Authors: Daqian Shao, Lukas Fesser, Marta Kwiatkowska
Robustness certification, which aims to formally certify the predictions of neural networks against adversarial inputs, has become an integral part of important tool for safety-critical applications. Despite considerable progress, existing certification methods are limited to elementary architectures, such as convolutional networks, recurrent networks and recently Transformers, on benchmark datasets such as MNIST. In this paper, we focus on the robustness certification of scene text recognition (STR), which is a complex and extensively deployed image-based sequence prediction problem. We tackle three types of STR model architectures, including the standard STR pipelines and the Vision Transformer. We propose STR-Cert, the first certification method for STR models, by significantly extending the DeepPoly polyhedral verification framework via deriving novel polyhedral bounds and algorithms for key STR model components. Finally, we certify and compare STR models on six datasets, demonstrating the efficiency and scalability of robustness certification, particularly for the Vision Transformer.
Authors: Toby Chong, Alina Chadwick, I-chao Shen, Haoran Xie, Takeo Igarashi
In this paper, we present a cosmetic-specific skin image dataset. It consists of skin images from $45$ patches ($5$ skin patches each from $9$ participants) of size $8mm^*8mm$ under three cosmetic products (i.e., foundation, blusher, and highlighter). We designed a novel capturing device inspired by Light Stage. Using the device, we captured over $600$ images of each skin patch under diverse lighting conditions in $30$ seconds. We repeated the process for the same skin patch under three cosmetic products. Finally, we demonstrate the viability of the dataset with an image-to-image translation-based pipeline for cosmetic rendering and compared our data-driven approach to an existing cosmetic rendering method.
Authors: Sankeerth Durvasula, Adrian Zhao, Fan Chen, Ruofan Liang, Pawan Kumar Sanjaya, Nandita Vijaykumar
Differentiable rendering is a technique used in an important emerging class of visual computing applications that involves representing a 3D scene as a model that is trained from 2D images using gradient descent. Recent works (e.g. 3D Gaussian Splatting) use a rasterization pipeline to enable rendering high quality photo-realistic imagery at high speeds from these learned 3D models. These methods have been demonstrated to be very promising, providing state-of-art quality for many important tasks. However, training a model to represent a scene is still a time-consuming task even when using powerful GPUs. In this work, we observe that the gradient computation phase during training is a significant bottleneck on GPUs due to the large number of atomic operations that need to be processed. These atomic operations overwhelm atomic units in the L2 partitions causing stalls. To address this challenge, we leverage the observations that during the gradient computation: (1) for most warps, all threads atomically update the same memory locations; and (2) warps generate varying amounts of atomic traffic (since some threads may be inactive). We propose DISTWAR, a software-approach to accelerate atomic operations based on two key ideas: First, we enable warp-level reduction of threads at the SM sub-cores using registers to leverage the locality in intra-warp atomic updates. Second, we distribute the atomic computation between the warp-level reduction at the SM and the L2 atomic units to increase the throughput of atomic computation. Warps with many threads performing atomic updates to the same memory locations are scheduled at the SM, and the rest using L2 atomic units. We implement DISTWAR using existing warp-level primitives. We evaluate DISTWAR on widely used raster-based differentiable rendering workloads. We demonstrate significant speedups of 2.44x on average (up to 5.7x).
Authors: Ziyun Li, Christoph Meinel, Haojin Yang
Generalized Class Discovery (GCD) plays a pivotal role in discerning both known and unknown categories from unlabeled datasets by harnessing the insights derived from a labeled set comprising recognized classes. A significant limitation in prevailing GCD methods is their presumption of an equitably distributed category occurrence in unlabeled data. Contrary to this assumption, visual classes in natural environments typically exhibit a long-tailed distribution, with known or prevalent categories surfacing more frequently than their rarer counterparts. Our research endeavors to bridge this disconnect by focusing on the long-tailed Generalized Category Discovery (Long-tailed GCD) paradigm, which echoes the innate imbalances of real-world unlabeled datasets. In response to the unique challenges posed by Long-tailed GCD, we present a robust methodology anchored in two strategic regularizations: (i) a reweighting mechanism that bolsters the prominence of less-represented, tail-end categories, and (ii) a class prior constraint that aligns with the anticipated class distribution. Comprehensive experiments reveal that our proposed method surpasses previous state-of-the-art GCD methods by achieving an improvement of approximately 6 - 9% on ImageNet100 and competitive performance on CIFAR100.
Authors: Ziyun Li, Ben Dai, Furkan Simsek, Christoph Meinel, Haojin Yang
Generalized class discovery (GCD) aims to infer known and unknown categories in an unlabeled dataset leveraging prior knowledge of a labeled set comprising known classes. Existing research implicitly/explicitly assumes that the frequency of occurrence for each category, whether known or unknown, is approximately the same in the unlabeled data. However, in nature, we are more likely to encounter known/common classes than unknown/uncommon ones, according to the long-tailed property of visual classes. Therefore, we present a challenging and practical problem, Imbalanced Generalized Category Discovery (ImbaGCD), where the distribution of unlabeled data is imbalanced, with known classes being more frequent than unknown ones. To address these issues, we propose ImbaGCD, A novel optimal transport-based expectation maximization framework that accomplishes generalized category discovery by aligning the marginal class prior distribution. ImbaGCD also incorporates a systematic mechanism for estimating the imbalanced class prior distribution under the GCD setup. Our comprehensive experiments reveal that ImbaGCD surpasses previous state-of-the-art GCD methods by achieving an improvement of approximately 2 - 4% on CIFAR-100 and 15 - 19% on ImageNet-100, indicating its superior effectiveness in solving the Imbalanced GCD problem.
