enhance cross-view consistency and alleviate content drift, CoSER rapidly scan all views in spiral bidirectional manner to aware holistic information and then scores each point based on semantic material. Subsequently, we conduct weighted down-sampling along the spatial dimension based on scores, thereby facilitating prominent information fusion across all views with lightweight computation. Technically, the core module is built by integrating the attention mechanism with a selective state space model, exploiting the robust learning capabilities of the former and the low overhead of the latter. Extensive evaluation shows that CoSER is capable of producing dense, high-fidelity, content-consistent multiview images that can be flexibly integrated into various 3D generation models.
Authors: Gwenol\'e Quellec, Mathieu Lamard, Bruno Lay, Alexandre Le Guilcher, Ali Erginay, B\'eatrice Cochener, Pascale Massin
Abstract: The purpose of this study is to evaluate the performance of the OphtAI system for the automatic detection of referable diabetic retinopathy (DR) and the automatic assessment of DR severity using color fundus photography. OphtAI relies on ensembles of convolutional neural networks trained to recognize eye laterality, detect referable DR and assess DR severity. The system can either process single images or full examination records. To document the automatic diagnoses, accurate heatmaps are generated. The system was developed and validated using a dataset of 763,848 images from 164,660 screening procedures from the OPHDIAT screening program. For comparison purposes, it was also evaluated in the public Messidor-2 dataset. Referable DR can be detected with an area under the ROC curve of AUC = 0.989 in the Messidor-2 dataset, using the University of Iowa's reference standard (95% CI: 0.984-0.994). This is better than the only AI system authorized by the FDA, evaluated in the exact same conditions (AUC = 0.980). OphtAI can also detect vision-threatening DR with an AUC of 0.997 (95% CI: 0.996-0.998) and proliferative DR with an AUC of 0.997 (95% CI: 0.995-0.999). The system runs in 0.3 seconds using a graphics processing unit and less than 2 seconds without. OphtAI is safer, faster and more comprehensive than the only AI system authorized by the FDA so far. Instant DR diagnosis is now possible, which is expected to streamline DR screening and to give easy access to DR screening to more diabetic patients.
Authors: Prashant Rajaram, Puneet Manchanda
Abstract: Influencer marketing videos have surged in popularity, yet significant gaps remain in understanding the relationship between video features and engagement. This challenge is intensified by the complexities of interpreting unstructured data. While deep learning models effectively leverage unstructured data to predict business outcomes, they often function as black boxes with limited interpretability, particularly when human validation is hindered by the absence of a known ground truth. To address this issue, the authors develop an "interpretable deep learning framework" that not only makes good out-of-sample predictions using unstructured data but also provides insights into the captured relationships. Inspired by visual attention in print advertising, the interpretation approach uses measures of model attention to video features, eliminating spurious associations through a two-step process and shortlisting relationships for formal causal testing. This method is applicable across well-known attention mechanisms - additive attention, scaled dot-product attention, and gradient-based attention - when analyzing text, audio, or video image data. Validated using simulations, this approach outperforms benchmark feature selection methods. This framework is applied to YouTube influencer videos, linking video features to measures of shallow and deep engagement developed based on the dual-system framework of thinking. The findings guide influencers and brands in prioritizing video features associated with deep engagement.
Authors: Hideaki Okamoto, Quan Huu Cap, Takakiyo Nomura, Kazuhito Nabeshima, Jun Hashimoto, Hitoshi Iyatomi
Abstract: Endoscopy is widely used to diagnose gastric cancer and has a high diagnostic performance, but it must be performed by a physician, which limits the number of people who can be diagnosed. In contrast, gastric X-rays can be performed by technicians and screen a much larger number of patients, but accurate diagnosis requires experience. We propose an unprecedented and practical gastric cancer diagnosis support system for gastric X-ray images, enabling more people to be screened. The system is based on a general deep learning-based object detection model and incorporates two novel techniques: refined probabilistic stomach image augmentation (R-sGAIA) and hard boundary box training (HBBT). R-sGAIA enhances the probabilistic gastric fold region, providing more learning patterns for cancer detection models. HBBT is an efficient training method that improves model performance by allowing the use of unannotated negative (i.e., healthy control) samples, which are typically unusable in conventional detection models. The proposed system achieves a sensitivity (SE) for gastric cancer of 90.2%, higher than that of an expert (85.5%). Additionally, two out of five detected candidate boxes are cancerous, maintaining high precision while processing images at a speed of 0.51 seconds per image. The system also outperforms methods using the same object detection model and state-of-the-art data augmentation, showing a 5.9-point improvement in the F1 score. In summary, this system efficiently identifies areas for radiologists to examine within a practical timeframe, significantly reducing their workload.
