Authors: Isaac YL Lee, Thanh Nguyen-Duc, Ryo Ueno, Jesse Smith, Peter Y Chan
Abstract: Excessive caregiver workload in hospital nurses has been implicated in poorer patient care and increased worker burnout. Measurement of this workload in the Intensive Care Unit (ICU) is often done using the Nursing Activities Score (NAS), but this is usually recorded manually and sporadically. Previous work has made use of Ambient Intelligence (AmI) by using computer vision to passively derive caregiver-patient interaction times to monitor staff workload. In this letter, we propose using a Multiscale Vision Transformer (MViT) to passively predict the NAS from low-resolution thermal videos recorded in an ICU. 458 videos were obtained from an ICU in Melbourne, Australia and used to train a MViTv2 model using an indirect prediction and a direct prediction method. The indirect method predicted 1 of 8 potentially identifiable NAS activities from the video before inferring the NAS. The direct method predicted the NAS score immediately from the video. The indirect method yielded an average 5-fold accuracy of 57.21%, an area under the receiver operating characteristic curve (ROC AUC) of 0.865, a F1 score of 0.570 and a mean squared error (MSE) of 28.16. The direct method yielded a MSE of 18.16. We also showed that the MViTv2 outperforms similar models such as R(2+1)D and ResNet50-LSTM under identical settings. This study shows the feasibility of using a MViTv2 to passively predict the NAS in an ICU and monitor staff workload automatically. Our results above also show an increased accuracy in predicting NAS directly versus predicting NAS indirectly. We hope that our study can provide a direction for future work and further improve the accuracy of passive NAS monitoring.
Authors: Amandeep Kumar, Muhammad Awais, Sanath Narayan, Hisham Cholakkal, Salman Khan, Rao Muhammad Anwer
Abstract: Drawing upon StyleGAN's expressivity and disentangled latent space, existing 2D approaches employ textual prompting to edit facial images with different attributes. In contrast, 3D-aware approaches that generate faces at different target poses require attribute-specific classifiers, learning separate model weights for each attribute, and are not scalable for novel attributes. In this work, we propose an efficient, plug-and-play, 3D-aware face editing framework based on attribute-specific prompt learning, enabling the generation of facial images with controllable attributes across various target poses. To this end, we introduce a text-driven learnable style token-based latent attribute editor (LAE). The LAE harnesses a pre-trained vision-language model to find text-guided attribute-specific editing direction in the latent space of any pre-trained 3D-aware GAN. It utilizes learnable style tokens and style mappers to learn and transform this editing direction to 3D latent space. To train LAE with multiple attributes, we use directional contrastive loss and style token loss. Furthermore, to ensure view consistency and identity preservation across different poses and attributes, we employ several 3D-aware identity and pose preservation losses. Our experiments show that our proposed framework generates high-quality images with 3D awareness and view consistency while maintaining attribute-specific features. We demonstrate the effectiveness of our method on different facial attributes, including hair color and style, expression, and others. Code: https://github.com/VIROBO-15/Efficient-3D-Aware-Facial-Image-Editing.
URLs: https://github.com/VIROBO-15/Efficient-3D-Aware-Facial-Image-Editing.
Authors: Sergio Casas, Ben Agro, Jiageng Mao, Thomas Gilles, Alexander Cui, Thomas Li, Raquel Urtasun
Abstract: The tasks of object detection and trajectory forecasting play a crucial role in understanding the scene for autonomous driving. These tasks are typically executed in a cascading manner, making them prone to compounding errors. Furthermore, there is usually a very thin interface between the two tasks, creating a lossy information bottleneck. To address these challenges, our approach formulates the union of the two tasks as a trajectory refinement problem, where the first pose is the detection (current time), and the subsequent poses are the waypoints of the multiple forecasts (future time). To tackle this unified task, we design a refinement transformer that infers the presence, pose, and multi-modal future behaviors of objects directly from LiDAR point clouds and high-definition maps. We call this model DeTra, short for object Detection and Trajectory forecasting. In our experiments, we observe that \ourmodel{} outperforms the state-of-the-art on Argoverse 2 Sensor and Waymo Open Dataset by a large margin, across a broad range of metrics. Last but not least, we perform extensive ablation studies that show the value of refinement for this task, that every proposed component contributes positively to its performance, and that key design choices were made.
Authors: Haokun Zhou, Yipeng Hong
Abstract: This study assesses the ability of Large Vision-Language Models (LVLMs) to differentiate between AI-generated and human-generated images. It introduces a new automated benchmark construction method for this evaluation. The experiment compared common LVLMs with human participants using a mixed dataset of AI and human-created images. Results showed that LVLMs could distinguish between the image types to some extent but exhibited a rightward bias, and perform significantly worse compared to humans. To build on these findings, we developed an automated benchmark construction process using AI. This process involved topic retrieval, narrative script generation, error embedding, and image generation, creating a diverse set of text-image pairs with intentional errors. We validated our method through constructing two caparable benchmarks. This study highlights the strengths and weaknesses of LVLMs in real-world understanding and advances benchmark construction techniques, providing a scalable and automatic approach for AI model evaluation.
Authors: Maying Shen, Hongxu Yin, Pavlo Molchanov, Lei Mao, Jose M. Alvarez
Abstract: Data often arrives in sequence over time in real-world deep learning applications such as autonomous driving. When new training data is available, training the model from scratch undermines the benefit of leveraging the learned knowledge, leading to significant training costs. Warm-starting from a previously trained checkpoint is the most intuitive way to retain knowledge and advance learning. However, existing literature suggests that this warm-starting degrades generalization. In this paper, we advocate for warm-starting but stepping out of the previous converging point, thus allowing a better adaptation to new data without compromising previous knowledge. We propose Knowledge Consolidation and Acquisition (CKCA), a continuous model improvement algorithm with two novel components. First, a novel feature regularization (FeatReg) to retain and refine knowledge from existing checkpoints; Second, we propose adaptive knowledge distillation (AdaKD), a novel approach to forget mitigation and knowledge transfer. We tested our method on ImageNet using multiple splits of the training data. Our approach achieves up to $8.39\%$ higher top1 accuracy than the vanilla warm-starting and consistently outperforms the prior art with a large margin.
Authors: Abdelrahman Abdallah, Mahmoud Abdalla, Mahmoud SalahEldin Kasem, Mohamed Mahmoud, Ibrahim Abdelhalim, Mohamed Elkasaby, Yasser ElBendary, Adam Jatowt
Abstract: In the fields of Optical Character Recognition (OCR) and Natural Language Processing (NLP), integrating multilingual capabilities remains a critical challenge, especially when considering languages with complex scripts such as Arabic. This paper introduces the Comprehensive Post-OCR Parsing and Receipt Understanding Dataset (CORU), a novel dataset specifically designed to enhance OCR and information extraction from receipts in multilingual contexts involving Arabic and English. CORU consists of over 20,000 annotated receipts from diverse retail settings, including supermarkets and clothing stores, alongside 30,000 annotated images for OCR that were utilized to recognize each detected line, and 10,000 items annotated for detailed information extraction. These annotations capture essential details such as merchant names, item descriptions, total prices, receipt numbers, and dates. They are structured to support three primary computational tasks: object detection, OCR, and information extraction. We establish the baseline performance for a range of models on CORU to evaluate the effectiveness of traditional methods, like Tesseract OCR, and more advanced neural network-based approaches. These baselines are crucial for processing the complex and noisy document layouts typical of real-world receipts and for advancing the state of automated multilingual document processing. Our datasets are publicly accessible (https://github.com/Update-For-Integrated-Business-AI/CORU).
URLs: https://github.com/Update-For-Integrated-Business-AI/CORU).
Authors: Dujian Ding, Bicheng Xu, Laks V. S. Lakshmanan
Abstract: Image classification is a fundamental building block for a majority of computer vision applications. With the growing popularity and capacity of machine learning models, people can easily access trained image classifiers as a service online or offline. However, model use comes with a cost and classifiers of higher capacity usually incur higher inference costs. To harness the respective strengths of different classifiers, we propose a principled approach, OCCAM, to compute the best classifier assignment strategy over image classification queries (termed as the optimal model portfolio) so that the aggregated accuracy is maximized, under user-specified cost budgets. Our approach uses an unbiased and low-variance accuracy estimator and effectively computes the optimal solution by solving an integer linear programming problem. On a variety of real-world datasets, OCCAM achieves 40% cost reduction with little to no accuracy drop.
Authors: F. A. Mamun, S. A. H. Chowdhury, J. E. Giti, H. Sarker
Abstract: The use of convolutional neural networks (CNNs) has accelerated the progress of handwritten character classification/recognition. Handwritten character recognition (HCR) has found applications in various domains, such as traffic signal detection, language translation, and document information extraction. However, the widespread use of existing HCR technology is yet to be seen as it does not provide reliable character recognition with outstanding accuracy. One of the reasons for unreliable HCR is that existing HCR methods do not take the handwriting styles of non-native writers into account. Hence, further improvement is needed to ensure the reliability and extensive deployment of character recognition technologies for critical tasks. In this work, the classification of English characters written by non-native users is performed by proposing a custom-tailored CNN model. We train this CNN with a new dataset called the handwritten isolated English character (HIEC) dataset. This dataset consists of 16,496 images collected from 260 persons. This paper also includes an ablation study of our CNN by adjusting hyperparameters to identify the best model for the HIEC dataset. The proposed model with five convolutional layers and one hidden layer outperforms state-of-the-art models in terms of character recognition accuracy and achieves an accuracy of $\mathbf{97.04}$%. Compared with the second-best model, the relative improvement of our model in terms of classification accuracy is $\mathbf{4.38}$%.
Authors: Ionu\c{t} Grigore, C\u{a}lin-Adrian Popa
Abstract: In the field of self-supervised depth estimation, Convolutional Neural Networks (CNNs) and Transformers have traditionally been dominant. However, both architectures struggle with efficiently handling long-range dependencies due to their local focus or computational demands. To overcome this limitation, we present MambaDepth, a versatile network tailored for self-supervised depth estimation. Drawing inspiration from the strengths of the Mamba architecture, renowned for its adept handling of lengthy sequences and its ability to capture global context efficiently through a State Space Model (SSM), we introduce MambaDepth. This innovative architecture combines the U-Net's effectiveness in self-supervised depth estimation with the advanced capabilities of Mamba. MambaDepth is structured around a purely Mamba-based encoder-decoder framework, incorporating skip connections to maintain spatial information at various levels of the network. This configuration promotes an extensive feature learning process, enabling the capture of fine details and broader contexts within depth maps. Furthermore, we have developed a novel integration technique within the Mamba blocks to facilitate uninterrupted connectivity and information flow between the encoder and decoder components, thereby improving depth accuracy. Comprehensive testing across the established KITTI dataset demonstrates MambaDepth's superiority over leading CNN and Transformer-based models in self-supervised depth estimation task, allowing it to achieve state-of-the-art performance. Moreover, MambaDepth proves its superior generalization capacities on other datasets such as Make3D and Cityscapes. MambaDepth's performance heralds a new era in effective long-range dependency modeling for self-supervised depth estimation.
Authors: Luyang Zhu, Yingwei Li, Nan Liu, Hao Peng, Dawei Yang, Ira Kemelmacher-Shlizerman
Abstract: We present M&M VTO, a mix and match virtual try-on method that takes as input multiple garment images, text description for garment layout and an image of a person. An example input includes: an image of a shirt, an image of a pair of pants, "rolled sleeves, shirt tucked in", and an image of a person. The output is a visualization of how those garments (in the desired layout) would look like on the given person. Key contributions of our method are: 1) a single stage diffusion based model, with no super resolution cascading, that allows to mix and match multiple garments at 1024x512 resolution preserving and warping intricate garment details, 2) architecture design (VTO UNet Diffusion Transformer) to disentangle denoising from person specific features, allowing for a highly effective finetuning strategy for identity preservation (6MB model per individual vs 4GB achieved with, e.g., dreambooth finetuning); solving a common identity loss problem in current virtual try-on methods, 3) layout control for multiple garments via text inputs specifically finetuned over PaLI-3 for virtual try-on task. Experimental results indicate that M&M VTO achieves state-of-the-art performance both qualitatively and quantitatively, as well as opens up new opportunities for virtual try-on via language-guided and multi-garment try-on.
Authors: Sabri Mustafa Kahya, Boran Hamdi Sivrikaya, Muhammet Sami Yavuz, Eckehard Steinbach
Abstract: This paper proposes a short-range FMCW radar-based facial authentication and out-of-distribution (OOD) detection framework. Our pipeline jointly estimates the correct classes for the in-distribution (ID) samples and detects the OOD samples to prevent their inaccurate prediction. Our reconstruction-based architecture consists of a main convolutional block with one encoder and multi-decoder configuration, and intermediate linear encoder-decoder parts. Together, these elements form an accurate human face classifier and a robust OOD detector. For our dataset, gathered using a 60 GHz short-range FMCW radar, our network achieves an average classification accuracy of 98.07% in identifying in-distribution human faces. As an OOD detector, it achieves an average Area Under the Receiver Operating Characteristic (AUROC) curve of 98.50% and an average False Positive Rate at 95% True Positive Rate (FPR95) of 6.20%. Also, our extensive experiments show that the proposed approach outperforms previous OOD detectors in terms of common OOD detection metrics.
Authors: Reyhane Askari Hemmat, Melissa Hall, Alicia Sun, Candace Ross, Michal Drozdzal, Adriana Romero-Soriano
Abstract: With the growing popularity of text-to-image generative models, there has been increasing focus on understanding their risks and biases. Recent work has found that state-of-the-art models struggle to depict everyday objects with the true diversity of the real world and have notable gaps between geographic regions. In this work, we aim to increase the diversity of generated images of common objects such that per-region variations are representative of the real world. We introduce an inference time intervention, contextualized Vendi Score Guidance (c-VSG), that guides the backwards steps of latent diffusion models to increase the diversity of a sample as compared to a "memory bank" of previously generated images while constraining the amount of variation within that of an exemplar set of real-world contextualizing images. We evaluate c-VSG with two geographically representative datasets and find that it substantially increases the diversity of generated images, both for the worst performing regions and on average, while simultaneously maintaining or improving image quality and consistency. Additionally, qualitative analyses reveal that diversity of generated images is significantly improved, including along the lines of reductive region portrayals present in the original model. We hope that this work is a step towards text-to-image generative models that reflect the true geographic diversity of the world.
Authors: Matthew Rodda, Sofia McLeod, Ky Cuong Pham, Tat-Jun Chin
Abstract: As space missions aim to explore increasingly hazardous terrain, accurate and timely position estimates are required to ensure safe navigation. Vision-based navigation achieves this goal through correlating impact craters visible through onboard imagery with a known database to estimate a craft's pose. However, existing literature has not sufficiently evaluated crater-detection algorithm (CDA) performance from imagery containing off-nadir view angles. In this work, we evaluate the performance of Mask R-CNN for crater detection, comparing models pretrained on simulated data containing off-nadir view angles and to pretraining on real-lunar images. We demonstrate pretraining on real-lunar images is superior despite the lack of images containing off-nadir view angles, achieving detection performance of 63.1 F1-score and ellipse-regression performance of 0.701 intersection over union. This work provides the first quantitative analysis of performance of CDAs on images containing off-nadir view angles. Towards the development of increasingly robust CDAs, we additionally provide the first annotated CDA dataset with off-nadir view angles from the Chang'e 5 Landing Camera.
Authors: Yiheng Zhang, Yunkang Cao, Tianhang Zhang, Weiming Shen
Abstract: This study targets Multi-Lighting Image Anomaly Detection (MLIAD), where multiple lighting conditions are utilized to enhance imaging quality and anomaly detection performance. While numerous image anomaly detection methods have been proposed, they lack the capacity to handle multiple inputs for a single sample, like multi-lighting images in MLIAD. Hence, this study proposes Attention Fusion Reverse Distillation (AFRD) to handle multiple inputs in MLIAD. For this purpose, AFRD utilizes a pre-trained teacher network to extract features from multiple inputs. Then these features are aggregated into fused features through an attention module. Subsequently, a corresponding student net-work is utilized to regress the attention fused features. The regression errors are denoted as anomaly scores during inference. Experiments on Eyecandies demonstrates that AFRD achieves superior MLIAD performance than other MLIAD alternatives, also highlighting the benefit of using multiple lighting conditions for anomaly detection.
Authors: Deshui Miao, Xin Li, Zhenyu He, Yaowei Wang, Ming-Hsuan Yang
Abstract: Tracking and segmenting multiple objects in complex scenes has always been a challenge in the field of video object segmentation, especially in scenarios where objects are occluded and split into parts. In such cases, the definition of objects becomes very ambiguous. The motivation behind the MOSE dataset is how to clearly recognize and distinguish objects in complex scenes. In this challenge, we propose a semantic embedding video object segmentation model and use the salient features of objects as query representations. The semantic understanding helps the model to recognize parts of the objects and the salient feature captures the more discriminative features of the objects. Trained on a large-scale video object segmentation dataset, our model achieves first place (\textbf{84.45\%}) in the test set of PVUW Challenge 2024: Complex Video Object Segmentation Track.
Authors: Xinquan Yang, Xuguang Li, Xiaoling Luo, Leilei Zeng, Yudi Zhang, Linlin Shen, Yongqiang Deng
Abstract: Surgical guide plate is an important tool for the dental implant surgery. However, the design process heavily relies on the dentist to manually simulate the implant angle and depth. When deep neural networks have been applied to assist the dentist quickly locates the implant position, most of them are not able to determine the implant depth. Inspired by the video grounding task which localizes the starting and ending time of the target video segment, in this paper, we simplify the implant depth prediction as video grounding and develop a Texture Perceive Implant Depth Prediction Network (TPNet), which enables us to directly output the implant depth without complex measurements of oral bone. TPNet consists of an implant region detector (IRD) and an implant depth prediction network (IDPNet). IRD is an object detector designed to crop the candidate implant volume from the CBCT, which greatly saves the computation resource. IDPNet takes the cropped CBCT data to predict the implant depth. A Texture Perceive Loss (TPL) is devised to enable the encoder of IDPNet to perceive the texture variation among slices. Extensive experiments on a large dental implant dataset demonstrated that the proposed TPNet achieves superior performance than the existing methods.
Authors: Peng Xing, Dong Zhang, Jinhui Tang, Zechao li
Abstract: Anomaly detection (AD) has been extensively studied and applied in a wide range of scenarios in the recent past. However, there are still gaps between achieved and desirable levels of recognition accuracy for making AD for practical applications. In this paper, we start from an insightful analysis of two types of fundamental yet representative failure cases in the baseline model, and reveal reasons that hinder current AD methods from achieving a higher recognition accuracy. Specifically, by Case-1, we found that the main reasons detrimental to current AD methods is that the inputs to the recovery model contain a large number of detailed features to be recovered, which leads to the normal/abnormal area has-not/has been recovered into its original state. By Case-2, we surprisingly found that the abnormal area that cannot be recognized in image-level representations can be easily recognized in the feature-level representation. Based on the above observations, we propose a novel Recover-then-Discriminate (ReDi) framework for AD. ReDi takes a self-generated feature map and a selected prompted image as explicit input information to solve problems in case-1. Concurrently, a feature-level discriminative network is proposed to enhance abnormal differences between the recovered representation and the input representation. Extensive experimental results on two popular yet challenging AD datasets validate that ReDi achieves the new state-of-the-art accuracy.
Authors: Vipin V
Abstract: A novel approach for forest fire detection using image processing technique is proposed. A rule-based color model for fire pixel classification is used. The proposed algorithm uses RGB and YCbCr color space. The advantage of using YCbCr color space is that it can separate the luminance from the chrominance more effectively than RGB color space. The performance of the proposed algorithm is tested on two sets of images, one of which contains fire; the other contains fire-like regions. Standard methods are used for calculating the performance of the algorithm. The proposed method has both higher detection rate and lower false alarm rate. Since the algorithm is cheap in computation, it can be used for real-time forest fire detection.
Authors: Zenghao Chai, Chen Tang, Yongkang Wong, Mohan Kankanhalli
Abstract: The creation of 4D avatars (i.e., animated 3D avatars) from text description typically uses text-to-image (T2I) diffusion models to synthesize 3D avatars in the canonical space and subsequently applies animation with target motions. However, such an optimization-by-animation paradigm has several drawbacks. (1) For pose-agnostic optimization, the rendered images in canonical pose for naive Score Distillation Sampling (SDS) exhibit domain gap and cannot preserve view-consistency using only T2I priors, and (2) For post hoc animation, simply applying the source motions to target 3D avatars yields translation artifacts and misalignment. To address these issues, we propose Skeleton-aware Text-based 4D Avatar generation with in-network motion Retargeting (STAR). STAR considers the geometry and skeleton differences between the template mesh and target avatar, and corrects the mismatched source motion by resorting to the pretrained motion retargeting techniques. With the informatively retargeted and occlusion-aware skeleton, we embrace the skeleton-conditioned T2I and text-to-video (T2V) priors, and propose a hybrid SDS module to coherently provide multi-view and frame-consistent supervision signals. Hence, STAR can progressively optimize the geometry, texture, and motion in an end-to-end manner. The quantitative and qualitative experiments demonstrate our proposed STAR can synthesize high-quality 4D avatars with vivid animations that align well with the text description. Additional ablation studies shows the contributions of each component in STAR. The source code and demos are available at: \href{https://star-avatar.github.io}{https://star-avatar.github.io}.
URLs: https://star-avatar.github.io, https://star-avatar.github.io
Authors: Yuchao Wang, Peirui Cheng, Pengju Tian, Ziyang Yuan, Liangjin Zhao, Jing Tian, Wensheng Wang, Zhirui Wang, Xian Sun
Abstract: With the advancement of collaborative perception, the role of aerial-ground collaborative perception, a crucial component, is becoming increasingly important. The demand for collaborative perception across different perspectives to construct more comprehensive perceptual information is growing. However, challenges arise due to the disparities in the field of view between cross-domain agents and their varying sensitivity to information in images. Additionally, when we transform image features into Bird's Eye View (BEV) features for collaboration, we need accurate depth information. To address these issues, we propose a framework specifically designed for aerial-ground collaboration. First, to mitigate the lack of datasets for aerial-ground collaboration, we develop a virtual dataset named V2U-COO for our research. Second, we design a Cross-Domain Cross-Adaptation (CDCA) module to align the target information obtained from different domains, thereby achieving more accurate perception results. Finally, we introduce a Collaborative Depth Optimization (CDO) module to obtain more precise depth estimation results, leading to more accurate perception outcomes. We conduct extensive experiments on both our virtual dataset and a public dataset to validate the effectiveness of our framework. Our experiments on the V2U-COO dataset and the DAIR-V2X dataset demonstrate that our method improves detection accuracy by 6.1% and 2.7%, respectively.
Authors: Pengju Tian, Peirui Cheng, Yuchao Wang, Zhechao Wang, Zhirui Wang, Menglong Yan, Xue Yang, Xian Sun
Abstract: Multi-UAV collaborative 3D object detection can perceive and comprehend complex environments by integrating complementary information, with applications encompassing traffic monitoring, delivery services and agricultural management. However, the extremely broad observations in aerial remote sensing and significant perspective differences across multiple UAVs make it challenging to achieve precise and consistent feature mapping from 2D images to 3D space in multi-UAV collaborative 3D object detection paradigm. To address the problem, we propose an unparalleled camera-based multi-UAV collaborative 3D object detection paradigm called UCDNet. Specifically, the depth information from the UAVs to the ground is explicitly utilized as a strong prior to provide a reference for more accurate and generalizable feature mapping. Additionally, we design a homologous points geometric consistency loss as an auxiliary self-supervision, which directly influences the feature mapping module, thereby strengthening the global consistency of multi-view perception. Experiments on AeroCollab3D and CoPerception-UAVs datasets show our method increases 4.7% and 10% mAP respectively compared to the baseline, which demonstrates the superiority of UCDNet.
Authors: Zengyuan Lai, Jiarui Yang, Songpengcheng Xia, Qi Wu, Zhen Sun, Wenxian Yu, Ling Pei
Abstract: Patients with mental disorders often exhibit risky abnormal actions, such as climbing walls or hitting windows, necessitating intelligent video behavior monitoring for smart healthcare with the rising Internet of Things (IoT) technology. However, the development of vision-based Human Action Recognition (HAR) for these actions is hindered by the lack of specialized algorithms and datasets. In this paper, we innovatively propose to build a vision-based HAR dataset including abnormal actions often occurring in the mental disorder group and then introduce a novel Scene-Motion-aware Action Recognition Technology framework, named SMART, consisting of two technical modules. First, we propose a scene perception module to extract human motion trajectory and human-scene interaction features, which introduces additional scene information for a supplementary semantic representation of the above actions. Second, the multi-stage fusion module fuses the skeleton motion, motion trajectory, and human-scene interaction features, enhancing the semantic association between the skeleton motion and the above supplementary representation, thus generating a comprehensive representation with both human motion and scene information. The effectiveness of our proposed method has been validated on our self-collected HAR dataset (MentalHAD), achieving 94.9% and 93.1% accuracy in un-seen subjects and scenes and outperforming state-of-the-art approaches by 6.5% and 13.2%, respectively. The demonstrated subject- and scene- generalizability makes it possible for SMART's migration to practical deployment in smart healthcare systems for mental disorder patients in medical settings. The code and dataset will be released publicly for further research: https://github.com/Inowlzy/SMART.git.
Authors: Dongkai Wang, Shiyu Xuan, Shiliang Zhang
Abstract: The capacity of existing human keypoint localization models is limited by keypoint priors provided by the training data. To alleviate this restriction and pursue more general model, this work studies keypoint localization from a different perspective by reasoning locations based on keypiont clues in text descriptions. We propose LocLLM, the first Large-Language Model (LLM) based keypoint localization model that takes images and text instructions as inputs and outputs the desired keypoint coordinates. LocLLM leverages the strong reasoning capability of LLM and clues of keypoint type, location, and relationship in textual descriptions for keypoint localization. To effectively tune LocLLM, we construct localization-based instruction conversations to connect keypoint description with corresponding coordinates in input image, and fine-tune the whole model in a parameter-efficient training pipeline. LocLLM shows remarkable performance on standard 2D/3D keypoint localization benchmarks. Moreover, incorporating language clues into the localization makes LocLLM show superior flexibility and generalizable capability in cross dataset keypoint localization, and even detecting novel type of keypoints unseen during training.
Authors: Zhenting Wang, Chen Chen, Vikash Sehwag, Minzhou Pan, Lingjuan Lyu
Abstract: The popularity of visual generative AI models like DALL-E 3, Stable Diffusion XL, Stable Video Diffusion, and Sora has been increasing. Through extensive evaluation, we discovered that the state-of-the-art visual generative models can generate content that bears a striking resemblance to characters protected by intellectual property rights held by major entertainment companies (such as Sony, Marvel, and Nintendo), which raises potential legal concerns. This happens when the input prompt contains the character's name or even just descriptive details about their characteristics. To mitigate such IP infringement problems, we also propose a defense method against it. In detail, we develop a revised generation paradigm that can identify potentially infringing generated content and prevent IP infringement by utilizing guidance techniques during the diffusion process. It has the capability to recognize generated content that may be infringing on intellectual property rights, and mitigate such infringement by employing guidance methods throughout the diffusion process without retrain or fine-tune the pretrained models. Experiments on well-known character IPs like Spider-Man, Iron Man, and Superman demonstrate the effectiveness of the proposed defense method. Our data and code can be found at https://github.com/ZhentingWang/GAI_IP_Infringement.
Authors: Sanjoy Chowdhury, Sayan Nag, K J Joseph, Balaji Vasan Srinivasan, Dinesh Manocha
Abstract: Music is a universal language that can communicate emotions and feelings. It forms an essential part of the whole spectrum of creative media, ranging from movies to social media posts. Machine learning models that can synthesize music are predominantly conditioned on textual descriptions of it. Inspired by how musicians compose music not just from a movie script, but also through visualizations, we propose MeLFusion, a model that can effectively use cues from a textual description and the corresponding image to synthesize music. MeLFusion is a text-to-music diffusion model with a novel "visual synapse", which effectively infuses the semantics from the visual modality into the generated music. To facilitate research in this area, we introduce a new dataset MeLBench, and propose a new evaluation metric IMSM. Our exhaustive experimental evaluation suggests that adding visual information to the music synthesis pipeline significantly improves the quality of generated music, measured both objectively and subjectively, with a relative gain of up to 67.98% on the FAD score. We hope that our work will gather attention to this pragmatic, yet relatively under-explored research area.
Authors: Zehong Ma, Shiliang Zhang, Longhui Wei, Qi Tian
Abstract: The challenge of open-vocabulary recognition lies in the model has no clue of new categories it is applied to. Existing works have proposed different methods to embed category cues into the model, \eg, through few-shot fine-tuning, providing category names or textual descriptions to Vision-Language Models. Fine-tuning is time-consuming and degrades the generalization capability. Textual descriptions could be ambiguous and fail to depict visual details. This paper tackles open-vocabulary recognition from a different perspective by referring to multi-modal clues composed of textual descriptions and exemplar images. Our method, named OVMR, adopts two innovative components to pursue a more robust category cues embedding. A multi-modal classifier is first generated by dynamically complementing textual descriptions with image exemplars. A preference-based refinement module is hence applied to fuse uni-modal and multi-modal classifiers, with the aim to alleviate issues of low-quality exemplar images or textual descriptions. The proposed OVMR is a plug-and-play module, and works well with exemplar images randomly crawled from the Internet. Extensive experiments have demonstrated the promising performance of OVMR, \eg, it outperforms existing methods across various scenarios and setups. Codes are publicly available at \href{https://github.com/Zehong-Ma/OVMR}{https://github.com/Zehong-Ma/OVMR}.
URLs: https://github.com/Zehong-Ma/OVMR, https://github.com/Zehong-Ma/OVMR
Authors: Doyi Kim, Minseok Seo, Yeji Choi
Abstract: Recently, data-driven weather forecasting methods have received significant attention for surpassing the RMSE performance of traditional NWP (Numerical Weather Prediction)-based methods. However, data-driven models are tuned to minimize the loss between forecasted data and ground truths, often using pixel-wise loss. This can lead to models that produce blurred outputs, which, despite being significantly different in detail from the actual weather conditions, still demonstrate low RMSE values. Although evaluation metrics from the computer vision field, such as PSNR, SSIM, and FVD, can be used, they are not entirely suitable for weather variables. This is because weather variables exhibit continuous physical changes over time and lack the distinct boundaries of objects typically seen in computer vision images. To resolve these issues, we propose the advection and convection Error (ACE) metric, specifically designed to assess how well models predict advection and convection, which are significant atmospheric transfer methods. We have validated the ACE evaluation metric on the WeatherBench2 and MovingMNIST datasets.
Authors: Haolong Ma, Hui Li, Chunyang Cheng, Xiaoning Song, Zhongwei Shen
Abstract: As a common image processing technique, image decomposition is often used to extract complementary information between modalities. In current decomposition-based image fusion methods, typically, source images are decomposed into three parts at single scale (i.e., visible-exclusive part, infrared-exclusive part, and common part) and lacking interaction between modalities during the decomposition process. These results in the inability of fusion images to effectively focus on finer complementary information between modalities at various scales. To address the above issue, a novel decomposition mechanism, Continuous Decomposition Fusion (CDeFuse), is proposed. Firstly, CDeFuse extends the original three-part decomposition to a more general K-part decomposition at each scale through similarity constraints to fuse multi-scale information and achieve a finer representation of decomposition features. Secondly, a Continuous Decomposition Module (CDM) is introduced to assist K-part decomposition. Its core component, State Transformer (ST), efficiently captures complementary information between modalities by utilizing multi-head self-attention mechanism. Finally, a novel decomposition loss function and the corresponding computational optimization strategy are utilized to ensure the smooth progress of the decomposition process while maintaining linear growth in time complexity with the number of decomposition results K. Extensive experiments demonstrate that our CDeFuse achieves comparable performance compared to previous methods. The code will be publicly available.
Authors: Cong Yang, Zuchao Li, Lefei Zhang
Abstract: Recently, large multimodal models have built a bridge from visual to textual information, but they tend to underperform in remote sensing scenarios. This underperformance is due to the complex distribution of objects and the significant scale differences among targets in remote sensing images, leading to visual ambiguities and insufficient descriptions by these multimodal models. Moreover, the lack of multimodal fine-tuning data specific to the remote sensing field makes it challenging for the model's behavior to align with user queries. To address these issues, this paper proposes an attribute-guided \textbf{Multi-Granularity Instruction Multimodal Model (MGIMM)} for remote sensing image detailed description. MGIMM guides the multimodal model to learn the consistency between visual regions and corresponding text attributes (such as object names, colors, and shapes) through region-level instruction tuning. Then, with the multimodal model aligned on region-attribute, guided by multi-grain visual features, MGIMM fully perceives both region-level and global image information, utilizing large language models for comprehensive descriptions of remote sensing images. Due to the lack of a standard benchmark for generating detailed descriptions of remote sensing images, we construct a dataset featuring 38,320 region-attribute pairs and 23,463 image-detailed description pairs. Compared with various advanced methods on this dataset, the results demonstrate the effectiveness of MGIMM's region-attribute guided learning approach. Code can be available at https://github.com/yangcong356/MGIMM.git
Authors: Eduard Poesina, Adriana Valentina Costache, Adrian-Gabriel Chifu, Josiane Mothe, Radu Tudor Ionescu
Abstract: Text-to-image generation has recently emerged as a viable alternative to text-to-image retrieval, due to the visually impressive results of generative diffusion models. Although query performance prediction is an active research topic in information retrieval, to the best of our knowledge, there is no prior study that analyzes the difficulty of queries (prompts) in text-to-image generation, based on human judgments. To this end, we introduce the first dataset of prompts which are manually annotated in terms of image generation performance. In order to determine the difficulty of the same prompts in image retrieval, we also collect manual annotations that represent retrieval performance. We thus propose the first benchmark for joint text-to-image prompt and query performance prediction, comprising 10K queries. Our benchmark enables: (i) the comparative assessment of the difficulty of prompts/queries in image generation and image retrieval, and (ii) the evaluation of prompt/query performance predictors addressing both generation and retrieval. We present results with several pre-generation/retrieval and post-generation/retrieval performance predictors, thus providing competitive baselines for future research. Our benchmark and code is publicly available under the CC BY 4.0 license at https://github.com/Eduard6421/PQPP.
Authors: Huanhuan Ma, Jinghao Zhang, Qiang Liu, Shu Wu, Liang Wang
Abstract: The rapid spread of information through mobile devices and media has led to the widespread of false or deceptive news, causing significant concerns in society. Among different types of misinformation, image repurposing, also known as out-of-context misinformation, remains highly prevalent and effective. However, current approaches for detecting out-of-context misinformation often lack interpretability and offer limited explanations. In this study, we propose a logic regularization approach for out-of-context detection called LOGRAN (LOGic Regularization for out-of-context ANalysis). The primary objective of LOGRAN is to decompose the out-of-context detection at the phrase level. By employing latent variables for phrase-level predictions, the final prediction of the image-caption pair can be aggregated using logical rules. The latent variables also provide an explanation for how the final result is derived, making this fine-grained detection method inherently explanatory. We evaluate the performance of LOGRAN on the NewsCLIPpings dataset, showcasing competitive overall results. Visualized examples also reveal faithful phrase-level predictions of out-of-context images, accompanied by explanations. This highlights the effectiveness of our approach in addressing out-of-context detection and enhancing interpretability.
Authors: Yuan Tian, Guo Lu, Guangtao Zhai
Abstract: Most video compression methods focus on human visual perception, neglecting semantic preservation. This leads to severe semantic loss during the compression, hampering downstream video analysis tasks. In this paper, we propose a Masked Video Modeling (MVM)-powered compression framework that particularly preserves video semantics, by jointly mining and compressing the semantics in a self-supervised manner. While MVM is proficient at learning generalizable semantics through the masked patch prediction task, it may also encode non-semantic information like trivial textural details, wasting bitcost and bringing semantic noises. To suppress this, we explicitly regularize the non-semantic entropy of the compressed video in the MVM token space. The proposed framework is instantiated as a simple Semantic-Mining-then-Compression (SMC) model. Furthermore, we extend SMC as an advanced SMC++ model from several aspects. First, we equip it with a masked motion prediction objective, leading to better temporal semantic learning ability. Second, we introduce a Transformer-based compression module, to improve the semantic compression efficacy. Considering that directly mining the complex redundancy among heterogeneous features in different coding stages is non-trivial, we introduce a compact blueprint semantic representation to align these features into a similar form, fully unleashing the power of the Transformer-based compression module. Extensive results demonstrate the proposed SMC and SMC++ models show remarkable superiority over previous traditional, learnable, and perceptual quality-oriented video codecs, on three video analysis tasks and seven datasets. \textit{Codes and model are available at: \url{https://github.com/tianyuan168326/VideoSemanticCompression-Pytorch}.
URLs: https://github.com/tianyuan168326/VideoSemanticCompression-Pytorch
Authors: Xingkui Zhu, Yiran Guan, Dingkang Liang, Yuchao Chen, Yuliang Liu, Xiang Bai
Abstract: The sparsely activated mixture of experts (MoE) model presents a promising alternative to traditional densely activated (dense) models, enhancing both quality and computational efficiency. However, training MoE models from scratch demands extensive data and computational resources. Moreover, public repositories like timm mainly provide pre-trained dense checkpoints, lacking similar resources for MoE models, hindering their adoption. To bridge this gap, we introduce MoE Jetpack, an effective method for fine-tuning dense checkpoints into MoE models. MoE Jetpack incorporates two key techniques: (1) checkpoint recycling, which repurposes dense checkpoints as initial weights for MoE models, thereby accelerating convergence, enhancing accuracy, and alleviating the computational burden of pre-training; (2) hyperspherical adaptive MoE (SpheroMoE) layer, which optimizes the MoE architecture for better integration of dense checkpoints, enhancing fine-tuning performance. Our experiments on vision tasks demonstrate that MoE Jetpack significantly improves convergence speed and accuracy when fine-tuning dense checkpoints into MoE models. Our code will be publicly available at https://github.com/Adlith/MoE-Jetpack.
Authors: Bing Cao, Yinan Xia, Yi Ding, Changqing Zhang, Qinghua Hu
Abstract: Multimodal fusion is crucial in joint decision-making systems for rendering holistic judgments. Since multimodal data changes in open environments, dynamic fusion has emerged and achieved remarkable progress in numerous applications. However, most existing dynamic multimodal fusion methods lack theoretical guarantees and easily fall into suboptimal problems, yielding unreliability and instability. To address this issue, we propose a Predictive Dynamic Fusion (PDF) framework for multimodal learning. We proceed to reveal the multimodal fusion from a generalization perspective and theoretically derive the predictable Collaborative Belief (Co-Belief) with Mono- and Holo-Confidence, which provably reduces the upper bound of generalization error. Accordingly, we further propose a relative calibration strategy to calibrate the predicted Co-Belief for potential uncertainty. Extensive experiments on multiple benchmarks confirm our superiority. Our code is available at https://github.com/Yinan-Xia/PDF.
Authors: Jason Yoo, Dylan Green, Geoff Pleiss, Frank Wood
Abstract: Diffusion models have shown exceptional capabilities in generating realistic videos. Yet, their training has been predominantly confined to offline environments where models can repeatedly train on i.i.d. data to convergence. This work explores the feasibility of training diffusion models from a semantically continuous video stream, where correlated video frames sequentially arrive one at a time. To investigate this, we introduce two novel continual video generative modeling benchmarks, Lifelong Bouncing Balls and Windows 95 Maze Screensaver, each containing over a million video frames generated from navigating stationary environments. Surprisingly, our experiments show that diffusion models can be effectively trained online using experience replay, achieving performance comparable to models trained with i.i.d. samples given the same number of gradient steps.
Authors: Belen Esther Aleman, Moises Diaz, Miguel Angel Ferrer
Abstract: Handwriting is a complex task that involves the coordination of motor, perceptual and cognitive skills. It is a fundamental skill for the cognitive and academic development of children. However, the technological, and educational changes in recent decades have affected both the teaching and assessment of handwriting. This paper presents a literature review of handwriting analysis in children, including a bibliometric analysis of published articles, the study participants, and the methods of evaluating the graphonometric state of children. The aim is to synthesize the state of the art and provide an overview of the main study trends over the last decade. The review concludes that handwriting remains a fundamental tool for early estimation of cognitive problems and early intervention. The article analyzes graphonometric evaluation tools. Likewise, it reflects on the importance of graphonometric evaluation as a means to detect possible difficulties or disorders in learning to write. The article concludes by highlighting the need to agree on an evaluation methodology and to combine databases.
Authors: Ke Meng, Kai Chen
Abstract: Numerous techniques have been meticulously designed to achieve optimal architectures for convolutional neural networks (CNNs), yet a comparable focus on vision transformers (ViTs) has been somewhat lacking. Despite the remarkable success of ViTs in various vision tasks, their heavyweight nature presents challenges of computational costs. In this paper, we leverage the Gaussian process to systematically explore the nonlinear and uncertain relationship between performance and global architecture factors of MobileViT, such as resolution, width, and depth including the depth of in-verted residual blocks and the depth of ViT blocks, and joint factors including resolution-depth and resolution-width. We present design principles twisting magic 4D cube of the global architecture factors that minimize model sizes and computational costs with higher model accuracy. We introduce a formula for downsizing architectures by iteratively deriving smaller MobileViT V2, all while adhering to a specified constraint of multiply-accumulate operations (MACs). Experiment results show that our formula significantly outperforms CNNs and mobile ViTs across diversified datasets
Authors: Zijia An, Boyu Diao, Libo Huang, Ruiqi Liu, Zhulin An, Yongjun Xu
Abstract: Incremental object detection aims to simultaneously maintain old-class accuracy and detect emerging new-class objects in incremental data. Most existing distillation-based methods underperform when unlabeled old-class objects are absent in the incremental dataset. While the absence can be mitigated by generating old-class samples, it also incurs high computational costs. In this paper, we argue that the extra computational cost stems from the inconsistency between the detector and the generative model, along with redundant generation. To overcome this problem, we propose Efficient Generated Object Replay (EGOR). Specifically, we generate old-class samples by inversing the original detectors, thus eliminating the necessity of training and storing additional generative models. We also propose augmented replay to reuse the objects in generated samples, thereby reducing the redundant generation. In addition, we propose high-response knowledge distillation focusing on the knowledge related to the old class, which transfers the knowledge in generated objects to the incremental detector. With the addition of the generated objects and losses, we observe a bias towards old classes in the detector. We balance the losses for old and new classes to alleviate the bias, thereby increasing the overall detection accuracy. Extensive experiments conducted on MS COCO 2017 demonstrate that our method can efficiently improve detection performance in the absence of old-class objects.