Authors: Zongliang Wu, Ruiying Lu, Ying Fu, Xin Yuan
Abstract: Snapshot compressive spectral imaging reconstruction aims to reconstruct three-dimensional spatial-spectral images from a single-shot two-dimensional compressed measurement. Existing state-of-the-art methods are mostly based on deep unfolding structures but have intrinsic performance bottlenecks: $i$) the ill-posed problem of dealing with heavily degraded measurement, and $ii$) the regression loss-based reconstruction models being prone to recover images with few details. In this paper, we introduce a generative model, namely the latent diffusion model (LDM), to generate degradation-free prior to enhance the regression-based deep unfolding method. Furthermore, to overcome the large computational cost challenge in LDM, we propose a lightweight model to generate knowledge priors in deep unfolding denoiser, and integrate these priors to guide the reconstruction process for compensating high-quality spectral signal details. Numeric and visual comparisons on synthetic and real-world datasets illustrate the superiority of our proposed method in both reconstruction quality and computational efficiency. Code will be released.
Authors: Paul K. Mandal
Abstract: In the era of rapidly advancing medical technologies, the segmentation of medical data has become inevitable, necessitating the development of privacy preserving machine learning algorithms that can train on distributed data. Consolidating sensitive medical data is not always an option particularly due to the stringent privacy regulations imposed by the Health Insurance Portability and Accountability Act (HIPAA). In this paper, I introduce a HIPAA compliant framework that can train from distributed data. I then propose a multimodal vertical federated model for Alzheimer's Disease (AD) detection, a serious neurodegenerative condition that can cause dementia, severely impairing brain function and hindering simple tasks, especially without preventative care. This vertical federated learning (VFL) model offers a distributed architecture that enables collaborative learning across diverse sources of medical data while respecting privacy constraints imposed by HIPAA. The VFL architecture proposed herein offers a novel distributed architecture, enabling collaborative learning across diverse sources of medical data while respecting statutory privacy constraints. By leveraging multiple modalities of data, the robustness and accuracy of AD detection can be enhanced. This model not only contributes to the advancement of federated learning techniques but also holds promise for overcoming the hurdles posed by data segmentation in medical research.
Authors: Rizhong Lin, Ali Gholipour, Jean-Philippe Thiran, Davood Karimi, Hamza Kebiri, Meritxell Bach Cuadra
Abstract: Deep learning models have shown great promise in estimating tissue microstructure from limited diffusion magnetic resonance imaging data. However, these models face domain shift challenges when test and train data are from different scanners and protocols, or when the models are applied to data with inherent variations such as the developing brains of infants and children scanned at various ages. Several techniques have been proposed to address some of these challenges, such as data harmonization or domain adaptation in the adult brain. However, those techniques remain unexplored for the estimation of fiber orientation distribution functions in the rapidly developing brains of infants. In this work, we extensively investigate the age effect and domain shift within and across two different cohorts of 201 newborns and 165 babies using the Method of Moments and fine-tuning strategies. Our results show that reduced variations in the microstructural development of babies in comparison to newborns directly impact the deep learning models' cross-age performance. We also demonstrate that a small number of target domain samples can significantly mitigate domain shift problems.
Authors: Hadi M. Dolatabadi, Sarah M. Erfani, Christopher Leckie
Abstract: Deep neural networks (DNNs) are vulnerable to shortcut learning: rather than learning the intended task, they tend to draw inconclusive relationships between their inputs and outputs. Shortcut learning is ubiquitous among many failure cases of neural networks, and traces of this phenomenon can be seen in their generalizability issues, domain shift, adversarial vulnerability, and even bias towards majority groups. In this paper, we argue that this commonality in the cause of various DNN issues creates a significant opportunity that should be leveraged to find a unified solution for shortcut learning. To this end, we outline the recent advances in topological data analysis (TDA), and persistent homology (PH) in particular, to sketch a unified roadmap for detecting shortcuts in deep learning. We demonstrate our arguments by investigating the topological features of computational graphs in DNNs using two cases of unlearnable examples and bias in decision-making as our test studies. Our analysis of these two failure cases of DNNs reveals that finding a unified solution for shortcut learning in DNNs is not out of reach, and TDA can play a significant role in forming such a framework.
Authors: Yunxin Li, Xinyu Chen, Baotain Hu, Min Zhang
Abstract: Long video understanding is a significant and ongoing challenge in the intersection of multimedia and artificial intelligence. Employing large language models (LLMs) for comprehending video becomes an emerging and promising method. However, this approach incurs high computational costs due to the extensive array of video tokens, experiences reduced visual clarity as a consequence of token aggregation, and confronts challenges arising from irrelevant visual tokens while answering video-related questions. To alleviate these issues, we present an Interactive Visual Adapter (IVA) within LLMs, designed to enhance interaction with fine-grained visual elements. Specifically, we first transform long videos into temporal video tokens via leveraging a visual encoder alongside a pretrained causal transformer, then feed them into LLMs with the video instructions. Subsequently, we integrated IVA, which contains a lightweight temporal frame selector and a spatial feature interactor, within the internal blocks of LLMs to capture instruction-aware and fine-grained visual signals. Consequently, the proposed video-LLM facilitates a comprehensive understanding of long video content through appropriate long video modeling and precise visual interactions. We conducted extensive experiments on nine video understanding benchmarks and experimental results show that our interactive visual adapter significantly improves the performance of video LLMs on long video QA tasks. Ablation studies further verify the effectiveness of IVA in understanding long and short video.
Authors: Xiongri Shen, Zhenxi Song, Zhiguo Zhang
Abstract: Diagnosis of mild cognitive impairment (MCI) and subjective cognitive decline (SCD) from fMRI functional connectivity (FC) has gained popularity, but most FC-based diagnostic models are black boxes lacking casual reasoning so they contribute little to the knowledge about FC-based neural biomarkers of cognitive decline.To enhance the explainability of diagnostic models, we propose a generative counterfactual attention-guided network (GCAN), which introduces counterfactual reasoning to recognize cognitive decline-related brain regions and then uses these regions as attention maps to boost the prediction performance of diagnostic models. Furthermore, to tackle the difficulty in the generation of highly-structured and brain-atlas-constrained FC, which is essential in counterfactual reasoning, an Atlas-Aware Bidirectional Transformer (AABT) method is developed. AABT employs a bidirectional strategy to encode and decode the tokens from each network of brain atlas, thereby enhancing the generation of high-quality target label FC. In the experiments of hospital-collected and ADNI datasets, the generated attention maps closely resemble FC abnormalities in the literature on SCD and MCI. The diagnostic performance is also superior to baseline models. The code is available at https://github.com/SXR3015/GCAN
Authors: Boyuan Peng, Jiaju Chen, P. Bilha Githinji, Ijaz Gul, Qihui Ye, Minjiang Chen, Peiwu Qin, Xingru Huang, Chenggang Yan, Dongmei Yu, Jiansong Ji, Zhenglin Chen
Abstract: Cell segmentation is essential in biomedical research for analyzing cellular morphology and behavior. Deep learning methods, particularly convolutional neural networks (CNNs), have revolutionized cell segmentation by extracting intricate features from images. However, the robustness of these methods under microscope optical aberrations remains a critical challenge. This study evaluates cell image segmentation models under optical aberrations from fluorescence and bright field microscopy. By simulating different types of aberrations, including astigmatism, coma, spherical aberration, trefoil, and mixed aberrations, we conduct a thorough evaluation of various cell instance segmentation models using the DynamicNuclearNet (DNN) and LIVECell datasets, representing fluorescence and bright field microscopy cell datasets, respectively. We train and test several segmentation models, including the Otsu threshold method and Mask R-CNN with different network heads (FPN, C3) and backbones (ResNet, VGG, Swin Transformer), under aberrated conditions. Additionally, we provide usage recommendations for the Cellpose 2.0 Toolbox on complex cell degradation images. The results indicate that the combination of FPN and SwinS demonstrates superior robustness in handling simple cell images affected by minor aberrations. In contrast, Cellpose 2.0 proves effective for complex cell images under similar conditions. Furthermore, we innovatively propose the Point Spread Function Image Label Classification Model (PLCM). This model can quickly and accurately identify aberration types and amplitudes from PSF images, assisting researchers without optical training. Through PLCM, researchers can better apply our proposed cell segmentation guidelines.