Authors: Feiyu Pan, Hao Fang, Xiankai Lu
Abstract: Referring video object segmentation (RVOS) relies on natural language expressions to segment target objects in video, emphasizing modeling dense text-video relations. The current RVOS methods typically use independently pre-trained vision and language models as backbones, resulting in a significant domain gap between video and text. In cross-modal feature interaction, text features are only used as query initialization and do not fully utilize important information in the text. In this work, we propose using frozen pre-trained vision-language models (VLM) as backbones, with a specific emphasis on enhancing cross-modal feature interaction. Firstly, we use frozen convolutional CLIP backbone to generate feature-aligned vision and text features, alleviating the issue of domain gap and reducing training costs. Secondly, we add more cross-modal feature fusion in the pipeline to enhance the utilization of multi-modal information. Furthermore, we propose a novel video query initialization method to generate higher quality video queries. Without bells and whistles, our method achieved 51.5 J&F on the MeViS test set and ranked 3rd place for MeViS Track in CVPR 2024 PVUW workshop: Motion Expression guided Video Segmentation.
Authors: Yuhao Li, Muzammal Naseer, Jiale Cao, Yu Zhu, Jinqiu Sun, Yanning Zhang, Fahad Shahbaz Khan
Abstract: Most existing multi-object tracking methods typically learn visual tracking features via maximizing dis-similarities of different instances and minimizing similarities of the same instance. While such a feature learning scheme achieves promising performance, learning discriminative features solely based on visual information is challenging especially in case of environmental interference such as occlusion, blur and domain variance. In this work, we argue that multi-modal language-driven features provide complementary information to classical visual features, thereby aiding in improving the robustness to such environmental interference. To this end, we propose a new multi-object tracking framework, named LG-MOT, that explicitly leverages language information at different levels of granularity (scene-and instance-level) and combines it with standard visual features to obtain discriminative representations. To develop LG-MOT, we annotate existing MOT datasets with scene-and instance-level language descriptions. We then encode both instance-and scene-level language information into high-dimensional embeddings, which are utilized to guide the visual features during training. At inference, our LG-MOT uses the standard visual features without relying on annotated language descriptions. Extensive experiments on three benchmarks, MOT17, DanceTrack and SportsMOT, reveal the merits of the proposed contributions leading to state-of-the-art performance. On the DanceTrack test set, our LG-MOT achieves an absolute gain of 2.2\% in terms of target object association (IDF1 score), compared to the baseline using only visual features. Further, our LG-MOT exhibits strong cross-domain generalizability. The dataset and code will be available at ~\url{https://github.com/WesLee88524/LG-MOT}.
Authors: Aarya Patel, Hamid Laga, Ojaswa Sharma
Abstract: Neural implicit representations have emerged as a powerful paradigm for 3D reconstruction. However, despite their success, existing methods fail to capture fine geometric details and thin structures, especially in scenarios where only sparse RGB views of the objects of interest are available. We hypothesize that current methods for learning neural implicit representations from RGB or RGBD images produce 3D surfaces with missing parts and details because they only rely on 0-order differential properties, i.e. the 3D surface points and their projections, as supervisory signals. Such properties, however, do not capture the local 3D geometry around the points and also ignore the interactions between points. This paper demonstrates that training neural representations with first-order differential properties, i.e. surface normals, leads to highly accurate 3D surface reconstruction even in situations where only as few as two RGB (front and back) images are available. Given multiview RGB images of an object of interest, we first compute the approximate surface normals in the image space using the gradient of the depth maps produced using an off-the-shelf monocular depth estimator such as Depth Anything model. An implicit surface regressor is then trained using a loss function that enforces the first-order differential properties of the regressed surface to match those estimated from Depth Anything. Our extensive experiments on a wide range of real and synthetic datasets show that the proposed method achieves an unprecedented level of reconstruction accuracy even when using as few as two RGB views. The detailed ablation study also demonstrates that normal-based supervision plays a key role in this significant improvement in performance, enabling the 3D reconstruction of intricate geometric details and thin structures that were previously challenging to capture.
Authors: Tanvir Mahmud, Mustafa Munir, Radu Marculescu, Diana Marculescu
Abstract: Video-to-video synthesis models face significant challenges, such as ensuring consistent character generation across frames, maintaining smooth temporal transitions, and preserving quality during fast motion. The introduction of joint fully cross-frame self-attention mechanisms has improved character consistency, but this comes at the cost of increased computational complexity. This full cross-frame self-attention mechanism also incorporates redundant details and limits the number of frames that can be jointly edited due to its computational cost. Moreover, the lack of frames in cross-frame attention adversely affects temporal consistency and visual quality. To address these limitations, we propose a new adaptive motion-guided cross-frame attention mechanism that drastically reduces complexity while preserving semantic details and temporal consistency. Specifically, we selectively incorporate the moving regions of successive frames in cross-frame attention and sparsely include stationary regions based on optical flow sampling. This technique allows for an increased number of jointly edited frames without additional computational overhead. For longer duration of video editing, existing methods primarily focus on frame interpolation or flow-warping from jointly edited keyframes, which often results in blurry frames or reduced temporal consistency. To improve this, we introduce KV-caching of jointly edited frames and reuse the same KV across all intermediate frames, significantly enhancing both intermediate frame quality and temporal consistency. Overall, our motion-sampling method enables the use of around three times more keyframes than existing joint editing methods while maintaining superior prediction quality. Ada-VE achieves up to 4x speed-up when using fully-extended self-attention across 40 frames for joint editing, without compromising visual quality or temporal consistency.
Authors: Xiaobiao Du, Haiyang Sun, Shuyun Wang, Zhuojie Wu, Hongwei Sheng, Jiaying Ying, Ming Lu, Tianqing Zhu, Kun Zhan, Xin Yu
Abstract: 3D cars are commonly used in self-driving systems, virtual/augmented reality, and games. However, existing 3D car datasets are either synthetic or low-quality, presenting a significant gap toward the high-quality real-world 3D car datasets and limiting their applications in practical scenarios. In this paper, we propose the first large-scale 3D real car dataset, termed 3DRealCar, offering three distinctive features. (1) \textbf{High-Volume}: 2,500 cars are meticulously scanned by 3D scanners, obtaining car images and point clouds with real-world dimensions; (2) \textbf{High-Quality}: Each car is captured in an average of 200 dense, high-resolution 360-degree RGB-D views, enabling high-fidelity 3D reconstruction; (3) \textbf{High-Diversity}: The dataset contains various cars from over 100 brands, collected under three distinct lighting conditions, including reflective, standard, and dark. Additionally, we offer detailed car parsing maps for each instance to promote research in car parsing tasks. Moreover, we remove background point clouds and standardize the car orientation to a unified axis for the reconstruction only on cars without background and controllable rendering. We benchmark 3D reconstruction results with state-of-the-art methods across each lighting condition in 3DRealCar. Extensive experiments demonstrate that the standard lighting condition part of 3DRealCar can be used to produce a large number of high-quality 3D cars, improving various 2D and 3D tasks related to cars. Notably, our dataset brings insight into the fact that recent 3D reconstruction methods face challenges in reconstructing high-quality 3D cars under reflective and dark lighting conditions. \textcolor{red}{\href{https://xiaobiaodu.github.io/3drealcar/}{Our dataset is available here.}}
Authors: Abisek Rajakumar Kalarani, Pushpak Bhattacharyya, Sumit Shekhar
Abstract: Metaphors are a common communication tool used in our day-to-day life. The detection and generation of metaphors in textual form have been studied extensively but metaphors in other forms have been under-explored. Recent studies have shown that Vision-Language (VL) models cannot understand visual metaphors in memes and adverts. As of now, no probing studies have been done that involve complex language phenomena like metaphors with videos. Hence, we introduce a new VL task of describing the metaphors present in the videos in our work. To facilitate this novel task, we construct and release a manually created dataset with 705 videos and 2115 human-written captions, along with a new metric called Average Concept Distance (ACD), to automatically evaluate the creativity of the metaphors generated. We also propose a novel low-resource video metaphor captioning system: GIT-LLaVA, which obtains comparable performance to SoTA video language models on the proposed task. We perform a comprehensive analysis of existing video language models on this task and publish our dataset, models, and benchmark results to enable further research.
Authors: Lianghan Zhu, Yanqi Bao, Jing Huo, Jing Wu, Yu-Kun Lai, Wenbin Li, Yang Gao
Abstract: The burgeoning field of text-based video generation (T2V) has reignited significant interest in the research of controllable video editing. Although pre-trained T2V-based editing models have achieved efficient editing capabilities, current works are still plagued by two major challenges. Firstly, the inherent limitations of T2V models lead to content inconsistencies and motion discontinuities between frames. Secondly, the notorious issue of over-editing significantly disrupts areas that are intended to remain unaltered. To address these challenges, our work aims to explore a robust video-based editing paradigm based on score distillation. Specifically, we propose an Adaptive Sliding Score Distillation strategy, which not only enhances the stability of T2V supervision but also incorporates both global and local video guidance to mitigate the impact of generation errors. Additionally, we modify the self-attention layers during the editing process to further preserve the key features of the original video. Extensive experiments demonstrate that these strategies enable us to effectively address the aforementioned challenges, achieving superior editing performance compared to existing state-of-the-art methods.
Authors: Bingchen Zhao, Nico Lang, Serge Belongie, Oisin Mac Aodha
Abstract: Category discovery methods aim to find novel categories in unlabeled visual data. At training time, a set of labeled and unlabeled images are provided, where the labels correspond to the categories present in the images. The labeled data provides guidance during training by indicating what types of visual properties and features are relevant for performing discovery in the unlabeled data. As a result, changing the categories present in the labeled set can have a large impact on what is ultimately discovered in the unlabeled set. Despite its importance, the impact of labeled data selection has not been explored in the category discovery literature to date. We show that changing the labeled data can significantly impact discovery performance. Motivated by this, we propose two new approaches for automatically selecting the most suitable labeled data based on the similarity between the labeled and unlabeled data. Our observation is that, unlike in conventional supervised transfer learning, the best labeled is neither too similar, nor too dissimilar, to the unlabeled categories. Our resulting approaches obtains state-of-the-art discovery performance across a range of challenging fine-grained benchmark datasets.
Authors: Liting Huang, Zhihao Zhang, Yiran Zhang, Xiyue Zhou, Shoujin Wang
Abstract: The recent advancements in generative AI models, which can create realistic and human-like content, are significantly transforming how people communicate, create, and work. While the appropriate use of generative AI models can benefit the society, their misuse poses significant threats to data reliability and authentication. However, due to a lack of aligned multimodal datasets, effective and robust methods for detecting machine-generated content are still in the early stages of development. In this paper, we introduce RU-AI, a new large-scale multimodal dataset designed for the robust and efficient detection of machine-generated content in text, image, and voice. Our dataset is constructed from three large publicly available datasets: Flickr8K, COCO, and Places205, by combining the original datasets and their corresponding machine-generated pairs. Additionally, experimental results show that our proposed unified model, which incorporates a multimodal embedding module with a multilayer perceptron network, can effectively determine the origin of the data (i.e., original data samples or machine-generated ones) from RU-AI. However, future work is still required to address the remaining challenges posed by RU-AI. The source code and dataset are available at https://github.com/ZhihaoZhang97/RU-AI.
Authors: Ghjulia Sialelli, Torben Peters, Jan D. Wegner, Konrad Schindler
Abstract: Accurate estimates of Above Ground Biomass (AGB) are essential in addressing two of humanity's biggest challenges, climate change and biodiversity loss. Existing datasets for AGB estimation from satellite imagery are limited. Either they focus on specific, local regions at high resolution, or they offer global coverage at low resolution. There is a need for a machine learning-ready, globally representative, high-resolution benchmark. Our findings indicate significant variability in biomass estimates across different vegetation types, emphasizing the necessity for a dataset that accurately captures global diversity. To address these gaps, we introduce a comprehensive new dataset that is globally distributed, covers a range of vegetation types, and spans several years. This dataset combines AGB reference data from the GEDI mission with data from Sentinel-2 and PALSAR-2 imagery. Additionally, it includes pre-processed high-level features such as a dense canopy height map, an elevation map, and a land-cover classification map. We also produce a dense, high-resolution (10m) map of AGB predictions for the entire area covered by the dataset. Rigorously tested, our dataset is accompanied by several benchmark models and is publicly available. It can be easily accessed using a single line of code, offering a solid basis for efforts towards global AGB estimation. The GitHub repository github.com/ghjuliasialelli/AGBD serves as a one-stop shop for all code and data.
Authors: Tanvir Mahmud, Shentong Mo, Yapeng Tian, Diana Marculescu
Abstract: Recent advances in pre-trained vision transformers have shown promise in parameter-efficient audio-visual learning without audio pre-training. However, few studies have investigated effective methods for aligning multimodal features in parameter-efficient audio-visual transformers. In this paper, we propose MA-AVT, a new parameter-efficient audio-visual transformer employing deep modality alignment for corresponding multimodal semantic features. Specifically, we introduce joint unimodal and multimodal token learning for aligning the two modalities with a frozen modality-shared transformer. This allows the model to learn separate representations for each modality, while also attending to the cross-modal relationships between them. In addition, unlike prior work that only aligns coarse features from the output of unimodal encoders, we introduce blockwise contrastive learning to align coarse-to-fine-grain hierarchical features throughout the encoding phase. Furthermore, to suppress the background features in each modality from foreground matched audio-visual features, we introduce a robust discriminative foreground mining scheme. Through extensive experiments on benchmark AVE, VGGSound, and CREMA-D datasets, we achieve considerable performance improvements over SOTA methods.
Authors: Lanzino Romeo, Fontana Federico, Diko Anxhelo, Marini Marco Raoul, Cinque Luigi
Abstract: Deepfake detection aims to contrast the spread of deep-generated media that undermines trust in online content. While existing methods focus on large and complex models, the need for real-time detection demands greater efficiency. With this in mind, unlike previous work, we introduce a novel deepfake detection approach on images using Binary Neural Networks (BNNs) for fast inference with minimal accuracy loss. Moreover, our method incorporates Fast Fourier Transform (FFT) and Local Binary Pattern (LBP) as additional channel features to uncover manipulation traces in frequency and texture domains. Evaluations on COCOFake, DFFD, and CIFAKE datasets demonstrate our method's state-of-the-art performance in most scenarios with a significant efficiency gain of up to a $20\times$ reduction in FLOPs during inference. Finally, by exploring BNNs in deepfake detection to balance accuracy and efficiency, this work paves the way for future research on efficient deepfake detection.
Authors: Ahc\`ene Boubekki, Samuel G. Fadel, Sebastian Mair
Abstract: Saliency methods have become standard in the explanation toolkit of deep neural networks. Recent developments specific to image classifiers have investigated region-based explanations with either new methods or by adapting well-established ones using ad-hoc superpixel algorithms. In this paper, we aim to avoid relying on these segmenters by extracting a segmentation from the activations of a deep neural network image classifier without fine-tuning the network. Our so-called Neuro-Activated Superpixels (NAS) can isolate the regions of interest in the input relevant to the model's prediction, which boosts high-threshold weakly supervised object localization performance. This property enables the semi-supervised semantic evaluation of saliency methods. The aggregation of NAS with existing saliency methods eases their interpretation and reveals the inconsistencies of the widely used area under the relevance curve metric.
Authors: Wei Qian, Qi Li, Kun Li, Xinke Wang, Xiao Sun, Meng Wang, Dan Guo
Abstract: This paper briefly introduces the solutions developed by our team, HFUT-VUT, for Track 1 of self-supervised heart rate measurement in the 3rd Vision-based Remote Physiological Signal Sensing (RePSS) Challenge hosted at IJCAI 2024. The goal is to develop a self-supervised learning algorithm for heart rate (HR) estimation using unlabeled facial videos. To tackle this task, we present two self-supervised HR estimation solutions that integrate spatial-temporal modeling and contrastive learning, respectively. Specifically, we first propose a non-end-to-end self-supervised HR measurement framework based on spatial-temporal modeling, which can effectively capture subtle rPPG clues and leverage the inherent bandwidth and periodicity characteristics of rPPG to constrain the model. Meanwhile, we employ an excellent end-to-end solution based on contrastive learning, aiming to generalize across different scenarios from complementary perspectives. Finally, we combine the strengths of the above solutions through an ensemble strategy to generate the final predictions, leading to a more accurate HR estimation. As a result, our solutions achieved a remarkable RMSE score of 8.85277 on the test dataset, securing \textbf{2nd place} in Track 1 of the challenge.
Authors: Venkanna Babu Guthula, Stefan Oehmcke, Remigio Chilaule, Hui Zhang, Nico Lang, Ankit Kariryaa, Johan Mottelson, Christian Igel
Abstract: As low-quality housing and in particular certain roof characteristics are associated with an increased risk of malaria, classification of roof types based on remote sensing imagery can support the assessment of malaria risk and thereby help prevent the disease. To support research in this area, we release the Nacala-Roof-Material dataset, which contains high-resolution drone images from Mozambique with corresponding labels delineating houses and specifying their roof types. The dataset defines a multi-task computer vision problem, comprising object detection, classification, and segmentation. In addition, we benchmarked various state-of-the-art approaches on the dataset. Canonical U-Nets, YOLOv8, and a custom decoder on pretrained DINOv2 served as baselines. We show that each of the methods has its advantages but none is superior on all tasks, which highlights the potential of our dataset for future research in multi-task learning. While the tasks are closely related, accurate segmentation of objects does not necessarily imply accurate instance separation, and vice versa. We address this general issue by introducing a variant of the deep ordinal watershed (DOW) approach that additionally separates the interior of objects, allowing for improved object delineation and separation. We show that our DOW variant is a generic approach that improves the performance of both U-Net and DINOv2 backbones, leading to a better trade-off between semantic segmentation and instance segmentation.
Authors: Yu-Wen Pao, An-Jie Li
Abstract: In this work, we propose a novel pipeline that combines AdaIN and NeRF for the task of stylized Novel View Synthesis. Compared to previous works, we make the following contributions: 1) We simplify the pipeline. 2) We extend the capabilities of model to handle the multi-style task. 3) We modify the model architecture to perform well on styles with strong brush strokes. 4) We implement style interpolation on the multi-style model, allowing us to control the style between any two styles and the style intensity between the stylized output and the original scene, providing better control over the stylization strength.
Authors: Zihan Gao, Licheng Jiao, Lingling Li, Xu Liu, Fang Liu, Puhua Chen, Yuwei Guo
Abstract: Neural Radiance Fields (NeRF) have been successfully applied in various aerial scenes, yet they face challenges with sparse views due to limited supervision. The acquisition of dense aerial views is often prohibitive, as unmanned aerial vehicles (UAVs) may encounter constraints in perspective range and energy constraints. In this work, we introduce Multiplane Prior guided NeRF (MPNeRF), a novel approach tailored for few-shot aerial scene rendering-marking a pioneering effort in this domain. Our key insight is that the intrinsic geometric regularities specific to aerial imagery could be leveraged to enhance NeRF in sparse aerial scenes. By investigating NeRF's and Multiplane Image (MPI)'s behavior, we propose to guide the training process of NeRF with a Multiplane Prior. The proposed Multiplane Prior draws upon MPI's benefits and incorporates advanced image comprehension through a SwinV2 Transformer, pre-trained via SimMIM. Our extensive experiments demonstrate that MPNeRF outperforms existing state-of-the-art methods applied in non-aerial contexts, by tripling the performance in SSIM and LPIPS even with three views available. We hope our work offers insights into the development of NeRF-based applications in aerial scenes with limited data.
Authors: Chen Liang, Qiang Guo, Chongkai Yu, Chengjing Wu, Ting Liu, Luoqi Liu
Abstract: Pixel-level Video Understanding requires effectively integrating three-dimensional data in both spatial and temporal dimensions to learn accurate and stable semantic information from continuous frames. However, existing advanced models on the VSPW dataset have not fully modeled spatiotemporal relationships. In this paper, we present our solution for the PVUW competition, where we introduce masked video consistency (MVC) based on existing models. MVC enforces the consistency between predictions of masked frames where random patches are withheld. The model needs to learn the segmentation results of the masked parts through the context of images and the relationship between preceding and succeeding frames of the video. Additionally, we employed test-time augmentation, model aggeregation and a multimodal model-based post-processing method. Our approach achieves 67.27% mIoU performance on the VSPW dataset, ranking 2nd place in the PVUW2024 challenge VSS track.
Authors: Jie Deng, Wenhao Chai, Junsheng Huang, Zhonghan Zhao, Qixuan Huang, Mingyan Gao, Jianshu Guo, Shengyu Hao, Wenhao Hu, Jenq-Neng Hwang, Xi Li, Gaoang Wang
Abstract: City scene generation has gained significant attention in autonomous driving, smart city development, and traffic simulation. It helps enhance infrastructure planning and monitoring solutions. Existing methods have employed a two-stage process involving city layout generation, typically using Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), or Transformers, followed by neural rendering. These techniques often exhibit limited diversity and noticeable artifacts in the rendered city scenes. The rendered scenes lack variety, resembling the training images, resulting in monotonous styles. Additionally, these methods lack planning capabilities, leading to less realistic generated scenes. In this paper, we introduce CityCraft, an innovative framework designed to enhance both the diversity and quality of urban scene generation. Our approach integrates three key stages: initially, a diffusion transformer (DiT) model is deployed to generate diverse and controllable 2D city layouts. Subsequently, a Large Language Model(LLM) is utilized to strategically make land-use plans within these layouts based on user prompts and language guidelines. Based on the generated layout and city plan, we utilize the asset retrieval module and Blender for precise asset placement and scene construction. Furthermore, we contribute two new datasets to the field: 1)CityCraft-OSM dataset including 2D semantic layouts of urban areas, corresponding satellite images, and detailed annotations. 2) CityCraft-Buildings dataset, featuring thousands of diverse, high-quality 3D building assets. CityCraft achieves state-of-the-art performance in generating realistic 3D cities.
Authors: Yawen Lu, Dongfang Liu, Qifan Wang, Cheng Han, Yiming Cui, Zhiwen Cao, Xueling Zhang, Yingjie Victor Chen, Heng Fan
Abstract: In this work, we introduce ProMotion, a unified prototypical framework engineered to model fundamental motion tasks. ProMotion offers a range of compelling attributes that set it apart from current task-specific paradigms. We adopt a prototypical perspective, establishing a unified paradigm that harmonizes disparate motion learning approaches. This novel paradigm streamlines the architectural design, enabling the simultaneous assimilation of diverse motion information. We capitalize on a dual mechanism involving the feature denoiser and the prototypical learner to decipher the intricacies of motion. This approach effectively circumvents the pitfalls of ambiguity in pixel-wise feature matching, significantly bolstering the robustness of motion representation. We demonstrate a profound degree of transferability across distinct motion patterns. This inherent versatility reverberates robustly across a comprehensive spectrum of both 2D and 3D downstream tasks. Empirical results demonstrate that ProMotion outperforms various well-known specialized architectures, achieving 0.54 and 0.054 Abs Rel error on the Sintel and KITTI depth datasets, 1.04 and 2.01 average endpoint error on the clean and final pass of Sintel flow benchmark, and 4.30 F1-all error on the KITTI flow benchmark. For its efficacy, we hope our work can catalyze a paradigm shift in universal models in computer vision.
Authors: Lianyu Pang, Jian Yin, Baoquan Zhao, Feize Wu, Fu Lee Wang, Qing Li, Xudong Mao
Abstract: Recent advances in text-to-image models have enabled high-quality personalized image synthesis of user-provided concepts with flexible textual control. In this work, we analyze the limitations of two primary techniques in text-to-image personalization: Textual Inversion and DreamBooth. When integrating the learned concept into new prompts, Textual Inversion tends to overfit the concept, while DreamBooth often overlooks it. We attribute these issues to the incorrect learning of the embedding alignment for the concept. We introduce AttnDreamBooth, a novel approach that addresses these issues by separately learning the embedding alignment, the attention map, and the subject identity in different training stages. We also introduce a cross-attention map regularization term to enhance the learning of the attention map. Our method demonstrates significant improvements in identity preservation and text alignment compared to the baseline methods.
Authors: Zahra Golpayegani, Patrick St-Amant, Nizar Bouguila
Abstract: Deep learning models can perform well when evaluated on images from the same distribution as the training set. However, applying small perturbations in the forms of noise, artifacts, occlusions, blurring, etc. to a model's input image and feeding the model with out-of-distribution (OOD) data can significantly drop the model's accuracy, making it not applicable to real-world scenarios. Data augmentation is one of the well-practiced methods to improve model robustness against OOD data; however, examining which augmentation type to choose and how it affects the OOD robustness remains understudied. There is a growing belief that augmenting datasets using data augmentations that improve a model's bias to shape-based features rather than texture-based features results in increased OOD robustness for Convolutional Neural Networks trained on the ImageNet-1K dataset. This is usually stated as ``an increase in the model's shape bias results in an increase in its OOD robustness". Based on this hypothesis, some works in the literature aim to find augmentations with higher effects on model shape bias and use those for data augmentation. By evaluating 39 types of data augmentations on a widely used OOD dataset, we demonstrate the impact of each data augmentation on the model's robustness to OOD data and further show that the mentioned hypothesis is not true; an increase in shape bias does not necessarily result in higher OOD robustness. By analyzing the results, we also find some biases in the ImageNet-1K dataset that can easily be reduced using proper data augmentation. Our evaluation results further show that there is not necessarily a trade-off between in-domain accuracy and OOD robustness, and choosing the proper augmentations can help increase both in-domain accuracy and OOD robustness simultaneously.
Authors: Shakhnaz Akhmedova, Nils K\"orber
Abstract: Generative adversarial networks (GANs) are machine learning models that are used to estimate the underlying statistical structure of a given dataset and as a result can be used for a variety of tasks such as image generation or anomaly detection. Despite their initial simplicity, designing an effective loss function for training GANs remains challenging, and various loss functions have been proposed aiming to improve the performance and stability of the generative models. In this study, loss function design for GANs is presented as an optimization problem solved using the genetic programming (GP) approach. Initial experiments were carried out using small Deep Convolutional GAN (DCGAN) model and the MNIST dataset, in order to search experimentally for an improved loss function. The functions found were evaluated on CIFAR10, with the best function, named GANetic loss, showing exceptionally better performance and stability compared to the losses commonly used for GAN training. To further evalute its general applicability on more challenging problems, GANetic loss was applied for two medical applications: image generation and anomaly detection. Experiments were performed with histopathological, gastrointestinal or glaucoma images to evaluate the GANetic loss in medical image generation, resulting in improved image quality compared to the baseline models. The GANetic Loss used for polyp and glaucoma images showed a strong improvement in the detection of anomalies. In summary, the GANetic loss function was evaluated on multiple datasets and applications where it consistently outperforms alternative loss functions. Moreover, GANetic loss leads to stable training and reproducible results, a known weak spot of GANs.
Authors: Shentong Mo
Abstract: Recent advancements in sequence modeling have led to the development of the Mamba architecture, noted for its selective state space approach, offering a promising avenue for efficient long sequence handling. However, its application in 3D shape generation, particularly at high resolutions, remains underexplored. Traditional diffusion transformers (DiT) with self-attention mechanisms, despite their potential, face scalability challenges due to the cubic complexity of attention operations as input length increases. This complexity becomes a significant hurdle when dealing with high-resolution voxel sizes. To address this challenge, we introduce a novel diffusion architecture tailored for 3D point clouds generation-Diffusion Mamba (DiM-3D). This architecture forgoes traditional attention mechanisms, instead utilizing the inherent efficiency of the Mamba architecture to maintain linear complexity with respect to sequence length. DiM-3D is characterized by fast inference times and substantially lower computational demands, quantified in reduced Gflops, thereby addressing the key scalability issues of prior models. Our empirical results on the ShapeNet benchmark demonstrate that DiM-3D achieves state-of-the-art performance in generating high-fidelity and diverse 3D shapes. Additionally, DiM-3D shows superior capabilities in tasks like 3D point cloud completion. This not only proves the model's scalability but also underscores its efficiency in generating detailed, high-resolution voxels necessary for advanced 3D shape modeling, particularly excelling in environments requiring high-resolution voxel sizes. Through these findings, we illustrate the exceptional scalability and efficiency of the Diffusion Mamba framework in 3D shape generation, setting a new standard for the field and paving the way for future explorations in high-resolution 3D modeling technologies.
Authors: Yani Zhang, Dongming Wu, Wencheng Han, Xingping Dong
Abstract: Referring multi-object tracking (RMOT) aims at detecting and tracking multiple objects following human instruction represented by a natural language expression. Existing RMOT benchmarks are usually formulated through manual annotations, integrated with static regulations. This approach results in a dearth of notable diversity and a constrained scope of implementation. In this work, our key idea is to bootstrap the task of referring multi-object tracking by introducing discriminative language words as much as possible. In specific, we first develop Refer-KITTI into a large-scale dataset, named Refer-KITTI-V2. It starts with 2,719 manual annotations, addressing the issue of class imbalance and introducing more keywords to make it closer to real-world scenarios compared to Refer-KITTI. They are further expanded to a total of 9,758 annotations by prompting large language models, which create 617 different words, surpassing previous RMOT benchmarks. In addition, the end-to-end framework in RMOT is also bootstrapped by a simple yet elegant temporal advancement strategy, which achieves better performance than previous approaches. The source code and dataset is available at https://github.com/zyn213/TempRMOT.
Authors: Yumin Zhang, Hongliu Li, Yajun Gao, Haoran Duan, Yawen Huang, Yefeng Zheng
Abstract: Few-shot medical image segmentation has achieved great progress in improving accuracy and efficiency of medical analysis in the biomedical imaging field. However, most existing methods cannot explore inter-class relations among base and novel medical classes to reason unseen novel classes. Moreover, the same kind of medical class has large intra-class variations brought by diverse appearances, shapes and scales, thus causing ambiguous visual characterization to degrade generalization performance of these existing methods on unseen novel classes. To address the above challenges, in this paper, we propose a \underline{\textbf{P}}rototype correlation \underline{\textbf{M}}atching and \underline{\textbf{C}}lass-relation \underline{\textbf{R}}easoning (i.e., \textbf{PMCR}) model. The proposed model can effectively mitigate false pixel correlation matches caused by large intra-class variations while reasoning inter-class relations among different medical classes. Specifically, in order to address false pixel correlation match brought by large intra-class variations, we propose a prototype correlation matching module to mine representative prototypes that can characterize diverse visual information of different appearances well. We aim to explore prototype-level rather than pixel-level correlation matching between support and query features via optimal transport algorithm to tackle false matches caused by intra-class variations. Meanwhile, in order to explore inter-class relations, we design a class-relation reasoning module to segment unseen novel medical objects via reasoning inter-class relations between base and novel classes. Such inter-class relations can be well propagated to semantic encoding of local query features to improve few-shot segmentation performance. Quantitative comparisons illustrates the large performance improvement of our model over other baseline methods.
Authors: Chaerin Min, Srinath Sridhar
Abstract: Grasping is an important human activity that has long been studied in robotics, computer vision, and cognitive science. Most existing works study grasping from the perspective of synthesizing hand poses conditioned on 3D or 2D object representations. We propose GenHeld to address the inverse problem of synthesizing held objects conditioned on 3D hand model or 2D image. Given a 3D model of hand, GenHeld 3D can select a plausible held object from a large dataset using compact object representations called object codes.The selected object is then positioned and oriented to form a plausible grasp without changing hand pose. If only a 2D hand image is available, GenHeld 2D can edit this image to add or replace a held object. GenHeld 2D operates by combining the abilities of GenHeld 3D with diffusion-based image editing. Results and experiments show that we outperform baselines and can generate plausible held objects in both 2D and 3D. Our experiments demonstrate that our method achieves high quality and plausibility of held object synthesis in both 3D and 2D.
Authors: Benjamin Fresz, Lena L\"orcher, Marco Huber
Abstract: Decision processes of computer vision models - especially deep neural networks - are opaque in nature, meaning that these decisions cannot be understood by humans. Thus, over the last years, many methods to provide human-understandable explanations have been proposed. For image classification, the most common group are saliency methods, which provide (super-)pixelwise feature attribution scores for input images. But their evaluation still poses a problem, as their results cannot be simply compared to the unknown ground truth. To overcome this, a slew of different proxy metrics have been defined, which are - as the explainability methods themselves - often built on intuition and thus, are possibly unreliable. In this paper, new evaluation metrics for saliency methods are developed and common saliency methods are benchmarked on ImageNet. In addition, a scheme for reliability evaluation of such metrics is proposed that is based on concepts from psychometric testing. The used code can be found at https://github.com/lelo204/ClassificationMetricsForImageExplanations .
URLs: https://github.com/lelo204/ClassificationMetricsForImageExplanations
Authors: Chinmaya Devaraj, Cornelia Fermuller, Yiannis Aloimonos
Abstract: Videos are more informative than images because they capture the dynamics of the scene. By representing motion in videos, we can capture dynamic activities. In this work, we introduce GPT-4 generated motion descriptions that capture fine-grained motion descriptions of activities and apply them to three action datasets. We evaluated several video-text models on the task of retrieval of motion descriptions. We found that they fall far behind human expert performance on two action datasets, raising the question of whether video-text models understand motion in videos. To address it, we introduce a method of improving motion understanding in video-text models by utilizing motion descriptions. This method proves to be effective on two action datasets for the motion description retrieval task. The results draw attention to the need for quality captions involving fine-grained motion information in existing datasets and demonstrate the effectiveness of the proposed pipeline in understanding fine-grained motion during video-text retrieval.
Authors: Xingrui Wang, Xin Li, Zhibo Chen
Abstract: Tuning-free long video diffusion has been proposed to generate extended-duration videos with enriched content by reusing the knowledge from pre-trained short video diffusion model without retraining. However, most works overlook the fine-grained long-term video consistency modeling, resulting in limited scene consistency (i.e., unreasonable object or background transitions), especially with multiple text inputs. To mitigate this, we propose the Consistency Noise Injection, dubbed CoNo, which introduces the "look-back" mechanism to enhance the fine-grained scene transition between different video clips, and designs the long-term consistency regularization to eliminate the content shifts when extending video contents through noise prediction. In particular, the "look-back" mechanism breaks the noise scheduling process into three essential parts, where one internal noise prediction part is injected into two video-extending parts, intending to achieve a fine-grained transition between two video clips. The long-term consistency regularization focuses on explicitly minimizing the pixel-wise distance between the predicted noises of the extended video clip and the original one, thereby preventing abrupt scene transitions. Extensive experiments have shown the effectiveness of the above strategies by performing long-video generation under both single- and multi-text prompt conditions. The project has been available in https://wxrui182.github.io/CoNo.github.io/.
Authors: Christian Giannetti
Abstract: Rainfall estimation through the analysis of its impact on electromagnetic waves has sparked increasing interest in the research community. Recent studies have delved into its effects on cellular network performance, demonstrating the potential to forecast rainfall levels based on electromagnetic wave attenuation during precipitations. This paper aims to solve the problem of identifying the nature of specific weather phenomena from the received signal level (RSL) in 4G/LTE mobile terminals. Specifically, utilizing time-series data representing RSL, we propose a novel approach to encode time series as images and model the task as an image classification problem, which we finally address using convolutional neural networks (CNNs). The main benefit of the abovementioned procedure is the opportunity to utilize various data augmentation techniques simultaneously. This encompasses applying traditional approaches, such as moving averages, to the time series and enhancing the generated images. We have investigated various image data augmentation methods to identify the most effective combination for this scenario. In the upcoming sections, we will introduce the task of rainfall estimation and conduct a comprehensive analysis of the dataset used. Subsequently, we will formally propose a new approach for converting time series into images. To conclude, the paper's final section will present and discuss the experiments conducted, providing the reader with a brief yet comprehensive overview of the results.
Authors: Lukas Helff, Felix Friedrich, Manuel Brack, Kristian Kersting, Patrick Schramowski
Abstract: We introduce LlavaGuard, a family of VLM-based safeguard models, offering a versatile framework for evaluating the safety compliance of visual content. Specifically, we designed LlavaGuard for dataset annotation and generative model safeguarding. To this end, we collected and annotated a high-quality visual dataset incorporating a broad safety taxonomy, which we use to tune VLMs on context-aware safety risks. As a key innovation, LlavaGuard's new responses contain comprehensive information, including a safety rating, the violated safety categories, and an in-depth rationale. Further, our introduced customizable taxonomy categories enable the context-specific alignment of LlavaGuard to various scenarios. Our experiments highlight the capabilities of LlavaGuard in complex and real-world applications. We provide checkpoints ranging from 7B to 34B parameters demonstrating state-of-the-art performance, with even the smallest models outperforming baselines like GPT-4. We make our dataset and model weights publicly available and invite further research to address the diverse needs of communities and contexts.
Authors: Shruti Joshi, Aiswarya Akumalla, Seth Haney, Maxim Bazhenov
Abstract: Mammalian brains handle complex reasoning by integrating information across brain regions specialized for particular sensory modalities. This enables improved robustness and generalization versus deep neural networks, which typically process one modality and are vulnerable to perturbations. While defense methods exist, they do not generalize well across perturbations. We developed a fusion model combining background and foreground features from CNNs trained on Imagenet and Places365. We tested its robustness to human-perceivable perturbations on MS COCO. The fusion model improved robustness, especially for classes with greater context variability. Our proposed solution for integrating multiple modalities provides a new approach to enhance robustness and may be complementary to existing methods.
Authors: Shengqiong Wu, Hao Fei, Xiangtai Li, Jiayi Ji, Hanwang Zhang, Tat-Seng Chua, Shuicheng Yan
Abstract: Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in processing vision-language tasks. One of the crux of MLLMs lies in vision tokenization, which involves efficiently transforming input visual signals into feature representations that are most beneficial for LLMs. However, existing vision tokenizers, essential for semantic alignment between vision and language, remain problematic. Existing methods aggressively fragment visual input, corrupting the visual semantic integrity. To address this, this paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok), which groups visual features into semantic units via a dynamic clustering algorithm, flexibly determining the number of tokens based on image complexity. The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features. The proposed MLLM (Setokim) equipped with SeTok significantly demonstrates superior performance across various tasks, as evidenced by our experimental results. The project page is at https://chocowu.github.io/SeTok-web/.
Authors: Zahra Golpayegani, Nizar Bouguila
Abstract: Storing data is particularly a challenge when dealing with image data which often involves large file sizes due to the high resolution and complexity of images. Efficient image compression algorithms are crucial to better manage data storage costs. In this paper, we propose a novel region-based lossy image compression technique, called PatchSVD, based on the Singular Value Decomposition (SVD) algorithm. We show through experiments that PatchSVD outperforms SVD-based image compression with respect to three popular image compression metrics. Moreover, we compare PatchSVD compression artifacts with those of Joint Photographic Experts Group (JPEG) and SVD-based image compression and illustrate some cases where PatchSVD compression artifacts are preferable compared to JPEG and SVD artifacts.
Authors: Keyhan Najafian, Farhad Maleki, Ian Stavness, Lingling Jin
Abstract: Video object segmentation approaches primarily rely on large-scale pixel-accurate human-annotated datasets for model development. In Dense Video Object Segmentation (DVOS) scenarios, each video frame encompasses hundreds of small, dense, and partially occluded objects. Accordingly, the labor-intensive manual annotation of even a single frame often takes hours, which hinders the development of DVOS for many applications. Furthermore, in videos with dense patterns, following a large number of objects that move in different directions poses additional challenges. To address these challenges, we proposed a semi-self-supervised spatiotemporal approach for DVOS utilizing a diffusion-based method through multi-task learning. Emulating real videos' optical flow and simulating their motion, we developed a methodology to synthesize computationally annotated videos that can be used for training DVOS models; The model performance was further improved by utilizing weakly labeled (computationally generated but imprecise) data. To demonstrate the utility and efficacy of the proposed approach, we developed DVOS models for wheat head segmentation of handheld and drone-captured videos, capturing wheat crops in fields of different locations across various growth stages, spanning from heading to maturity. Despite using only a few manually annotated video frames, the proposed approach yielded high-performing models, achieving a Dice score of 0.82 when tested on a drone-captured external test set. While we showed the efficacy of the proposed approach for wheat head segmentation, its application can be extended to other crops or DVOS in other domains, such as crowd analysis or microscopic image analysis.
Authors: Jianing Yang, Xuweiyi Chen, Nikhil Madaan, Madhavan Iyengar, Shengyi Qian, David F. Fouhey, Joyce Chai
Abstract: The integration of language and 3D perception is crucial for developing embodied agents and robots that comprehend and interact with the physical world. While large language models (LLMs) have demonstrated impressive language understanding and generation capabilities, their adaptation to 3D environments (3D-LLMs) remains in its early stages. A primary challenge is the absence of large-scale datasets that provide dense grounding between language and 3D scenes. In this paper, we introduce 3D-GRAND, a pioneering large-scale dataset comprising 40,087 household scenes paired with 6.2 million densely-grounded scene-language instructions. Our results show that instruction tuning with 3D-GRAND significantly enhances grounding capabilities and reduces hallucinations in 3D-LLMs. As part of our contributions, we propose a comprehensive benchmark 3D-POPE to systematically evaluate hallucination in 3D-LLMs, enabling fair comparisons among future models. Our experiments highlight a scaling effect between dataset size and 3D-LLM performance, emphasizing the critical role of large-scale 3D-text datasets in advancing embodied AI research. Notably, our results demonstrate early signals for effective sim-to-real transfer, indicating that models trained on large synthetic data can perform well on real-world 3D scans. Through 3D-GRAND and 3D-POPE, we aim to equip the embodied AI community with essential resources and insights, setting the stage for more reliable and better-grounded 3D-LLMs. Project website: https://3d-grand.github.io
Authors: Shruthi Bannur, Kenza Bouzid, Daniel C. Castro, Anton Schwaighofer, Sam Bond-Taylor, Maximilian Ilse, Fernando P\'erez-Garc\'ia, Valentina Salvatelli, Harshita Sharma, Felix Meissen, Mercy Ranjit, Shaury Srivastav, Julia Gong, Fabian Falck, Ozan Oktay, Anja Thieme, Matthew P. Lungren, Maria Teodora Wetscherek, Javier Alvarez-Valle, Stephanie L. Hyland
Abstract: Radiology reporting is a complex task that requires detailed image understanding, integration of multiple inputs, including comparison with prior imaging, and precise language generation. This makes it ideal for the development and use of generative multimodal models. Here, we extend report generation to include the localisation of individual findings on the image - a task we call grounded report generation. Prior work indicates that grounding is important for clarifying image understanding and interpreting AI-generated text. Therefore, grounded reporting stands to improve the utility and transparency of automated report drafting. To enable evaluation of grounded reporting, we propose a novel evaluation framework - RadFact - leveraging the reasoning capabilities of large language models (LLMs). RadFact assesses the factuality of individual generated sentences, as well as correctness of generated spatial localisations when present. We introduce MAIRA-2, a large multimodal model combining a radiology-specific image encoder with a LLM, and trained for the new task of grounded report generation on chest X-rays. MAIRA-2 uses more comprehensive inputs than explored previously: the current frontal image, the current lateral image, the prior frontal image and prior report, as well as the Indication, Technique and Comparison sections of the current report. We demonstrate that these additions significantly improve report quality and reduce hallucinations, establishing a new state of the art on findings generation (without grounding) on MIMIC-CXR while demonstrating the feasibility of grounded reporting as a novel and richer task.