Authors: James K Ruffle, Samia Mohinta, Kelly Pegoretti Baruteau, Rebekah Rajiah, Faith Lee, Sebastian Brandner, Parashkev Nachev, Harpreet Hyare
Abstract: The VASARI MRI feature set is a quantitative system designed to standardise glioma imaging descriptions. Though effective, deriving VASARI is time-consuming and seldom used in clinical practice. This is a problem that machine learning could plausibly automate. Using glioma data from 1172 patients, we developed VASARI-auto, an automated labelling software applied to both open-source lesion masks and our openly available tumour segmentation model. In parallel, two consultant neuroradiologists independently quantified VASARI features in a subsample of 100 glioblastoma cases. We quantified: 1) agreement across neuroradiologists and VASARI-auto; 2) calibration of performance equity; 3) an economic workforce analysis; and 4) fidelity in predicting patient survival. Tumour segmentation was compatible with the current state of the art and equally performant regardless of age or sex. A modest inter-rater variability between in-house neuroradiologists was comparable to between neuroradiologists and VASARI-auto, with far higher agreement between VASARI-auto methods. The time taken for neuroradiologists to derive VASARI was substantially higher than VASARI-auto (mean time per case 317 vs. 3 seconds). A UK hospital workforce analysis forecast that three years of VASARI featurisation would demand 29,777 consultant neuroradiologist workforce hours ({\pounds}1,574,935), reducible to 332 hours of computing time (and {\pounds}146 of power) with VASARI-auto. The best-performing survival model utilised VASARI-auto features as opposed to those derived by neuroradiologists. VASARI-auto is a highly efficient automated labelling system with equitable performance across patient age or sex, a favourable economic profile if used as a decision support tool, and with non-inferior fidelity in downstream patient survival prediction. Future work should iterate upon and integrate such tools to enhance patient care.
Authors: Jorge Tapias Gomez, Aneesh Rangnekar, Hannah Williams, Hannah Thompson, Julio Garcia-Aguilar, Joshua Jesse Smith, Harini Veeraraghavan
Abstract: Endoscopic images are used at various stages of rectal cancer treatment starting from cancer screening, diagnosis, during treatment to assess response and toxicity from treatments such as colitis, and at follow up to detect new tumor or local regrowth (LR). However, subjective assessment is highly variable and can underestimate the degree of response in some patients, subjecting them to unnecessary surgery, or overestimate response that places patients at risk of disease spread. Advances in deep learning has shown the ability to produce consistent and objective response assessment for endoscopic images. However, methods for detecting cancers, regrowth, and monitoring response during the entire course of patient treatment and follow-up are lacking. This is because, automated diagnosis and rectal cancer response assessment requires methods that are robust to inherent imaging illumination variations and confounding conditions (blood, scope, blurring) present in endoscopy images as well as changes to the normal lumen and tumor during treatment. Hence, a hierarchical shifted window (Swin) transformer was trained to distinguish rectal cancer from normal lumen using endoscopy images. Swin as well as two convolutional (ResNet-50, WideResNet-50), and vision transformer (ViT) models were trained and evaluated on follow-up longitudinal images to detect LR on private dataset as well as on out-of-distribution (OOD) public colonoscopy datasets to detect pre/non-cancerous polyps. Color shifts were applied using optimal transport to simulate distribution shifts. Swin and ResNet models were similarly accurate in the in-distribution dataset. Swin was more accurate than other methods (follow-up: 0.84, OOD: 0.83) even when subject to color shifts (follow-up: 0.83, OOD: 0.87), indicating capability to provide robust performance for longitudinal cancer assessment.