Authors: Dongfu Jiang, Max Ku, Tianle Li, Yuansheng Ni, Shizhuo Sun, Rongqi Fan, Wenhu Chen
Abstract: Generative AI has made remarkable strides to revolutionize fields such as image and video generation. These advancements are driven by innovative algorithms, architecture, and data. However, the rapid proliferation of generative models has highlighted a critical gap: the absence of trustworthy evaluation metrics. Current automatic assessments such as FID, CLIP, FVD, etc often fail to capture the nuanced quality and user satisfaction associated with generative outputs. This paper proposes an open platform GenAI-Arena to evaluate different image and video generative models, where users can actively participate in evaluating these models. By leveraging collective user feedback and votes, GenAI-Arena aims to provide a more democratic and accurate measure of model performance. It covers three arenas for text-to-image generation, text-to-video generation, and image editing respectively. Currently, we cover a total of 27 open-source generative models. GenAI-Arena has been operating for four months, amassing over 6000 votes from the community. We describe our platform, analyze the data, and explain the statistical methods for ranking the models. To further promote the research in building model-based evaluation metrics, we release a cleaned version of our preference data for the three tasks, namely GenAI-Bench. We prompt the existing multi-modal models like Gemini, GPT-4o to mimic human voting. We compute the correlation between model voting with human voting to understand their judging abilities. Our results show existing multimodal models are still lagging in assessing the generated visual content, even the best model GPT-4o only achieves a Pearson correlation of 0.22 in the quality subscore, and behaves like random guessing in others.
Authors: Haotian Zhang, Junting Zhou, Haowei Lin, Hang Ye, Jianhua Zhu, Zihao Wang, Liangcai Gao, Yizhou Wang, Yitao Liang
Abstract: Continual Learning (CL) poses a significant challenge in Artificial Intelligence, aiming to mirror the human ability to incrementally acquire knowledge and skills. While extensive research has focused on CL within the context of classification tasks, the advent of increasingly powerful generative models necessitates the exploration of Continual Learning of Generative models (CLoG). This paper advocates for shifting the research focus from classification-based CL to CLoG. We systematically identify the unique challenges presented by CLoG compared to traditional classification-based CL. We adapt three types of existing CL methodologies, replay-based, regularization-based, and parameter-isolation-based methods to generative tasks and introduce comprehensive benchmarks for CLoG that feature great diversity and broad task coverage. Our benchmarks and results yield intriguing insights that can be valuable for developing future CLoG methods. Additionally, we will release a codebase designed to facilitate easy benchmarking and experimentation in CLoG publicly at https://github.com/linhaowei1/CLoG. We believe that shifting the research focus to CLoG will benefit the continual learning community and illuminate the path for next-generation AI-generated content (AIGC) in a lifelong learning paradigm.
Authors: Jongyun Shin, Seunjin Han, Jangho Kim
Abstract: Model agnostic meta-learning (MAML) is one of the most widely used gradient-based meta-learning, consisting of two optimization loops: an inner loop and outer loop. MAML learns the new task from meta-initialization parameters with an inner update and finds the meta-initialization parameters in the outer loop. In general, the injection of noise into the gradient of the model for augmenting the gradient is one of the widely used regularization methods. In this work, we propose a novel cooperative meta-learning framework dubbed CML which leverages gradient-level regularization with gradient augmentation. We inject learnable noise into the gradient of the model for the model generalization. The key idea of CML is introducing the co-learner which has no inner update but the outer loop update to augment gradients for finding better meta-initialization parameters. Since the co-learner does not update in the inner loop, it can be easily deleted after meta-training. Therefore, CML infers with only meta-learner without additional cost and performance degradation. We demonstrate that CML is easily applicable to gradient-based meta-learning methods and CML leads to increased performance in few-shot regression, few-shot image classification and few-shot node classification tasks. Our codes are at https://github.com/JJongyn/CML.
Authors: Qingze Bai, Tiange Liu, Zhi Liu, Yubing Tong, Drew Torigian, Jayaram Udupa
Abstract: In this paper, we present XctDiff, an algorithm framework for reconstructing CT from a single radiograph, which decomposes the reconstruction process into two easily controllable tasks: feature extraction and CT reconstruction. Specifically, we first design a progressive feature extraction strategy that is able to extract robust 3D priors from radiographs. Then, we use the extracted prior information to guide the CT reconstruction in the latent space. Moreover, we design a homogeneous spatial codebook to improve the reconstruction quality further. The experimental results show that our proposed method achieves state-of-the-art reconstruction performance and overcomes the blurring issue. We also apply XctDiff on self-supervised pre-training task. The effectiveness indicates that it has promising additional applications in medical image analysis. The code is available at:https://github.com/qingze-bai/XctDiff
Authors: Yixin Huang, Yiqi Jin, Ke Tao, Kaijian Xia, Jianfeng Gu, Lei Yu, Lan Du, Cunjian Chen
Abstract: May-Thurner Syndrome (MTS), also known as iliac vein compression syndrome or Cockett's syndrome, is a condition potentially impacting over 20 percent of the population, leading to an increased risk of iliofemoral deep venous thrombosis. In this paper, we present a 3D-based deep learning approach called MTS-Net for diagnosing May-Thurner Syndrome using CT scans. To effectively capture the spatial-temporal relationship among CT scans and emulate the clinical process of diagnosing MTS, we propose a novel attention module called the dual-enhanced positional multi-head self-attention (DEP-MHSA). The proposed DEP-MHSA reconsiders the role of positional embedding and incorporates a dual-enhanced positional embedding in both attention weights and residual connections. Further, we establish a new dataset, termed MTS-CT, consisting of 747 subjects. Experimental results demonstrate that our proposed approach achieves state-of-the-art MTS diagnosis results, and our self-attention design facilitates the spatial-temporal modeling. We believe that our DEP-MHSA is more suitable to handle CT image sequence modeling and the proposed dataset enables future research on MTS diagnosis. We make our code and dataset publicly available at: https://github.com/Nutingnon/MTS_dep_mhsa.
Authors: Yiheng Zhang, Yunkang Cao, Xiaohao Xu, Weiming Shen
Abstract: This paper presents LogiCode, a novel framework that leverages Large Language Models (LLMs) for identifying logical anomalies in industrial settings, moving beyond traditional focus on structural inconsistencies. By harnessing LLMs for logical reasoning, LogiCode autonomously generates Python codes to pinpoint anomalies such as incorrect component quantities or missing elements, marking a significant leap forward in anomaly detection technologies. A custom dataset "LOCO-Annotations" and a benchmark "LogiBench" are introduced to evaluate the LogiCode's performance across various metrics including binary classification accuracy, code generation success rate, and precision in reasoning. Findings demonstrate LogiCode's enhanced interpretability, significantly improving the accuracy of logical anomaly detection and offering detailed explanations for identified anomalies. This represents a notable shift towards more intelligent, LLM-driven approaches in industrial anomaly detection, promising substantial impacts on industry-specific applications.
Authors: Yu-Chang Wu, Shen-Huan Lyu, Haopu Shang, Xiangyu Wang, Chao Qian
Abstract: Selective classification enables models to make predictions only when they are sufficiently confident, aiming to enhance safety and reliability, which is important in high-stakes scenarios. Previous methods mainly use deep neural networks and focus on modifying the architecture of classification layers to enable the model to estimate the confidence of its prediction. This work provides a generalization bound for selective classification, disclosing that optimizing feature layers helps improve the performance of selective classification. Inspired by this theory, we propose to explicitly improve the selective classification model at the feature level for the first time, leading to a novel Confidence-aware Contrastive Learning method for Selective Classification, CCL-SC, which similarizes the features of homogeneous instances and differentiates the features of heterogeneous instances, with the strength controlled by the model's confidence. The experimental results on typical datasets, i.e., CIFAR-10, CIFAR-100, CelebA, and ImageNet, show that CCL-SC achieves significantly lower selective risk than state-of-the-art methods, across almost all coverage degrees. Moreover, it can be combined with existing methods to bring further improvement.
Authors: Michelle Espranita Liman, Daniel Rueckert, Florian J. Fintelmann, Philip M\"uller
Abstract: Field-of-view (FOV) recovery of truncated chest CT scans is crucial for accurate body composition analysis, which involves quantifying skeletal muscle and subcutaneous adipose tissue (SAT) on CT slices. This, in turn, enables disease prognostication. Here, we present a method for recovering truncated CT slices using generative image outpainting. We train a diffusion model and apply it to truncated CT slices generated by simulating a small FOV. Our model reliably recovers the truncated anatomy and outperforms the previous state-of-the-art despite being trained on 87% less data.
Authors: Sungho Jeon, Xinyue Ma, Kwang In Kim, Myeongjae Jeon
Abstract: On-device continual learning (CL) requires the co-optimization of model accuracy and resource efficiency to be practical. This is extremely challenging because it must preserve accuracy while learning new tasks with continuously drifting data and maintain both high energy and memory efficiency to be deployable on real-world devices. Typically, a CL method leverages one of two types of backbone networks: CNN or ViT. It is commonly believed that CNN-based CL excels in resource efficiency, whereas ViT-based CL is superior in model performance, making each option attractive only for a single aspect. In this paper, we revisit this comparison while embracing powerful pre-trained ViT models of various sizes, including ViT-Ti (5.8M parameters). Our detailed analysis reveals that many practical options exist today for making ViT-based methods more suitable for on-device CL, even when accuracy, energy, and memory are all considered. To further expand this impact, we introduce REP, which improves resource efficiency specifically targeting prompt-based rehearsal-free methods. Our key focus is on avoiding catastrophic trade-offs with accuracy while trimming computational and memory costs throughout the training process. We achieve this by exploiting swift prompt selection that enhances input data using a carefully provisioned model, and by developing two novel algorithms-adaptive token merging (AToM) and adaptive layer dropping (ALD)-that optimize the prompt updating stage. In particular, AToM and ALD perform selective skipping across the data and model-layer dimensions without compromising task-specific features in vision transformer models. Extensive experiments on three image classification datasets validate REP's superior resource efficiency over current state-of-the-art methods.
Authors: Sojung An, Tae-Jin Oh, Eunha Sohn, Donghyun Kim
Abstract: Deep learning-based time series forecasting has dominated the short-term precipitation forecasting field with the help of its ability to estimate motion flow in high-resolution datasets. The growing interest in precipitation nowcasting offers substantial opportunities for the advancement of current forecasting technologies. Nevertheless, there has been a scarcity of in-depth surveys of time series precipitation forecasting using deep learning. Thus, this paper systemically reviews recent progress in time series precipitation forecasting models. Specifically, we investigate the following key points within background components, covering: i) preprocessing, ii) objective functions, and iii) evaluation metrics. We then categorize forecasting models into \textit{recursive} and \textit{multiple} strategies based on their approaches to predict future frames, investigate the impacts of models using the strategies, and performance assessments. Finally, we evaluate current deep learning-based models for precipitation forecasting on a public benchmark, discuss their limitations and challenges, and present some promising research directions. Our contribution lies in providing insights for a better understanding of time series precipitation forecasting and in aiding the development of robust AI solutions for the future.
Authors: Yuxing Long, Wenzhe Cai, Hongcheng Wang, Guanqi Zhan, Hao Dong
Abstract: Enabling robots to navigate following diverse language instructions in unexplored environments is an attractive goal for human-robot interaction. However, this goal is challenging because different navigation tasks require different strategies. The scarcity of instruction navigation data hinders training an instruction navigation model with varied strategies. Therefore, previous methods are all constrained to one specific type of navigation instruction. In this work, we propose InstructNav, a generic instruction navigation system. InstructNav makes the first endeavor to handle various instruction navigation tasks without any navigation training or pre-built maps. To reach this goal, we introduce Dynamic Chain-of-Navigation (DCoN) to unify the planning process for different types of navigation instructions. Furthermore, we propose Multi-sourced Value Maps to model key elements in instruction navigation so that linguistic DCoN planning can be converted into robot actionable trajectories. With InstructNav, we complete the R2R-CE task in a zero-shot way for the first time and outperform many task-training methods. Besides, InstructNav also surpasses the previous SOTA method by 10.48% on the zero-shot Habitat ObjNav and by 86.34% on demand-driven navigation DDN. Real robot experiments on diverse indoor scenes further demonstrate our method's robustness in coping with the environment and instruction variations.
Authors: Feiyang Wang, Xingquan Zuo, Hai Huang, Gang Chen
Abstract: Many machine learning models are susceptible to adversarial attacks, with decision-based black-box attacks representing the most critical threat in real-world applications. These attacks are extremely stealthy, generating adversarial examples using hard labels obtained from the target machine learning model. This is typically realized by optimizing perturbation directions, guided by decision boundaries identified through query-intensive exact search, significantly limiting the attack success rate. This paper introduces a novel approach using the Approximation Decision Boundary (ADB) to efficiently and accurately compare perturbation directions without precisely determining decision boundaries. The effectiveness of our ADB approach (ADBA) hinges on promptly identifying suitable ADB, ensuring reliable differentiation of all perturbation directions. For this purpose, we analyze the probability distribution of decision boundaries, confirming that using the distribution's median value as ADB can effectively distinguish different perturbation directions, giving rise to the development of the ADBA-md algorithm. ADBA-md only requires four queries on average to differentiate any pair of perturbation directions, which is highly query-efficient. Extensive experiments on six well-known image classifiers clearly demonstrate the superiority of ADBA and ADBA-md over multiple state-of-the-art black-box attacks.
Authors: Dmitry Nechaev, Alexey Pchelnikov, Ekaterina Ivanova
Abstract: Pathology, the microscopic examination of diseased tissue, is critical for diagnosing various medical conditions, particularly cancers. Traditional methods are labor-intensive and prone to human error. Digital pathology, which converts glass slides into high-resolution digital images for analysis by computer algorithms, revolutionizes the field by enhancing diagnostic accuracy, consistency, and efficiency through automated image analysis and large-scale data processing. Foundational transformer pretraining is crucial for developing robust, generalizable models as it enables learning from vast amounts of unannotated data. This paper introduces the Hibou family of foundational vision transformers for pathology, leveraging the DINOv2 framework to pretrain two model variants, Hibou-B and Hibou-L, on a proprietary dataset of over 1 million whole slide images (WSIs) representing diverse tissue types and staining techniques. Our pretrained models demonstrate superior performance on both patch-level and slide-level benchmarks, surpassing existing state-of-the-art methods. Notably, Hibou-L achieves the highest average accuracy across multiple benchmark datasets. To support further research and application in the field, we have open-sourced the Hibou-B model, which can be accessed at https://github.com/HistAI/hibou
Authors: Thomas Decker, Ananta R. Bhattarai, Jindong Gu, Volker Tresp, Florian Buettner
Abstract: Using feature attributions for post-hoc explanations is a common practice to understand and verify the predictions of opaque machine learning models. Despite the numerous techniques available, individual methods often produce inconsistent and unstable results, putting their overall reliability into question. In this work, we aim to systematically improve the quality of feature attributions by combining multiple explanations across distinct methods or their variations. For this purpose, we propose a novel approach to derive optimal convex combinations of feature attributions that yield provable improvements of desired quality criteria such as robustness or faithfulness to the model behavior. Through extensive experiments involving various model architectures and popular feature attribution techniques, we demonstrate that our combination strategy consistently outperforms individual methods and existing baselines.
Authors: Sandesh Kamath, Albin Soutif-Cormerais, Joost van de Weijer, Bogdan Raducanu
Abstract: Recent research identified a temporary performance drop on previously learned tasks when transitioning to a new one. This drop is called the stability gap and has great consequences for continual learning: it complicates the direct employment of continually learning since the worse-case performance at task-boundaries is dramatic, it limits its potential as an energy-efficient training paradigm, and finally, the stability drop could result in a reduced final performance of the algorithm. In this paper, we show that the stability gap also occurs when applying joint incremental training of homogeneous tasks. In this scenario, the learner continues training on the same data distribution and has access to all data from previous tasks. In addition, we show that in this scenario, there exists a low-loss linear path to the next minima, but that SGD optimization does not choose this path. We perform further analysis including a finer batch-wise analysis which could provide insights towards potential solution directions.
Authors: Taha Entesari, Sina Sharifi, Mahyar Fazlyab
Abstract: A key challenge that threatens the widespread use of neural networks in safety-critical applications is their vulnerability to adversarial attacks. In this paper, we study the second-order behavior of continuously differentiable deep neural networks, focusing on robustness against adversarial perturbations. First, we provide a theoretical analysis of robustness and attack certificates for deep classifiers by leveraging local gradients and upper bounds on the second derivative (curvature constant). Next, we introduce a novel algorithm to analytically compute provable upper bounds on the second derivative of neural networks. This algorithm leverages the compositional structure of the model to propagate the curvature bound layer-by-layer, giving rise to a scalable and modular approach. The proposed bound can serve as a differentiable regularizer to control the curvature of neural networks during training, thereby enhancing robustness. Finally, we demonstrate the efficacy of our method on classification tasks using the MNIST and CIFAR-10 datasets.
Authors: Hartmut F\"uhr, Max Getter
Abstract: We analyze energy decay for deep convolutional neural networks employed as feature extractors, such as Mallat's wavelet scattering transform. For time-frequency scattering transforms based on Gabor filters, it has been established that energy decay is exponential, for arbitrary square-integrable input signals. Our main results allow to prove that this is wrong for wavelet scattering in arbitrary dimensions. In this setting, the energy decay of the scattering transform acting on a generic square-integrable signal turns out to be arbitrarily slow. The fact that this behavior holds for dense subsets of $L^2(\mathbb{R}^d)$ emphasizes that fast energy decay is generally not a stable property of signals. We complement these findings with positive results allowing to conclude fast (up to exponential) energy decay for generalized Sobolev spaces that are tailored to the frequency localization of the underlying filter bank. Both negative and positive results highlight that energy decay in scattering networks critically depends on the interplay of the respective frequency localizations of the signal on the one hand, and of the employed filters on the other.
Authors: Christos Sakaridis, Haoran Wang, Ke Li, Ren\'e Zurbr\"ugg, Arpit Jadon, Wim Abbeloos, Daniel Olmeda Reino, Luc Van Gool, Dengxin Dai
Abstract: Level-5 driving automation requires a robust visual perception system that can parse input images under any condition. However, existing driving datasets for dense semantic perception are either dominated by images captured under normal conditions or are small in scale. To address this, we introduce ACDC, the Adverse Conditions Dataset with Correspondences for training and testing methods for diverse semantic perception tasks on adverse visual conditions. ACDC consists of a large set of 8012 images, half of which (4006) are equally distributed between four common adverse conditions: fog, nighttime, rain, and snow. Each adverse-condition image comes with a high-quality pixel-level panoptic annotation, a corresponding image of the same scene under normal conditions, and a binary mask that distinguishes between intra-image regions of clear and uncertain semantic content. 1503 of the corresponding normal-condition images feature panoptic annotations, raising the total annotated images to 5509. ACDC supports the standard tasks of semantic segmentation, object detection, instance segmentation, and panoptic segmentation, as well as the newly introduced uncertainty-aware semantic segmentation. A detailed empirical study demonstrates the challenges that the adverse domains of ACDC pose to state-of-the-art supervised and unsupervised approaches and indicates the value of our dataset in steering future progress in the field. Our dataset and benchmark are publicly available at https://acdc.vision.ee.ethz.ch
Authors: Lei Wang
Abstract: Human action recognition still exists many challenging problems such as different viewpoints, occlusion, lighting conditions, human body size and the speed of action execution, although it has been widely used in different areas. To tackle these challenges, the Kinect depth sensor has been developed to record real time depth sequences, which are insensitive to the color of human clothes and illumination conditions. Many methods on recognizing human action have been reported in the literature such as HON4D, HOPC, RBD and HDG, which use the 4D surface normals, pointclouds, skeleton-based model and depth gradients respectively to capture discriminative information from depth videos or skeleton data. In this research project, the performance of four aforementioned algorithms will be analyzed and evaluated using five benchmark datasets, which cover challenging issues such as noise, change of viewpoints, background clutters and occlusions. We also implemented and improved the HDG algorithm, and applied it in cross-view action recognition using the UWA3D Multiview Activity dataset. Moreover, we used different combinations of individual feature vectors in HDG for performance evaluation. The experimental results show that our improvement of HDG outperforms other three state-of-the-art algorithms for cross-view action recognition.