Authors: Andrea Napoli, Paul White
Abstract: This paper considers the problem of model selection under domain shift. Motivated by principles from distributionally robust optimisation (DRO) and domain adaptation theory, it is proposed that the training-validation split should maximise the distribution mismatch between the two sets. By adopting the maximum mean discrepancy (MMD) as the measure of mismatch, it is shown that the partitioning problem reduces to kernel k-means clustering. A constrained clustering algorithm, which leverages linear programming to control the size, label, and (optionally) group distributions of the splits, is presented. The algorithm does not require additional metadata, and comes with convergence guarantees. In experiments, the technique consistently outperforms alternative splitting strategies across a range of datasets and training algorithms, for both domain generalisation (DG) and unsupervised domain adaptation (UDA) tasks. Analysis also shows the MMD between the training and validation sets to be strongly rank-correlated ($\rho=0.63$) with test domain accuracy, further substantiating the validity of this approach.
Authors: Maxime Di Folco, Cosmin I. Bercea, Emily Chan, Julia A. Schnabel
Abstract: Interpretability is essential in medical imaging to ensure that clinicians can comprehend and trust artificial intelligence models. Several approaches have been recently considered to encode attributes in the latent space to enhance its interpretability. Notably, attribute regularization aims to encode a set of attributes along the dimensions of a latent representation. However, this approach is based on Variational AutoEncoder and suffers from blurry reconstruction. In this paper, we propose an Attributed-regularized Soft Introspective Variational Autoencoder that combines attribute regularization of the latent space within the framework of an adversarially trained variational autoencoder. We demonstrate on short-axis cardiac Magnetic Resonance images of the UK Biobank the ability of the proposed method to address blurry reconstruction issues of variational autoencoder methods while preserving the latent space interpretability.
Authors: Yunbei Zhang, Akshay Mehra, Jihun Hamm
Abstract: Continual Test-Time Adaptation (CTTA) seeks to adapt a source pre-trained model to continually changing, unlabeled target domains. Existing TTA methods are typically designed for environments where domain changes occur sequentially and can struggle in more dynamic scenarios, as illustrated in Figure \ref{fig:settings}. Inspired by the principles of online K-Means, we introduce a novel approach to CTTA through visual prompting. We propose a \emph{Dynamic Prompt Coreset} that not only preserves knowledge from previously visited domains but also accommodates learning from new potential domains. This is complemented by a distance-based \emph{Weight Updating Mechanism} that ensures the coreset remains current and relevant. Our approach employs a fixed model architecture alongside the coreset and an innovative updating system to effectively mitigate challenges such as catastrophic forgetting and error accumulation. Extensive testing on four widely-used benchmarks demonstrates that our method consistently outperforms state-of-the-art alternatives in both classification and segmentation CTTA tasks across the structured and dynamic CTTA settings, with $99\%$ fewer trainable parameters.
Authors: Bao Gia Doan, Afshar Shamsi, Xiao-Yu Guo, Arash Mohammadi, Hamid Alinejad-Rokny, Dino Sejdinovic, Damith C. Ranasinghe, Ehsan Abbasnejad
Abstract: Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.
Authors: Jiahao Su, Kang You, Zekai Xu, Weizhi Xu, Zhezhi He
Abstract: Spiking neural networks (SNNs) are becoming a promising alternative to conventional artificial neural networks (ANNs) due to their rich neural dynamics and the implementation of energy-efficient neuromorphic chips. However, the non-differential binary communication mechanism makes SNN hard to converge to an ANN-level accuracy. When SNN encounters sequence learning, the situation becomes worse due to the difficulties in modeling long-range dependencies. To overcome these difficulties, researchers developed variants of LIF neurons and different surrogate gradients but still failed to obtain good results when the sequence became longer (e.g., $>$500). Unlike them, we obtain an optimal SNN in sequence learning by directly mapping parameters from a quantized CRNN. We design two sub-pipelines to support the end-to-end conversion of different structures in neural networks, which is called CNN-Morph (CNN $\rightarrow$ QCNN $\rightarrow$ BIFSNN) and RNN-Morph (RNN $\rightarrow$ QRNN $\rightarrow$ RBIFSNN). Using conversion pipelines and the s-analog encoding method, the conversion error of our framework is zero. Furthermore, we give the theoretical and experimental demonstration of the lossless CRNN-SNN conversion. Our results show the effectiveness of our method over short and long timescales tasks compared with the state-of-the-art learning- and conversion-based methods. We reach the highest accuracy of 99.16% (0.46 $\uparrow$) on S-MNIST, 94.95% (3.95 $\uparrow$) on PS-MNIST (sequence length of 784) respectively, and the lowest loss of 0.057 (0.013 $\downarrow$) within 8 time-steps in collision avoidance dataset.