Authors: Alex Vicente-Sola, Davide L. Manna, Paul Kirkland, Gaetano Di Caterina, Trevor Bihl
Abstract: Spiking Neural Networks (SNN) are characterised by their unique temporal dynamics, but the properties and advantages of such computations are still not well understood. In order to provide answers, in this work we demonstrate how Spiking neurons can enable temporal feature extraction in feed-forward neural networks without the need for recurrent synapses, and how recurrent SNNs can achieve comparable results to LSTM with a smaller number of parameters. This shows how their bio-inspired computing principles can be successfully exploited beyond energy efficiency gains and evidences their differences with respect to conventional artificial neural networks. These results are obtained through a new task, DVS-Gesture-Chain (DVS-GC), which allows, for the first time, to evaluate the perception of temporal dependencies in a real event-based action recognition dataset. Our study proves how the widely used DVS Gesture benchmark can be solved by networks without temporal feature extraction when its events are accumulated in frames, unlike the new DVS-GC which demands an understanding of the order in which events happen. Furthermore, this setup allowed us to reveal the role of the leakage rate in spiking neurons for temporal processing tasks and demonstrated the benefits of "hard reset" mechanisms. Additionally, we also show how time-dependent weights and normalization can lead to understanding order by means of temporal attention.
Authors: Xiantong Zhao, Yinan Han, Shengjing Tian, Jian Liu, Xiuping Liu
Abstract: Although recent Siamese network-based trackers have achieved impressive perceptual accuracy for single object tracking in LiDAR point clouds, they usually utilized heavy correlation operations to capture category-level characteristics only, and overlook the inherent merit of arbitrariness in contrast to multiple object tracking. In this work, we propose a radically novel one-stream network with the strength of the instance-level encoding, which avoids the correlation operations occurring in previous Siamese network, thus considerably reducing the computational effort. In particular, the proposed method mainly consists of a Template-aware Transformer Module (TTM) and a Multi-scale Feature Aggregation (MFA) module capable of fusing spatial and semantic information. The TTM stitches the specified template and the search region together and leverages an attention mechanism to establish the information flow, breaking the previous pattern of independent \textit{extraction-and-correlation}. As a result, this module makes it possible to directly generate template-aware features that are suitable for the arbitrary and continuously changing nature of the target, enabling the model to deal with unseen categories. In addition, the MFA is proposed to make spatial and semantic information complementary to each other, which is characterized by reverse directional feature propagation that aggregates information from shallow to deep layers. Extensive experiments on KITTI and nuScenes demonstrate that our method has achieved considerable performance not only for class-specific tracking but also for class-agnostic tracking with less computation and higher efficiency.
Authors: Jay Bhanushali, Manivannan Muniyandi, Praneeth Chakravarthula
Abstract: We present a cross-domain inference technique that learns from synthetic data to estimate depth and normals for in-the-wild omnidirectional 3D scenes encountered in real-world uncontrolled settings. To this end, we introduce UBotNet, an architecture that combines UNet and Bottleneck Transformer elements to predict consistent scene normals and depth. We also introduce the OmniHorizon synthetic dataset containing 24,335 omnidirectional images that represent a wide variety of outdoor environments, including buildings, streets, and diverse vegetation. This dataset is generated from expansive, lifelike virtual spaces and encompasses dynamic scene elements, such as changing lighting conditions, different times of day, pedestrians, and vehicles. Our experiments show that UBotNet achieves significantly improved accuracy in depth estimation and normal estimation compared to existing models. Lastly, we validate cross-domain synthetic-to-real depth and normal estimation on real outdoor images using UBotNet trained solely on our synthetic OmniHorizon dataset, demonstrating the potential of both the synthetic dataset and the proposed network for real-world scene understanding applications.
Authors: Ahmed Imtiaz Humayun, Randall Balestriero, Guha Balakrishnan, Richard Baraniuk
Abstract: Current Deep Network (DN) visualization and interpretability methods rely heavily on data space visualizations such as scoring which dimensions of the data are responsible for their associated prediction or generating new data features or samples that best match a given DN unit or representation. In this paper, we go one step further by developing the first provably exact method for computing the geometry of a DN's mapping - including its decision boundary - over a specified region of the data space. By leveraging the theory of Continuous Piece-Wise Linear (CPWL) spline DNs, SplineCam exactly computes a DNs geometry without resorting to approximations such as sampling or architecture simplification. SplineCam applies to any DN architecture based on CPWL nonlinearities, including (leaky-)ReLU, absolute value, maxout, and max-pooling and can also be applied to regression DNs such as implicit neural representations. Beyond decision boundary visualization and characterization, SplineCam enables one to compare architectures, measure generalizability and sample from the decision boundary on or off the manifold. Project Website: bit.ly/splinecam.
Authors: Adrian Shuai Li, Elisa Bertino, Xuan-Hong Dang, Ankush Singla, Yuhai Tu, Mark N Wegman
Abstract: The most effective domain adaptation (DA) involves the decomposition of data representation into a domain independent representation (DIRep), and a domain dependent representation (DDRep). A classifier is trained by using the DIRep of the labeled source images. Since the DIRep is domain invariant, the classifier can be "transferred" to make predictions for the target domain with no (or few) labels. However, information useful for classification in the target domain can "hide" in the DDRep in current DA algorithms such as Domain-Separation-Networks (DSN). DSN's weak constraint to enforce orthogonality of DIRep and DDRep, allows this hiding and can result in poor performance. To address this shortcoming, we developed a new algorithm wherein a stronger constraint is imposed to minimize the DDRep by using a KL divergent loss for the DDRep in order to create the maximal DIRep that enhances transfer learning performance. By using synthetic data sets, we show explicitly that depending on initialization DSN with its weaker constraint can lead to sub-optimal solutions with poorer DA performance whereas our algorithm with maximal DIRep is robust against such perturbations. We demonstrate the equal-or-better performance of our approach against state-of-the-art algorithms by using several standard benchmark image datasets including Office. We further highlight the compatibility of our algorithm with pretrained models, extending its applicability and versatility in real-world scenarios.
Authors: Ishaan Singh Rawal, Alexander Matyasko, Shantanu Jaiswal, Basura Fernando, Cheston Tan
Abstract: While VideoQA Transformer models demonstrate competitive performance on standard benchmarks, the reasons behind their success are not fully understood. Do these models capture the rich multimodal structures and dynamics from video and text jointly? Or are they achieving high scores by exploiting biases and spurious features? Hence, to provide insights, we design $\textit{QUAG}$ (QUadrant AveraGe), a lightweight and non-parametric probe, to conduct dataset-model combined representation analysis by impairing modality fusion. We find that the models achieve high performance on many datasets without leveraging multimodal representations. To validate QUAG further, we design $\textit{QUAG-attention}$, a less-expressive replacement of self-attention with restricted token interactions. Models with QUAG-attention achieve similar performance with significantly fewer multiplication operations without any finetuning. Our findings raise doubts about the current models' abilities to learn highly-coupled multimodal representations. Hence, we design the $\textit{CLAVI}$ (Complements in LAnguage and VIdeo) dataset, a stress-test dataset curated by augmenting real-world videos to have high modality coupling. Consistent with the findings of QUAG, we find that most of the models achieve near-trivial performance on CLAVI. This reasserts the limitations of current models for learning highly-coupled multimodal representations, that is not evaluated by the current datasets (project page: https://dissect-videoqa.github.io ).
Authors: Rahul Palnitkar, Jeova Farias Sales Rocha Neto
Abstract: Spectral Clustering is one of the most traditional methods to solve segmentation problems. Based on Normalized Cuts, it aims at partitioning an image using an objective function defined by a graph. Despite their mathematical attractiveness, spectral approaches are traditionally neglected by the scientific community due to their practical issues and underperformance. In this paper, we adopt a sparse graph formulation based on the inclusion of extra nodes to a simple grid graph. While the grid encodes the pixel spatial disposition, the extra nodes account for the pixel color data. Applying the original Normalized Cuts algorithm to this graph leads to a simple and scalable method for spectral image segmentation, with an interpretable solution. Our experiments also demonstrate that our proposed methodology over performs both traditional and modern unsupervised algorithms for segmentation in both real and synthetic data.
Authors: Siyi Du, Nourhan Bayasi, Ghassan Hamarneh, Rafeef Garbi
Abstract: Despite its clinical utility, medical image segmentation (MIS) remains a daunting task due to images' inherent complexity and variability. Vision transformers (ViTs) have recently emerged as a promising solution to improve MIS; however, they require larger training datasets than convolutional neural networks. To overcome this obstacle, data-efficient ViTs were proposed, but they are typically trained using a single source of data, which overlooks the valuable knowledge that could be leveraged from other available datasets. Naivly combining datasets from different domains can result in negative knowledge transfer (NKT), i.e., a decrease in model performance on some domains with non-negligible inter-domain heterogeneity. In this paper, we propose MDViT, the first multi-domain ViT that includes domain adapters to mitigate data-hunger and combat NKT by adaptively exploiting knowledge in multiple small data resources (domains). Further, to enhance representation learning across domains, we integrate a mutual knowledge distillation paradigm that transfers knowledge between a universal network (spanning all the domains) and auxiliary domain-specific branches. Experiments on 4 skin lesion segmentation datasets show that MDViT outperforms state-of-the-art algorithms, with superior segmentation performance and a fixed model size, at inference time, even as more domains are added. Our code is available at https://github.com/siyi-wind/MDViT.
Authors: Dongjia Zhao, Lei Qi, Xiao Shi, Yinghuan Shi, Xin Geng
Abstract: Single domain generalization aims to enhance the ability of the model to generalize to unknown domains when trained on a single source domain. However, the limited diversity in the training data hampers the learning of domain-invariant features, resulting in compromised generalization performance. To address this, data perturbation (augmentation) has emerged as a crucial method to increase data diversity. Nevertheless, existing perturbation methods often focus on either image-level or feature-level perturbations independently, neglecting their synergistic effects. To overcome these limitations, we propose CPerb, a simple yet effective cross-perturbation method. Specifically, CPerb utilizes both horizontal and vertical operations. Horizontally, it applies image-level and feature-level perturbations to enhance the diversity of the training data, mitigating the issue of limited diversity in single-source domains. Vertically, it introduces multi-route perturbation to learn domain-invariant features from different perspectives of samples with the same semantic category, thereby enhancing the generalization capability of the model. Additionally, we propose MixPatch, a novel feature-level perturbation method that exploits local image style information to further diversify the training data. Extensive experiments on various benchmark datasets validate the effectiveness of our method.
Authors: Sreyan Ghosh, Chandra Kiran Reddy Evuru, Sonal Kumar, Utkarsh Tyagi, Sakshi Singh, Sanjoy Chowdhury, Dinesh Manocha
Abstract: Neural image classifiers can often learn to make predictions by overly relying on non-predictive features that are spuriously correlated with the class labels in the training data. This leads to poor performance in real-world atypical scenarios where such features are absent. This paper presents ASPIRE (Language-guided Data Augmentation for SPurIous correlation REmoval), a simple yet effective solution for supplementing the training dataset with images without spurious features, for robust learning against spurious correlations via better generalization. ASPIRE, guided by language at various steps, can generate non-spurious images without requiring any group labeling or existing non-spurious images in the training set. Precisely, we employ LLMs to first extract foreground and background features from textual descriptions of an image, followed by advanced language-guided image editing to discover the features that are spuriously correlated with the class label. Finally, we personalize a text-to-image generation model using the edited images to generate diverse in-domain images without spurious features. ASPIRE is complementary to all prior robust training methods in literature, and we demonstrate its effectiveness across 4 datasets and 9 baselines and show that ASPIRE improves the worst-group classification accuracy of prior methods by 1% - 38%. We also contribute a novel test set for the challenging Hard ImageNet dataset.
Authors: Cheng Feng, Congxuan Zhang, Zhen Chen, Weiming Hu, Liyue Ge
Abstract: Depth sensing is of paramount importance for unmanned aerial and autonomous vehicles. Nonetheless, contemporary monocular depth estimation methods employing complex deep neural networks within Convolutional Neural Networks are inadequately expedient for real-time inference on embedded platforms. This paper endeavors to surmount this challenge by proposing two efficient and lightweight architectures, RT-MonoDepth and RT-MonoDepth-S, thereby mitigating computational complexity and latency. Our methodologies not only attain accuracy comparable to prior depth estimation methods but also yield faster inference speeds. Specifically, RT-MonoDepth and RT-MonoDepth-S achieve frame rates of 18.4&30.5 FPS on NVIDIA Jetson Nano and 253.0&364.1 FPS on Jetson AGX Orin, utilizing a single RGB image of resolution 640x192. The experimental results underscore the superior accuracy and faster inference speed of our methods in comparison to existing fast monocular depth estimation methodologies on the KITTI dataset.
Authors: Zhenying Fang, Jun Yu, Richang Hong
Abstract: Temporal action detection aims to recognize the action category and determine each action instance's starting and ending time in untrimmed videos. The mixed methods have achieved remarkable performance by seamlessly merging anchor-based and anchor-free approaches. Nonetheless, there are still two crucial issues within the mixed framework: (1) Brute-force merging and handcrafted anchor design hinder the substantial potential and practicality of the mixed methods. (2) Within-category predictions show a significant abundance of false positives. In this paper, we propose a novel Boundary Discretization and Reliable Classification Network (BDRC-Net) that addresses the issues above by introducing boundary discretization and reliable classification modules. Specifically, the boundary discretization module (BDM) elegantly merges anchor-based and anchor-free approaches in the form of boundary discretization, eliminating the need for the traditional handcrafted anchor design. Furthermore, the reliable classification module (RCM) predicts reliable global action categories to reduce false positives. Extensive experiments conducted on different benchmarks demonstrate that our proposed method achieves competitive detection performance. The code will be released at https://github.com/zhenyingfang/BDRC-Net.
Authors: Takeru Miyato, Bernhard Jaeger, Max Welling, Andreas Geiger
Abstract: As transformers are equivariant to the permutation of input tokens, encoding the positional information of tokens is necessary for many tasks. However, since existing positional encoding schemes have been initially designed for NLP tasks, their suitability for vision tasks, which typically exhibit different structural properties in their data, is questionable. We argue that existing positional encoding schemes are suboptimal for 3D vision tasks, as they do not respect their underlying 3D geometric structure. Based on this hypothesis, we propose a geometry-aware attention mechanism that encodes the geometric structure of tokens as relative transformation determined by the geometric relationship between queries and key-value pairs. By evaluating on multiple novel view synthesis (NVS) datasets in the sparse wide-baseline multi-view setting, we show that our attention, called Geometric Transform Attention (GTA), improves learning efficiency and performance of state-of-the-art transformer-based NVS models without any additional learned parameters and only minor computational overhead.
Authors: Hmrishav Bandyopadhyay, Subhadeep Koley, Ayan Das, Ayan Kumar Bhunia, Aneeshan Sain, Pinaki Nath Chowdhury, Tao Xiang, Yi-Zhe Song
Abstract: In this paper, we democratise 3D content creation, enabling precise generation of 3D shapes from abstract sketches while overcoming limitations tied to drawing skills. We introduce a novel part-level modelling and alignment framework that facilitates abstraction modelling and cross-modal correspondence. Leveraging the same part-level decoder, our approach seamlessly extends to sketch modelling by establishing correspondence between CLIPasso edgemaps and projected 3D part regions, eliminating the need for a dataset pairing human sketches and 3D shapes. Additionally, our method introduces a seamless in-position editing process as a byproduct of cross-modal part-aligned modelling. Operating in a low-dimensional implicit space, our approach significantly reduces computational demands and processing time.
Authors: Yunhan Zhao, Haoyu Ma, Shu Kong, Charless Fowlkes
Abstract: Egocentric sensors such as AR/VR devices capture human-object interactions and offer the potential to provide task-assistance by recalling 3D locations of objects of interest in the surrounding environment. This capability requires instance tracking in real-world 3D scenes from egocentric videos (IT3DEgo). We explore this problem by first introducing a new benchmark dataset, consisting of RGB and depth videos, per-frame camera pose, and instance-level annotations in both 2D camera and 3D world coordinates. We present an evaluation protocol which evaluates tracking performance in 3D coordinates with two settings for enrolling instances to track: (1) single-view online enrollment where an instance is specified on-the-fly based on the human wearer's interactions. and (2) multi-view pre-enrollment where images of an instance to be tracked are stored in memory ahead of time. To address IT3DEgo, we first re-purpose methods from relevant areas, e.g., single object tracking (SOT) -- running SOT methods to track instances in 2D frames and lifting them to 3D using camera pose and depth. We also present a simple method that leverages pretrained segmentation and detection models to generate proposals from RGB frames and match proposals with enrolled instance images. Our experiments show that our method (with no finetuning) significantly outperforms SOT-based approaches in the egocentric setting. We conclude by arguing that the problem of egocentric instance tracking is made easier by leveraging camera pose and using a 3D allocentric (world) coordinate representation.
Authors: Zhiming Hu, Jiahui Xu, Syn Schmitt, Andreas Bulling
Abstract: Human eye gaze plays a significant role in many virtual and augmented reality (VR/AR) applications, such as gaze-contingent rendering, gaze-based interaction, or eye-based activity recognition. However, prior works on gaze analysis and prediction have only explored eye-head coordination and were limited to human-object interactions. We first report a comprehensive analysis of eye-body coordination in various human-object and human-human interaction activities based on four public datasets collected in real-world (MoGaze), VR (ADT), as well as AR (GIMO and EgoBody) environments. We show that in human-object interactions, e.g. pick and place, eye gaze exhibits strong correlations with full-body motion while in human-human interactions, e.g. chat and teach, a person's gaze direction is correlated with the body orientation towards the interaction partner. Informed by these analyses we then present Pose2Gaze, a novel eye-body coordination model that uses a convolutional neural network and a spatio-temporal graph convolutional neural network to extract features from head direction and full-body poses, respectively, and then uses a convolutional neural network to predict eye gaze. We compare our method with state-of-the-art methods that predict eye gaze only from head movements and show that Pose2Gaze outperforms these baselines with an average improvement of 24.0% on MoGaze, 10.1% on ADT, 21.3% on GIMO, and 28.6% on EgoBody in mean angular error, respectively. We also show that our method significantly outperforms prior methods in the sample downstream task of eye-based activity recognition. These results underline the significant information content available in eye-body coordination during daily activities and open up a new direction for gaze prediction.
Authors: Yukun Zuo, Hantao Yao, Liansheng Zhuang, Changsheng Xu
Abstract: Audio-visual video recognition (AVVR) aims to integrate audio and visual clues to categorize videos accurately. While existing methods train AVVR models using provided datasets and achieve satisfactory results, they struggle to retain historical class knowledge when confronted with new classes in real-world situations. Currently, there are no dedicated methods for addressing this problem, so this paper concentrates on exploring Class Incremental Audio-Visual Video Recognition (CIAVVR). For CIAVVR, since both stored data and learned model of past classes contain historical knowledge, the core challenge is how to capture past data knowledge and past model knowledge to prevent catastrophic forgetting. We introduce Hierarchical Augmentation and Distillation (HAD), which comprises the Hierarchical Augmentation Module (HAM) and Hierarchical Distillation Module (HDM) to efficiently utilize the hierarchical structure of data and models, respectively. Specifically, HAM implements a novel augmentation strategy, segmental feature augmentation, to preserve hierarchical model knowledge. Meanwhile, HDM introduces newly designed hierarchical (video-distribution) logical distillation and hierarchical (snippet-video) correlative distillation to capture and maintain the hierarchical intra-sample knowledge of each data and the hierarchical inter-sample knowledge between data, respectively. Evaluations on four benchmarks (AVE, AVK-100, AVK-200, and AVK-400) demonstrate that the proposed HAD effectively captures hierarchical information in both data and models, resulting in better preservation of historical class knowledge and improved performance. Furthermore, we provide a theoretical analysis to support the necessity of the segmental feature augmentation strategy.
Authors: Bang An, Mucong Ding, Tahseen Rabbani, Aakriti Agrawal, Yuancheng Xu, Chenghao Deng, Sicheng Zhu, Abdirisak Mohamed, Yuxin Wen, Tom Goldstein, Furong Huang
Abstract: In the burgeoning age of generative AI, watermarks act as identifiers of provenance and artificial content. We present WAVES (Watermark Analysis Via Enhanced Stress-testing), a benchmark for assessing image watermark robustness, overcoming the limitations of current evaluation methods. WAVES integrates detection and identification tasks and establishes a standardized evaluation protocol comprised of a diverse range of stress tests. The attacks in WAVES range from traditional image distortions to advanced, novel variations of diffusive, and adversarial attacks. Our evaluation examines two pivotal dimensions: the degree of image quality degradation and the efficacy of watermark detection after attacks. Our novel, comprehensive evaluation reveals previously undetected vulnerabilities of several modern watermarking algorithms. We envision WAVES as a toolkit for the future development of robust watermarks. The project is available at https://wavesbench.github.io/
Authors: Li Meng, Morten Goodwin, Anis Yazidi, Paal Engelstad
Abstract: Vision Transformers implement multi-head self-attention via stacking multiple attention blocks. The query, key, and value are often intertwined and generated within those blocks via a single, shared linear transformation. This paper explores the concept of disentangling the key from the query and value, and adopting a manifold representation for the key. Our experiments reveal that decoupling and endowing the key with a manifold structure can enhance the model's performance. Specifically, ViT-B exhibits a 0.87% increase in top-1 accuracy, while Swin-T sees a boost of 0.52% in top-1 accuracy on the ImageNet-1K dataset, with eight charts in the manifold key. Our approach also yields positive results in object detection and instance segmentation tasks on the COCO dataset. We establish that these performance gains are not merely due to the simplicity of adding more parameters and computations. Future research may investigate strategies for cutting the budget of such representations and aim for further performance improvements based on our findings.