Authors: Christoph Auer, Maksym Lysak, Ahmed Nassar, Michele Dolfi, Nikolaos Livathinos, Panos Vagenas, Cesar Berrospi Ramis, Matteo Omenetti, Fabian Lindlbauer, Kasper Dinkla, Valery Weber, Lucas Morin, Ingmar Meijer, Viktor Kuropiatnyk, Peter W. J. Staar
Abstract: This technical report introduces Docling, an easy to use, self-contained, MIT-licensed open-source package for PDF document conversion. It is powered by state-of-the-art specialized AI models for layout analysis (DocLayNet) and table structure recognition (TableFormer), and runs efficiently on commodity hardware in a small resource budget. The code interface allows for easy extensibility and addition of new features and models.
Authors: Kai-liang Lu, Yu-meng Su, Zhuo Bi, Cheng Qiu, Wen-jun Zhang
Abstract: This paper compared physics-informed neural network (PINN), conventional neural network (NN) and traditional numerical discretization methods on solving differential equations (DEs) through literature investigation and experimental validation. We focused on the soft-constrained PINN approach and formalized its mathematical framework and computational flow for solving Ordinary DEs and Partial DEs (ODEs/PDEs). The working mechanism and its accuracy and efficiency were experimentally verified by solving typical linear and non-linear (e.g., Primer, Van der Pol, Duffing) oscillator ODEs. We demonstrate that the DeepXDE-based implementation of PINN is not only light code and efficient in training, but also flexible across CPU/GPU platforms. PINN greatly reduces the need for labeled data: when the nonlinearity of the ODE is weak, a very small amount of supervised training data plus a few unsupervised collocation points are sufficient to predict the solution; in the minimalist case, only one or two training points (with initial values) are needed for first- or second-order ODEs, respectively. We also find that, with the aid of collocation points and the use of physical information, PINN has the ability to extrapolate data outside the time domain of the training set, and especially is robust to noisy data, thus with enhanced generalization capabilities. Training is accelerated when the gains obtained along with the reduction in the amount of data outweigh the delay caused by the increase in the loss function terms. The soft-constrained PINN can easily impose a physical law (e.g., conservation of energy) constraint by adding a regularization term to the total loss function, thus improving the solution performance to ODEs that obey this physical law. Furthermore, PINN can also be used for stiff ODEs, PDEs, and other types of DEs, and is becoming a favorable catalyst for the era of Digital Twins.
Authors: Zhenye Lou, Qing Xu, Zekun Jiang, Xiangjian He, Zhen Chen, Yi Wang, Chenxin Li, Maggie M. He, Wenting Duan
Abstract: Domain-generalized nuclei segmentation refers to the generalizability of models to unseen domains based on knowledge learned from source domains and is challenged by various image conditions, cell types, and stain strategies. Recently, the Segment Anything Model (SAM) has made great success in universal image segmentation by interactive prompt modes (e.g., point and box). Despite its strengths, the original SAM presents limited adaptation to medical images. Moreover, SAM requires providing manual bounding box prompts for each object to produce satisfactory segmentation masks, so it is laborious in nuclei segmentation scenarios. To address these limitations, we propose a domain-generalizable framework for nuclei image segmentation, abbreviated to NuSegDG. Specifically, we first devise a Heterogeneous Space Adapter (HS-Adapter) to learn multi-dimensional feature representations of different nuclei domains by injecting a small number of trainable parameters into the image encoder of SAM. To alleviate the labor-intensive requirement of manual prompts, we introduce a Gaussian-Kernel Prompt Encoder (GKP-Encoder) to generate density maps driven by a single point, which guides segmentation predictions by mixing position prompts and semantic prompts. Furthermore, we present a Two-Stage Mask Decoder (TSM-Decoder) to effectively convert semantic masks to instance maps without the manual demand for morphological shape refinement. Based on our experimental evaluations, the proposed NuSegDG demonstrates state-of-the-art performance in nuclei instance segmentation, exhibiting superior domain generalization capabilities. The source code is available at https://github.com/xq141839/NuSegDG.