Authors: Juan Nathaniel, Yongquan Qu, Tung Nguyen, Sungduk Yu, Julius Busecke, Aditya Grover, Pierre Gentine
Abstract: Accurate prediction of climate in the subseasonal-to-seasonal scale is crucial for disaster preparedness and robust decision making amidst climate change. Yet, forecasting beyond the weather timescale is challenging because it deals with problems other than initial conditions, including boundary interaction, butterfly effect, and our inherent lack of physical understanding. At present, existing benchmarks tend to have shorter forecasting range of up-to 15 days, do not include a wide range of operational baselines, and lack physics-based constraints for explainability. Thus, we propose ChaosBench, a challenging benchmark to extend the predictability range of data-driven weather emulators to S2S timescale. First, ChaosBench is comprised of variables beyond the typical surface-atmospheric ERA5 to also include ocean, ice, and land reanalysis products that span over 45 years to allow for full Earth system emulation that respects boundary conditions. We also propose physics-based, in addition to deterministic and probabilistic metrics, to ensure a physically-consistent ensemble that accounts for butterfly effect. Furthermore, we evaluate on a diverse set of physics-based forecasts from four national weather agencies as baselines to our data-driven counterpart such as ClimaX, PanguWeather, GraphCast, and FourCastNetV2. Overall, we find methods originally developed for weather-scale applications fail on S2S task: their performance simply collapse to an unskilled climatology. Nonetheless, we outline and demonstrate several strategies that can potentially extend the predictability range of existing weather emulators, including the use of ensembles and robust control of error propagation. Our benchmark, datasets, and instructions are available at https://leap-stc.github.io/ChaosBench.
Authors: Jiakai Sun, Han Jiao, Guangyuan Li, Zhanjie Zhang, Lei Zhao, Wei Xing
Abstract: Constructing photo-realistic Free-Viewpoint Videos (FVVs) of dynamic scenes from multi-view videos remains a challenging endeavor. Despite the remarkable advancements achieved by current neural rendering techniques, these methods generally require complete video sequences for offline training and are not capable of real-time rendering. To address these constraints, we introduce 3DGStream, a method designed for efficient FVV streaming of real-world dynamic scenes. Our method achieves fast on-the-fly per-frame reconstruction within 12 seconds and real-time rendering at 200 FPS. Specifically, we utilize 3D Gaussians (3DGs) to represent the scene. Instead of the na\"ive approach of directly optimizing 3DGs per-frame, we employ a compact Neural Transformation Cache (NTC) to model the translations and rotations of 3DGs, markedly reducing the training time and storage required for each FVV frame. Furthermore, we propose an adaptive 3DG addition strategy to handle emerging objects in dynamic scenes. Experiments demonstrate that 3DGStream achieves competitive performance in terms of rendering speed, image quality, training time, and model storage when compared with state-of-the-art methods.
Authors: Takahiro Shindo, Kein Yamada, Taiju Watanabe, Hiroshi Watanabe
Abstract: Image Coding for Machines (ICM) is an image compression technique for image recognition. This technique is essential due to the growing demand for image recognition AI. In this paper, we propose a method for ICM that focuses on encoding and decoding only the edge information of object parts in an image, which we call SA-ICM. This is an Learned Image Compression (LIC) model trained using edge information created by Segment Anything. Our method can be used for image recognition models with various tasks. SA-ICM is also robust to changes in input data, making it effective for a variety of use cases. Additionally, our method provides benefits from a privacy point of view, as it removes human facial information on the encoder's side, thus protecting one's privacy. Furthermore, this LIC model training method can be used to train Neural Representations for Videos (NeRV), which is a video compression model. By training NeRV using edge information created by Segment Anything, it is possible to create a NeRV that is effective for image recognition (SA-NeRV). Experimental results confirm the advantages of SA-ICM, presenting the best performance in image compression for image recognition. We also show that SA-NeRV is superior to ordinary NeRV in video compression for machines. Code is available at https://github.com/final-0/SA-ICM.
Authors: Sahand Sharifzadeh, Christos Kaplanis, Shreya Pathak, Dharshan Kumaran, Anastasija Ilic, Jovana Mitrovic, Charles Blundell, Andrea Banino
Abstract: The creation of high-quality human-labeled image-caption datasets presents a significant bottleneck in the development of Visual-Language Models (VLMs). In this work, we investigate an approach that leverages the strengths of Large Language Models (LLMs) and image generation models to create synthetic image-text pairs for efficient and effective VLM training. Our method employs a pretrained text-to-image model to synthesize image embeddings from captions generated by an LLM. Despite the text-to-image model and VLM initially being trained on the same data, our approach leverages the image generator's ability to create novel compositions, resulting in synthetic image embeddings that expand beyond the limitations of the original dataset. Extensive experiments demonstrate that our VLM, finetuned on synthetic data achieves comparable performance to models trained solely on human-annotated data, while requiring significantly less data. Furthermore, we perform a set of analyses on captions which reveals that semantic diversity and balance are key aspects for better downstream performance. Finally, we show that synthesizing images in the image embedding space is 25\% faster than in the pixel space. We believe our work not only addresses a significant challenge in VLM training but also opens up promising avenues for the development of self-improving multi-modal models.
Authors: Chenyang Ma, Kai Lu, Ta-Ying Cheng, Niki Trigoni, Andrew Markham
Abstract: Current state-of-the-art spatial reasoning-enhanced VLMs are trained to excel at spatial visual question answering (VQA). However, we believe that higher-level 3D-aware tasks, such as articulating dynamic scene changes and motion planning, require a fundamental and explicit 3D understanding beyond current spatial VQA datasets. In this work, we present SpatialPIN, a framework designed to enhance the spatial reasoning capabilities of VLMs through prompting and interacting with priors from multiple 3D foundation models in a zero-shot, training-free manner. Extensive experiments demonstrate that our spatial reasoning-imbued VLM performs well on various forms of spatial VQA and can extend to help in various downstream robotics tasks such as pick and stack and trajectory planning.
Authors: Zhitong Xiong, Yi Wang, Fahong Zhang, Adam J. Stewart, Jo\"elle Hanna, Damian Borth, Ioannis Papoutsis, Bertrand Le Saux, Gustau Camps-Valls, Xiao Xiang Zhu
Abstract: The development of foundation models has revolutionized our ability to interpret the Earth's surface using satellite observational data. Traditional models have been siloed, tailored to specific sensors or data types like optical, radar, and hyperspectral, each with its own unique characteristics. This specialization hinders the potential for a holistic analysis that could benefit from the combined strengths of these diverse data sources. Our novel approach introduces the Dynamic One-For-All (DOFA) model, leveraging the concept of neural plasticity in brain science to integrate various data modalities into a single framework adaptively. This dynamic hypernetwork, adjusting to different wavelengths, enables a single versatile Transformer jointly trained on data from five sensors to excel across 12 distinct Earth observation tasks, including sensors never seen during pretraining. DOFA's innovative design offers a promising leap towards more accurate, efficient, and unified Earth observation analysis, showcasing remarkable adaptability and performance in harnessing the potential of multimodal Earth observation data.
Authors: Zelin Zhao, Fenglei Fan, Wenlong Liao, Junchi Yan
Abstract: Many contemporary studies utilize grid-based models for neural field representation, but a systematic analysis of grid-based models is still missing, hindering the improvement of those models. Therefore, this paper introduces a theoretical framework for grid-based models. This framework points out that these models' approximation and generalization behaviors are determined by grid tangent kernels (GTK), which are intrinsic properties of grid-based models. The proposed framework facilitates a consistent and systematic analysis of diverse grid-based models. Furthermore, the introduced framework motivates the development of a novel grid-based model named the Multiplicative Fourier Adaptive Grid (MulFAGrid). The numerical analysis demonstrates that MulFAGrid exhibits a lower generalization bound than its predecessors, indicating its robust generalization performance. Empirical studies reveal that MulFAGrid achieves state-of-the-art performance in various tasks, including 2D image fitting, 3D signed distance field (SDF) reconstruction, and novel view synthesis, demonstrating superior representation ability. The project website is available at https://sites.google.com/view/cvpr24-2034-submission/home.
URLs: https://sites.google.com/view/cvpr24-2034-submission/home.
Authors: Yijia Weng, Bowen Wen, Jonathan Tremblay, Valts Blukis, Dieter Fox, Leonidas Guibas, Stan Birchfield
Abstract: We address the problem of building digital twins of unknown articulated objects from two RGBD scans of the object at different articulation states. We decompose the problem into two stages, each addressing distinct aspects. Our method first reconstructs object-level shape at each state, then recovers the underlying articulation model including part segmentation and joint articulations that associate the two states. By explicitly modeling point-level correspondences and exploiting cues from images, 3D reconstructions, and kinematics, our method yields more accurate and stable results compared to prior work. It also handles more than one movable part and does not rely on any object shape or structure priors. Project page: https://github.com/NVlabs/DigitalTwinArt
Authors: Jinseo Jeong, Junseo Koo, Qimeng Zhang, Gunhee Kim
Abstract: Existing NeRF-based inverse rendering methods suppose that scenes are exclusively illuminated by distant light sources, neglecting the potential influence of emissive sources within a scene. In this work, we confront this limitation using LDR multi-view images captured with emissive sources turned on and off. Two key issues must be addressed: 1) ambiguity arising from the limited dynamic range along with unknown lighting details, and 2) the expensive computational cost in volume rendering to backtrace the paths leading to final object colors. We present a novel approach, ESR-NeRF, leveraging neural networks as learnable functions to represent ray-traced fields. By training networks to satisfy light transport segments, we regulate outgoing radiances, progressively identifying emissive sources while being aware of reflection areas. The results on scenes encompassing emissive sources with various properties demonstrate the superiority of ESR-NeRF in qualitative and quantitative ways. Our approach also extends its applicability to the scenes devoid of emissive sources, achieving lower CD metrics on the DTU dataset.
Authors: Anudeep Das, Vasisht Duddu, Rui Zhang, N. Asokan
Abstract: Diffusion-based text-to-image (T2I) models generate high-fidelity images for given textual prompts. They are trained on large datasets scraped from the Internet, potentially containing unacceptable concepts (e.g., copyright infringing or unsafe). Retraining T2I models after filtering out unacceptable concepts in the training data is inefficient and degrades utility. Hence, there is a need for concept removal techniques (CRTs) which are effective in removing unacceptable concepts, utility-preserving on acceptable concepts, and robust against evasion with adversarial prompts. None of the prior filtering and fine-tuning CRTs satisfy all these requirements simultaneously. We introduce Espresso, the first robust concept filter based on Contrastive Language-Image Pre-Training (CLIP). It identifies unacceptable concepts by projecting the generated image's embedding onto the vector connecting unacceptable and acceptable concepts in the joint text-image embedding space. This ensures robustness by restricting the adversary to adding noise only along this vector, in the direction of the acceptable concept. Further fine-tuning Espresso to separate embeddings of acceptable and unacceptable concepts, while preserving their pairing with image embeddings, ensures both effectiveness and utility. We evaluate Espresso on eleven concepts to show that it is effective (~5% CLIP accuracy on unacceptable concepts), utility-preserving (~93% normalized CLIP score on acceptable concepts), and robust (~4% CLIP accuracy on adversarial prompts for unacceptable concepts). Finally, we present theoretical bounds for the certified robustness of Espresso against adversarial prompts, and an empirical analysis.
Authors: Yuan Zhang, Fei Xiao, Tao Huang, Chun-Kai Fan, Hongyuan Dong, Jiawen Li, Jiacong Wang, Kuan Cheng, Shanghang Zhang, Haoyuan Guo
Abstract: Large vision-language models (LVLMs) have recently achieved rapid progress, exhibiting great perception and reasoning abilities concerning visual information. However, when faced with prompts in different sizes of solution spaces, LVLMs fail to always give consistent answers regarding the same knowledge point. This inconsistency of answers between different solution spaces is prevalent in LVLMs and erodes trust. To this end, we provide a multi-modal benchmark ConBench, to intuitively analyze how LVLMs perform when the solution space of a prompt revolves around a knowledge point. Based on the ConBench tool, we are the first to reveal the tapestry and get the following findings: (1) In the discriminate realm, the larger the solution space of the prompt, the lower the accuracy of the answers. (2) Establish the relationship between the discriminative and generative realms: the accuracy of the discriminative question type exhibits a strong positive correlation with its Consistency with the caption. (3) Compared to open-source models, closed-source models exhibit a pronounced bias advantage in terms of Consistency. Eventually, we ameliorate the consistency of LVLMs by trigger-based diagnostic refinement, indirectly improving the performance of their caption. We hope this paper will accelerate the research community in better evaluating their models and encourage future advancements in the consistency domain.
Authors: Shuaixin Liu, Kunqian Li, Yilin Ding, Qi Qi
Abstract: Underwater Image Enhancement (UIE) aims to improve the visual quality from a low-quality input. Unlike other image enhancement tasks, underwater images suffer from the unavailability of real reference images. Although existing works exploit synthetic images and manually select well-enhanced images as reference images to train enhancement networks, their upper performance bound is limited by the reference domain. To address this challenge, we propose CLIP-UIE, a novel framework that leverages the potential of Contrastive Language-Image Pretraining (CLIP) for the UIE task. Specifically, we propose employing color transfer to yield synthetic images by degrading in-air natural images into corresponding underwater images, guided by the real underwater domain. This approach enables the diffusion model to capture the prior knowledge of mapping transitions from the underwater degradation domain to the real in-air natural domain. Still, fine-tuning the diffusion model for specific downstream tasks is inevitable and may result in the loss of this prior knowledge. To migrate this drawback, we combine the prior knowledge of the in-air natural domain with CLIP to train a CLIP-Classifier. Subsequently, we integrate this CLIP-Classifier with UIE benchmark datasets to jointly fine-tune the diffusion model, guiding the enhancement results towards the in-air natural domain. Additionally, for image enhancement tasks, we observe that both the image-to-image diffusion model and CLIP-Classifier primarily focus on the high-frequency region during fine-tuning. Therefore, we propose a new fine-tuning strategy that specifically targets the high-frequency region, which can be up to 10 times faster than traditional strategies. Extensive experiments demonstrate that our method exhibits a more natural appearance.
Authors: Juntae Kim, Sungwon Woo, Jongho Nang
Abstract: This paper addresses image copy detection, a task in online sharing platforms for copyright protection. While previous approaches have performed exceptionally well, the large size of their networks and descriptors remains a significant disadvantage, complicating their practical application. In this paper, we propose a novel method that achieves a competitive performance by using a lightweight network and compact descriptors. By utilizing relational self-supervised distillation to transfer knowledge from a large network to a small network, we enable the training of lightweight networks with a small descriptor size. Our approach, which we call Relational self-supervised Distillation with Compact Descriptors (RDCD), introduces relational self-supervised distillation (RSD) for flexible representation in a smaller feature space and applies contrastive learning with a hard negative (HN) loss to prevent dimensional collapse. We demonstrate the effectiveness of our method using the DISC2021, Copydays, and NDEC benchmark datasets, with which our lightweight network with compact descriptors achieves a competitive performance. For the DISC2021 benchmark, ResNet-50/EfficientNet- B0 are used as a teacher and student respectively, the micro average precision improved by 5.0%/4.9%/5.9% for 64/128/256 descriptor sizes compared to the baseline method.
Authors: Jun Zheng, Fuwei Zhao, Youjiang Xu, Xin Dong, Xiaodan Liang
Abstract: Video try-on stands as a promising area for its tremendous real-world potential. Prior works are limited to transferring product clothing images onto person videos with simple poses and backgrounds, while underperforming on casually captured videos. Recently, Sora revealed the scalability of Diffusion Transformer (DiT) in generating lifelike videos featuring real-world scenarios. Inspired by this, we explore and propose the first DiT-based video try-on framework for practical in-the-wild applications, named VITON-DiT. Specifically, VITON-DiT consists of a garment extractor, a Spatial-Temporal denoising DiT, and an identity preservation ControlNet. To faithfully recover the clothing details, the extracted garment features are fused with the self-attention outputs of the denoising DiT and the ControlNet. We also introduce novel random selection strategies during training and an Interpolated Auto-Regressive (IAR) technique at inference to facilitate long video generation. Unlike existing attempts that require the laborious and restrictive construction of a paired training dataset, severely limiting their scalability, VITON-DiT alleviates this by relying solely on unpaired human dance videos and a carefully designed multi-stage training strategy. Furthermore, we curate a challenging benchmark dataset to evaluate the performance of casual video try-on. Extensive experiments demonstrate the superiority of VITON-DiT in generating spatio-temporal consistent try-on results for in-the-wild videos with complicated human poses.
Authors: Abrar Fahim, Alex Murphy, Alona Fyshe
Abstract: Multi-modal contrastive models such as CLIP achieve state-of-the-art performance in zero-shot classification by embedding input images and texts on a joint representational space. Recently, a modality gap has been reported in two-encoder contrastive models like CLIP, meaning that the image and text embeddings reside in disjoint areas of the latent space. Previous studies suggest that this gap exists due to 1) the cone effect, 2) mismatched pairs in the dataset, and 3) insufficient training. We show that, even when accounting for all these factors, and even when using the same modality, the contrastive loss actually creates a gap during training. As a result, We propose that the modality gap is inherent to the two-encoder contrastive loss and rename it the contrastive gap. We present evidence that attributes this contrastive gap to low uniformity in CLIP space, resulting in embeddings that occupy only a small portion of the latent space. To close the gap, we adapt the uniformity and alignment properties of unimodal contrastive loss to the multi-modal setting and show that simply adding these terms to the CLIP loss distributes the embeddings more uniformly in the representational space, closing the gap. In our experiments, we show that the modified representational space achieves better performance than default CLIP loss in downstream tasks such as zero-shot image classification and multi-modal arithmetic.
Authors: Shuai Yuan, Guancong Lin, Lixian Zhang, Runmin Dong, Jinxiao Zhang, Shuang Chen, Juepeng Zheng, Jie Wang, Haohuan Fu
Abstract: Fine urban change segmentation using multi-temporal remote sensing images is essential for understanding human-environment interactions in urban areas. Although there have been advances in high-quality land cover datasets that reveal the physical features of urban landscapes, the lack of fine-grained land use datasets hinders a deeper understanding of how human activities are distributed across the landscape and the impact of these activities on the environment, thus constraining proper technique development. To address this, we introduce FUSU, the first fine-grained land use change segmentation dataset for Fine-grained Urban Semantic Understanding. FUSU features the most detailed land use classification system to date, with 17 classes and 30 billion pixels of annotations. It includes bi-temporal high-resolution satellite images with 0.2-0.5 m ground sample distance and monthly optical and radar satellite time series, covering 847 km^2 across five urban areas in the southern and northern of China with different geographical features. The fine-grained land use pixel-wise annotations and high spatial-temporal resolution data provide a robust foundation for developing proper deep learning models to provide contextual insights on human activities and urbanization. To fully leverage FUSU, we propose a unified time-series architecture for both change detection and segmentation. We benchmark FUSU on various methods for several tasks. Dataset and code are available at: https://github.com/yuanshuai0914/FUSU.
Authors: Wenbo Hu, Zi-Yi Dou, Liunian Harold Li, Amita Kamath, Nanyun Peng, Kai-Wei Chang
Abstract: Large Vision-Language Models (LVLMs) typically encode an image into a fixed number of visual tokens (e.g., 576) and process these tokens with a language model. Despite their strong performance, LVLMs face challenges in adapting to varying computational constraints. This raises the question: can we achieve flexibility in the number of visual tokens to suit different tasks and computational resources? We answer this with an emphatic yes. Inspired by Matryoshka Representation Learning, we introduce the Matryoshka Query Transformer (MQT), capable of encoding an image into m visual tokens during inference, where m can be any number up to a predefined maximum. This is achieved by employing a query transformer with M latent query tokens to compress the visual embeddings. During each training step, we randomly select m <= M latent query tokens and train the model using only these first m tokens, discarding the rest. Combining MQT with LLaVA, we train a single model once, and flexibly and drastically reduce the number of inference-time visual tokens while maintaining similar or better performance compared to training independent models for each number of tokens. Our model, MQT-LLAVA, matches LLaVA-1.5 performance across 11 benchmarks using a maximum of 256 tokens instead of LLaVA's fixed 576. Reducing to 16 tokens (8x less TFLOPs) only sacrifices the performance by 2.4 points on MMBench. On certain tasks such as ScienceQA and MMMU, we can even go down to only 2 visual tokens with performance drops of just 3% and 6% each. Our exploration of the trade-off between the accuracy and computational cost brought about by the number of visual tokens facilitates future research to achieve the best of both worlds.
Authors: Bowen Zheng, Tianming Yang
Abstract: Diffusion Models (DMs) have achieved great success in image generation and other fields. By fine sampling through the trajectory defined by the SDE/ODE solver based on a well-trained score model, DMs can generate remarkable high-quality results. However, this precise sampling often requires multiple steps and is computationally demanding. To address this problem, instance-based distillation methods have been proposed to distill a one-step generator from a DM by having a simpler student model mimic a more complex teacher model. Yet, our research reveals an inherent limitations in these methods: the teacher model, with more steps and more parameters, occupies different local minima compared to the student model, leading to suboptimal performance when the student model attempts to replicate the teacher. To avoid this problem, we introduce a novel distributional distillation method, which uses an exclusive distributional loss. This method exceeds state-of-the-art (SOTA) results while requiring significantly fewer training images. Additionally, we show that DMs' layers are differentially activated at different time steps, leading to an inherent capability to generate images in a single step. Freezing most of the convolutional layers in a DM during distributional distillation enables this innate capability and leads to further performance improvements. Our method achieves the SOTA results on CIFAR-10 (FID 1.54), AFHQv2 64x64 (FID 1.23), FFHQ 64x64 (FID 0.85) and ImageNet 64x64 (FID 1.16) with great efficiency. Most of those results are obtained with only 5 million training images within 6 hours on 8 A100 GPUs.