Authors: Chaoya Jiang, Jia Hongrui, Haiyang Xu, Wei Ye, Mengfan Dong, Ming Yan, Ji Zhang, Fei Huang, Shikun Zhang
Abstract: This paper presents MaVEn, an innovative Multi-granularity Visual Encoding framework designed to enhance the capabilities of Multimodal Large Language Models (MLLMs) in multi-image reasoning. Current MLLMs primarily focus on single-image visual understanding, limiting their ability to interpret and integrate information across multiple images. MaVEn addresses this limitation by combining discrete visual symbol sequences, which abstract coarse-grained semantic concepts, with traditional continuous representation sequences that model fine-grained features. This dual approach bridges the semantic gap between visual and textual data, thereby improving the model's ability to process and interpret information from multiple images effectively. Additionally, we design a dynamic reduction mechanism by for long-sequence continuous features to enhance multi-image processing efficiency. Experimental results demonstrate that MaVEn significantly enhances MLLMs' understanding in complex multi-image scenarios, while also improving performance in single-image contexts.
Authors: Haojun Shi, Suyu Ye, Xinyu Fang, Chuanyang Jin, Leyla Isik, Yen-Ling Kuo, Tianmin Shu
Abstract: Understanding people's social interactions in complex real-world scenarios often relies on intricate mental reasoning. To truly understand how and why people interact with one another, we must infer the underlying mental states that give rise to the social interactions, i.e., Theory of Mind reasoning in multi-agent interactions. Additionally, social interactions are often multi-modal -- we can watch people's actions, hear their conversations, and/or read about their past behaviors. For AI systems to successfully and safely interact with people in real-world environments, they also need to understand people's mental states as well as their inferences about each other's mental states based on multi-modal information about their interactions. For this, we introduce MuMA-ToM, a Multi-modal Multi-Agent Theory of Mind benchmark. MuMA-ToM is the first multi-modal Theory of Mind benchmark that evaluates mental reasoning in embodied multi-agent interactions. In MuMA-ToM, we provide video and text descriptions of people's multi-modal behavior in realistic household environments. Based on the context, we then ask questions about people's goals, beliefs, and beliefs about others' goals. We validated MuMA-ToM in a human experiment and provided a human baseline. We also proposed a novel multi-modal, multi-agent ToM model, LIMP (Language model-based Inverse Multi-agent Planning). Our experimental results show that LIMP significantly outperforms state-of-the-art methods, including large multi-modal models (e.g., GPT-4o, Gemini-1.5 Pro) and a recent multi-modal ToM model, BIP-ALM.
Authors: Ling Lin, Yihang Zhou, Zhanqi Hu, Dian Jiang, Congcong Liu, Shuo Zhou, Yanjie Zhu, Jianxiang Liao, Dong Liang, Hairong Zheng, Haifeng Wang
Abstract: Tuberous sclerosis complex (TSC) manifests as a multisystem disorder with significant neurological implications. This study addresses the critical need for robust classification models tailored to TSC in pediatric patients, introducing QResNet,a novel deep learning model seamlessly integrating conventional convolutional neural networks with quantum neural networks. The model incorporates a two-layer quantum layer (QL), comprising ZZFeatureMap and Ansatz layers, strategically designed for processing classical data within a quantum framework. A comprehensive evaluation, demonstrates the superior performance of QResNet in TSC MRI image classification compared to conventional 3D-ResNet models. These compelling findings underscore the potential of quantum computing to revolutionize medical imaging and diagnostics.Remarkably, this method surpasses conventional CNNs in accuracy and Area Under the Curve (AUC) metrics with the current dataset. Future research endeavors may focus on exploring the scalability and practical implementation of quantum algorithms in real-world medical imaging scenarios.