Authors: Zhiming Meng, Hui Li, Zeyang Zhang, Zhongwei Shen, Yunlong Yu, Xiaoning Song, Xiaojun Wu
Abstract: Generative models are widely utilized to model the distribution of fused images in the field of infrared and visible image fusion. However, current generative models based fusion methods often suffer from unstable training and slow inference speed. To tackle this problem, a novel fusion method based on consistency model is proposed, termed as CoMoFusion, which can generate the high-quality images and achieve fast image inference speed. In specific, the consistency model is used to construct multi-modal joint features in the latent space with the forward and reverse process. Then, the infrared and visible features extracted by the trained consistency model are fed into fusion module to generate the final fused image. In order to enhance the texture and salient information of fused images, a novel loss based on pixel value selection is also designed. Extensive experiments on public datasets illustrate that our method obtains the SOTA fusion performance compared with the existing fusion methods.
Authors: Yunbin Tu, Liang Li, Li Su, Zheng-Jun Zha, Chenggang Yan, Qingming Huang
Abstract: Multi-change captioning aims to describe complex and coupled changes within an image pair in natural language. Compared with single-change captioning, this task requires the model to have higher-level cognition ability to reason an arbitrary number of changes. In this paper, we propose a novel context-aware difference distilling (CARD) network to capture all genuine changes for yielding sentences. Given an image pair, CARD first decouples context features that aggregate all similar/dissimilar semantics, termed common/difference context features. Then, the consistency and independence constraints are designed to guarantee the alignment/discrepancy of common/difference context features. Further, the common context features guide the model to mine locally unchanged features, which are subtracted from the pair to distill locally difference features. Next, the difference context features augment the locally difference features to ensure that all changes are distilled. In this way, we obtain an omni-representation of all changes, which is translated into linguistic sentences by a transformer decoder. Extensive experiments on three public datasets show CARD performs favourably against state-of-the-art methods.The code is available at https://github.com/tuyunbin/CARD.
Authors: Yue Ma, Hongyu Liu, Hongfa Wang, Heng Pan, Yingqing He, Junkun Yuan, Ailing Zeng, Chengfei Cai, Heung-Yeung Shum, Wei Liu, Qifeng Chen
Abstract: We present Follow-Your-Emoji, a diffusion-based framework for portrait animation, which animates a reference portrait with target landmark sequences. The main challenge of portrait animation is to preserve the identity of the reference portrait and transfer the target expression to this portrait while maintaining temporal consistency and fidelity. To address these challenges, Follow-Your-Emoji equipped the powerful Stable Diffusion model with two well-designed technologies. Specifically, we first adopt a new explicit motion signal, namely expression-aware landmark, to guide the animation process. We discover this landmark can not only ensure the accurate motion alignment between the reference portrait and target motion during inference but also increase the ability to portray exaggerated expressions (i.e., large pupil movements) and avoid identity leakage. Then, we propose a facial fine-grained loss to improve the model's ability of subtle expression perception and reference portrait appearance reconstruction by using both expression and facial masks. Accordingly, our method demonstrates significant performance in controlling the expression of freestyle portraits, including real humans, cartoons, sculptures, and even animals. By leveraging a simple and effective progressive generation strategy, we extend our model to stable long-term animation, thus increasing its potential application value. To address the lack of a benchmark for this field, we introduce EmojiBench, a comprehensive benchmark comprising diverse portrait images, driving videos, and landmarks. We show extensive evaluations on EmojiBench to verify the superiority of Follow-Your-Emoji.
Authors: Fang Chen, Gourav Datta, Mujahid Al Rafi, Hyeran Jeon, Meng Tang
Abstract: The expansion of neural network sizes and the enhancement of image resolution through modern camera sensors result in heightened memory and power demands for neural networks. Reducing peak memory, which is the maximum memory consumed during the execution of a neural network, is critical to deploy neural networks on edge devices with limited memory budget. A naive approach to reducing peak memory is aggressive down-sampling of feature maps via pooling with large stride, which often results in unacceptable degradation in network performance. To mitigate this problem, we propose residual encoded distillation (ReDistill) for peak memory reduction in a teacher-student framework, in which a student network with less memory is derived from the teacher network using aggressive pooling. We apply our distillation method to multiple problems in computer vision including image classification and diffusion based image generation. For image classification, our method yields 2x-3.2x measured peak memory on an edge GPU with negligible degradation in accuracy for most CNN based architectures. Additionally, our method yields improved test accuracy for tiny vision transformer (ViT) based models distilled from large CNN based teacher architectures. For diffusion-based image generation, our proposed distillation method yields a denoising network with 4x lower theoretical peak memory while maintaining decent diversity and fidelity for image generation. Experiments demonstrate our method's superior performance compared to other feature-based and response-based distillation methods.
Authors: Nan Zhang, Xidan Zhang, Jianing Wei, Fangjun Wang, Zhiming Tan
Abstract: This report describes the winning solution to the WeatherProof Dataset Challenge (CVPR 2024 UG2+ Track 3). Details regarding the challenge are available at https://cvpr2024ug2challenge.github.io/track3.html. We propose an enhanced semantic segmentation pipeline for this challenge. Firstly, we improve semantic segmentation models, using backbone pretrained with Depth Anything to improve UperNet model and SETRMLA model, and adding language guidance based on both weather and category information to InternImage model. Secondly, we introduce a new dataset WeatherProofExtra with wider viewing angle and employ data augmentation methods, including adverse weather and super-resolution. Finally, effective training strategies and ensemble method are applied to improve final performance further. Our solution is ranked 1st on the final leaderboard. Code will be available at https://github.com/KaneiGi/WeatherProofChallenge.
URLs: https://cvpr2024ug2challenge.github.io/track3.html., https://github.com/KaneiGi/WeatherProofChallenge.
Authors: Ruipu Wu, Jifei Che, Han Li, Chengjing Wu, Ting Liu, Luoqi Liu
Abstract: Video panoptic segmentation is an advanced task that extends panoptic segmentation by applying its concept to video sequences. In the hope of addressing the challenge of video panoptic segmentation in diverse conditions, We utilize DVIS++ as our baseline model and enhance it by introducing a comprehensive approach centered on the query-wise ensemble, supplemented by additional techniques. Our proposed approach achieved a VPQ score of 57.01 on the VIPSeg test set, and ranked 3rd in the VPS track of the 3rd Pixel-level Video Understanding in the Wild Challenge.
Authors: Salvatore Esposito, Qingshan Xu, Kacper Kania, Charlie Hewitt, Octave Mariotti, Lohit Petikam, Julien Valentin, Arno Onken, Oisin Mac Aodha
Abstract: We introduce a new generative approach for synthesizing 3D geometry and images from single-view collections. Most existing approaches predict volumetric density to render multi-view consistent images. By employing volumetric rendering using neural radiance fields, they inherit a key limitation: the generated geometry is noisy and unconstrained, limiting the quality and utility of the output meshes. To address this issue, we propose GeoGen, a new SDF-based 3D generative model trained in an end-to-end manner. Initially, we reinterpret the volumetric density as a Signed Distance Function (SDF). This allows us to introduce useful priors to generate valid meshes. However, those priors prevent the generative model from learning details, limiting the applicability of the method to real-world scenarios. To alleviate that problem, we make the transformation learnable and constrain the rendered depth map to be consistent with the zero-level set of the SDF. Through the lens of adversarial training, we encourage the network to produce higher fidelity details on the output meshes. For evaluation, we introduce a synthetic dataset of human avatars captured from 360-degree camera angles, to overcome the challenges presented by real-world datasets, which often lack 3D consistency and do not cover all camera angles. Our experiments on multiple datasets show that GeoGen produces visually and quantitatively better geometry than the previous generative models based on neural radiance fields.
Authors: Qihao Liu, Yi Zhang, Song Bai, Adam Kortylewski, Alan Yuille
Abstract: We present DIRECT-3D, a diffusion-based 3D generative model for creating high-quality 3D assets (represented by Neural Radiance Fields) from text prompts. Unlike recent 3D generative models that rely on clean and well-aligned 3D data, limiting them to single or few-class generation, our model is directly trained on extensive noisy and unaligned `in-the-wild' 3D assets, mitigating the key challenge (i.e., data scarcity) in large-scale 3D generation. In particular, DIRECT-3D is a tri-plane diffusion model that integrates two innovations: 1) A novel learning framework where noisy data are filtered and aligned automatically during the training process. Specifically, after an initial warm-up phase using a small set of clean data, an iterative optimization is introduced in the diffusion process to explicitly estimate the 3D pose of objects and select beneficial data based on conditional density. 2) An efficient 3D representation that is achieved by disentangling object geometry and color features with two separate conditional diffusion models that are optimized hierarchically. Given a prompt input, our model generates high-quality, high-resolution, realistic, and complex 3D objects with accurate geometric details in seconds. We achieve state-of-the-art performance in both single-class generation and text-to-3D generation. We also demonstrate that DIRECT-3D can serve as a useful 3D geometric prior of objects, for example to alleviate the well-known Janus problem in 2D-lifting methods such as DreamFusion. The code and models are available for research purposes at: https://github.com/qihao067/direct3d.
Authors: Fangfu Liu, Hanyang Wang, Shunyu Yao, Shengjun Zhang, Jie Zhou, Yueqi Duan
Abstract: In recent years, there has been rapid development in 3D generation models, opening up new possibilities for applications such as simulating the dynamic movements of 3D objects and customizing their behaviors. However, current 3D generative models tend to focus only on surface features such as color and shape, neglecting the inherent physical properties that govern the behavior of objects in the real world. To accurately simulate physics-aligned dynamics, it is essential to predict the physical properties of materials and incorporate them into the behavior prediction process. Nonetheless, predicting the diverse materials of real-world objects is still challenging due to the complex nature of their physical attributes. In this paper, we propose \textbf{Physics3D}, a novel method for learning various physical properties of 3D objects through a video diffusion model. Our approach involves designing a highly generalizable physical simulation system based on a viscoelastic material model, which enables us to simulate a wide range of materials with high-fidelity capabilities. Moreover, we distill the physical priors from a video diffusion model that contains more understanding of realistic object materials. Extensive experiments demonstrate the effectiveness of our method with both elastic and plastic materials. Physics3D shows great potential for bridging the gap between the physical world and virtual neural space, providing a better integration and application of realistic physical principles in virtual environments. Project page: https://liuff19.github.io/Physics3D.
Authors: Joseph Sobek, Jose R. Medina Inojosa, Betsy J. Medina Inojosa, S. M. Rassoulinejad-Mousavi, Gian Marco Conte, Francisco Lopez-Jimenez, Bradley J. Erickson
Abstract: Artificial intelligence-enhanced identification of organs, lesions, and other structures in medical imaging is typically done using convolutional neural networks (CNNs) designed to make voxel-accurate segmentations of the region of interest. However, the labels required to train these CNNs are time-consuming to generate and require attention from subject matter experts to ensure quality. For tasks where voxel-level precision is not required, object detection models offer a viable alternative that can reduce annotation effort. Despite this potential application, there are few options for general purpose object detection frameworks available for 3-D medical imaging. We report on MedYOLO, a 3-D object detection framework using the one-shot detection method of the YOLO family of models and designed for use with medical imaging. We tested this model on four different datasets: BRaTS, LIDC, an abdominal organ Computed Tomography (CT) dataset, and an ECG-gated heart CT dataset. We found our models achieve high performance on commonly present medium and large-sized structures such as the heart, liver, and pancreas even without hyperparameter tuning. However, the models struggle with very small or rarely present structures.
Authors: Xiang Liu, Jiahong Chen, Bin Chen, Zimo Liu, Baoyi An, Shu-Tao Xia, Zhi Wang
Abstract: Displaying high-quality images on edge devices, such as augmented reality devices, is essential for enhancing the user experience. However, these devices often face power consumption and computing resource limitations, making it challenging to apply many deep learning-based image compression algorithms in this field. Implicit Neural Representation (INR) for image compression is an emerging technology that offers two key benefits compared to cutting-edge autoencoder models: low computational complexity and parameter-free decoding. It also outperforms many traditional and early neural compression methods in terms of quality. In this study, we introduce a new Mixed AutoRegressive Model (MARM) to significantly reduce the decoding time for the current INR codec, along with a new synthesis network to enhance reconstruction quality. MARM includes our proposed AutoRegressive Upsampler (ARU) blocks, which are highly computationally efficient, and ARM from previous work to balance decoding time and reconstruction quality. We also propose enhancing ARU's performance using a checkerboard two-stage decoding strategy. Moreover, the ratio of different modules can be adjusted to maintain a balance between quality and speed. Comprehensive experiments demonstrate that our method significantly improves computational efficiency while preserving image quality. With different parameter settings, our method can achieve over a magnitude acceleration in decoding time without industrial level optimization, or achieve state-of-the-art reconstruction quality compared with other INR codecs. To the best of our knowledge, our method is the first INR-based codec comparable with Hyperprior in both decoding speed and quality while maintaining low complexity.
Authors: Anke Tang, Li Shen, Yong Luo, Nan Yin, Lefei Zhang, Dacheng Tao
Abstract: Merging various task-specific Transformer-based models trained on different tasks into a single unified model can execute all the tasks concurrently. Previous methods, exemplified by task arithmetic, have been proven to be both effective and scalable. Existing methods have primarily focused on seeking a static optimal solution within the original model parameter space. A notable challenge is mitigating the interference between parameters of different models, which can substantially deteriorate performance. In this paper, we propose to merge most of the parameters while upscaling the MLP of the Transformer layers to a weight-ensembling mixture of experts (MoE) module, which can dynamically integrate shared and task-specific knowledge based on the input, thereby providing a more flexible solution that can adapt to the specific needs of each instance. Our key insight is that by identifying and separating shared knowledge and task-specific knowledge, and then dynamically integrating them, we can mitigate the parameter interference problem to a great extent. We conduct the conventional multi-task model merging experiments and evaluate the generalization and robustness of our method. The results demonstrate the effectiveness of our method and provide a comprehensive understanding of our method. The code is available at https://github.com/tanganke/weight-ensembling_MoE
Authors: Yejun Yoon, Seunghyun Yoon, Kunwoo Park
Abstract: This paper addresses the critical challenge of assessing the representativeness of news thumbnail images, which often serve as the first visual engagement for readers when an article is disseminated on social media. We focus on whether a news image represents the actors discussed in the news text. To serve the challenge, we introduce NewsTT, a manually annotated dataset of 1000 news thumbnail images and text pairs. We found that the pretrained vision and language models, such as BLIP-2, struggle with this task. Since news subjects frequently involve named entities or proper nouns, the pretrained models could have a limited capability to match news actors' visual and textual appearances. We hypothesize that learning to contrast news text with its counterfactual, of which named entities are replaced, can enhance the cross-modal matching ability of vision and language models. We propose CFT-CLIP, a contrastive learning framework that updates vision and language bi-encoders according to the hypothesis. We found that our simple method can boost the performance for assessing news thumbnail representativeness, supporting our assumption. Code and data can be accessed at https://github.com/ssu-humane/news-images-acl24.
Authors: Ahmed Imtiaz Humayun, Randall Balestriero, Richard Baraniuk
Abstract: Grokking, or delayed generalization, is a phenomenon where generalization in a deep neural network (DNN) occurs long after achieving near zero training error. Previous studies have reported the occurrence of grokking in specific controlled settings, such as DNNs initialized with large-norm parameters or transformers trained on algorithmic datasets. We demonstrate that grokking is actually much more widespread and materializes in a wide range of practical settings, such as training of a convolutional neural network (CNN) on CIFAR10 or a Resnet on Imagenette. We introduce the new concept of delayed robustness, whereby a DNN groks adversarial examples and becomes robust, long after interpolation and/or generalization. We develop an analytical explanation for the emergence of both delayed generalization and delayed robustness based on the local complexity of a DNN's input-output mapping. Our local complexity measures the density of so-called linear regions (aka, spline partition regions) that tile the DNN input space and serves as a utile progress measure for training. We provide the first evidence that, for classification problems, the linear regions undergo a phase transition during training whereafter they migrate away from the training samples (making the DNN mapping smoother there) and towards the decision boundary (making the DNN mapping less smooth there). Grokking occurs post phase transition as a robust partition of the input space thanks to the linearization of the DNN mapping around the training points. Website: https://bit.ly/grok-adversarial
Authors: Joseph Cho, Fachrina Dewi Puspitasari, Sheng Zheng, Jingyao Zheng, Lik-Hang Lee, Tae-Ho Kim, Choong Seon Hong, Chaoning Zhang
Abstract: The evolution of video generation from text, starting with animating MNIST numbers to simulating the physical world with Sora, has progressed at a breakneck speed over the past seven years. While often seen as a superficial expansion of the predecessor text-to-image generation model, text-to-video generation models are developed upon carefully engineered constituents. Here, we systematically discuss these elements consisting of but not limited to core building blocks (vision, language, and temporal) and supporting features from the perspective of their contributions to achieving a world model. We employ the PRISMA framework to curate 97 impactful research articles from renowned scientific databases primarily studying video synthesis using text conditions. Upon minute exploration of these manuscripts, we observe that text-to-video generation involves more intricate technologies beyond the plain extension of text-to-image generation. Our additional review into the shortcomings of Sora-generated videos pinpoints the call for more in-depth studies in various enabling aspects of video generation such as dataset, evaluation metric, efficient architecture, and human-controlled generation. Finally, we conclude that the study of the text-to-video generation may still be in its infancy, requiring contribution from the cross-discipline research community towards its advancement as the first step to realize artificial general intelligence (AGI).
Authors: Zichen Wang, Xi Deng, Ziyi Zhang, Wenzel Jakob, Steve Marschner
Abstract: We present a simple algorithm for differentiable rendering of surfaces represented by Signed Distance Fields (SDF), which makes it easy to integrate rendering into gradient-based optimization pipelines. To tackle visibility-related derivatives that make rendering non-differentiable, existing physically based differentiable rendering methods often rely on elaborate guiding data structures or reparameterization with a global impact on variance. In this article, we investigate an alternative that embraces nonzero bias in exchange for low variance and architectural simplicity. Our method expands the lower-dimensional boundary integral into a thin band that is easy to sample when the underlying surface is represented by an SDF. We demonstrate the performance and robustness of our formulation in end-to-end inverse rendering tasks, where it obtains results that are competitive with or superior to existing work.
Authors: Desislav Ivanov, Carlo Alberto Barbano, Marco Grangetto
Abstract: Traditional staining normalization approaches, e.g. Macenko, typically rely on the choice of a single representative reference image, which may not adequately account for the diverse staining patterns of datasets collected in practical scenarios. In this study, we introduce a novel approach that leverages multiple reference images to enhance robustness against stain variation. Our method is parameter-free and can be adopted in existing computational pathology pipelines with no significant changes. We evaluate the effectiveness of our method through experiments using a deep-learning pipeline for automatic nuclei segmentation on colorectal images. Our results show that by leveraging multiple reference images, better results can be achieved when generalizing to external data, where the staining can widely differ from the training set.
Authors: Advait Gadhikar, Sree Harsha Nelaturu, Rebekka Burkholz
Abstract: The success of iterative pruning methods in achieving state-of-the-art sparse networks has largely been attributed to improved mask identification and an implicit regularization induced by pruning. We challenge this hypothesis and instead posit that their repeated cyclic training schedules enable improved optimization. To verify this, we show that pruning at initialization is significantly boosted by repeated cyclic training, even outperforming standard iterative pruning methods. The dominant mechanism how this is achieved, as we conjecture, can be attributed to a better exploration of the loss landscape leading to a lower training loss. However, at high sparsity, repeated cyclic training alone is not enough for competitive performance. A strong coupling between learnt parameter initialization and mask seems to be required. Standard methods obtain this coupling via expensive pruning-training iterations, starting from a dense network. To achieve this with sparse training instead, we propose SCULPT-ing, i.e., repeated cyclic training of any sparse mask followed by a single pruning step to couple the parameters and the mask, which is able to match the performance of state-of-the-art iterative pruning methods in the high sparsity regime at reduced computational cost.
Authors: Faisal Tareque Shohan, Mir Tafseer Nayeem, Samsul Islam, Abu Ubaida Akash, Shafiq Joty
Abstract: Millions of news articles published online daily can overwhelm readers. Headlines and entity (topic) tags are essential for guiding readers to decide if the content is worth their time. While headline generation has been extensively studied, tag generation remains largely unexplored, yet it offers readers better access to topics of interest. The need for conciseness in capturing readers' attention necessitates improved content selection strategies for identifying salient and relevant segments within lengthy articles, thereby guiding language models effectively. To address this, we propose to leverage auxiliary information such as images and captions embedded in the articles to retrieve relevant sentences and utilize instruction tuning with variations to generate both headlines and tags for news articles in a multilingual context. To make use of the auxiliary information, we have compiled a dataset named XL-HeadTags, which includes 20 languages across 6 diverse language families. Through extensive evaluation, we demonstrate the effectiveness of our plug-and-play multimodal-multilingual retrievers for both tasks. Additionally, we have developed a suite of tools for processing and evaluating multilingual texts, significantly contributing to the research community by enabling more accurate and efficient analysis across languages.