Authors: Atah Nuh Mih, Hung Cao, Asfia Kawnine, Monica Wachowicz
Resource constraints have restricted several EdgeAI applications to machine learning inference approaches, where models are trained on the cloud and deployed to the edge device. This poses challenges such as bandwidth, latency, and privacy associated with storing data off-site for model building. Training on the edge device can overcome these challenges by eliminating the need to transfer data to another device for storage and model development. On-device training also provides robustness to data variations as models can be retrained on newly acquired data to improve performance. We, therefore, propose a lightweight EdgeAI architecture modified from Xception, for on-device training in a resource-constraint edge environment. We evaluate our model on a PCB defect detection task and compare its performance against existing lightweight models - MobileNetV2, EfficientNetV2B0, and MobileViT-XXS. The results of our experiment show that our model has a remarkable performance with a test accuracy of 73.45% without pre-training. This is comparable to the test accuracy of non-pre-trained MobileViT-XXS (75.40%) and much better than other non-pre-trained models (MobileNetV2 - 50.05%, EfficientNetV2B0 - 54.30%). The test accuracy of our model without pre-training is comparable to pre-trained MobileNetV2 model - 75.45% and better than pre-trained EfficientNetV2B0 model - 58.10%. In terms of memory efficiency, our model performs better than EfficientNetV2B0 and MobileViT-XXS. We find that the resource efficiency of machine learning models does not solely depend on the number of parameters but also depends on architectural considerations. Our method can be applied to other resource-constraint applications while maintaining significant performance.
Authors: Ziqi Yuan, Liyuan Wang, Wenbo Ding, Xingxing Zhang, Jiachen Zhong, Jianyong Ai, Jianmin Li, Jun Zhu
In real-world applications, an object detector often encounters object instances from new classes and needs to accommodate them effectively. Previous work formulated this critical problem as incremental object detection (IOD), which assumes the object instances of new classes to be fully annotated in incremental data. However, as supervisory signals are usually rare and expensive, the supervised IOD may not be practical for implementation. In this work, we consider a more realistic setting named semi-supervised IOD (SSIOD), where the object detector needs to learn new classes incrementally from a few labelled data and massive unlabelled data without catastrophic forgetting of old classes. A commonly-used strategy for supervised IOD is to encourage the current model (as a student) to mimic the behavior of the old model (as a teacher), but it generally fails in SSIOD because a dominant number of object instances from old and new classes are coexisting and unlabelled, with the teacher only recognizing a fraction of them. Observing that learning only the classes of interest tends to preclude detection of other classes, we propose to bridge the coexistence of unlabelled classes by constructing two teacher models respectively for old and new classes, and using the concatenation of their predictions to instruct the student. This approach is referred to as DualTeacher, which can serve as a strong baseline for SSIOD with limited resource overhead and no extra hyperparameters. We build various benchmarks for SSIOD and perform extensive experiments to demonstrate the superiority of our approach (e.g., the performance lead is up to 18.28 AP on MS-COCO). Our code is available at \url{https://github.com/chuxiuhong/DualTeacher}.
Authors: Amirreza Hashemi
This study presents a comprehensive evaluation of tools available on the HuggingFace platform for two pivotal applications in artificial intelligence: image segmentation and voice conversion. The primary objective was to identify the top three tools within each category and subsequently install and configure these tools on Linux systems. We leveraged the power of pre-trained segmentation models such as SAM and DETR Model with ResNet-50 backbone for image segmentation, and the so-vits-svc-fork model for voice conversion. This paper delves into the methodologies and challenges encountered during the implementation process, and showcases the successful combination of video segmentation and voice conversion in a unified project named AutoVisual Fusion Suite.
Authors: Francisco Troncoso-Pastoriza, Pablo Eguía-Oller, Rebeca P. Díaz-Redondo, Enrique Granada-Álvarez
In this paper we introduce a method that supports the detection, identification and localization of lamps in a building, with the main goal of automatically feeding its energy model by means of Building Information Modeling (BIM) methods. The proposed method, thus, provides useful information to apply energy-saving strategies to reduce energy consumption in the building sector through the correct management of the lighting infrastructure. Based on the unique geometry and brightness of lamps and the use of only greyscale images, our methodology is able to obtain accurate results despite its low computational needs, resulting in near-real-time processing. The main novelty is that the focus of the candidate search is not over the entire image but instead only on a limited region that summarizes the specific characteristics of the lamp. The information obtained from our approach was used on the Green Building XML Schema to illustrate the automatic generation of BIM data from the results of the algorithm.
Authors: Vikas Singh
Noise is inevitably common in digital images, leading to visual image deterioration. Therefore, a suitable filtering method is required to lessen the noise while preserving the image features (edges, corners, etc.). This paper presents the efficient type-2 fuzzy weighted mean filter with an adaptive threshold to remove the SAP noise. The present filter has two primary steps: The first stage categorizes images as lightly, medium, and heavily corrupted based on an adaptive threshold by comparing the M-ALD of processed pixels with the upper and lower MF of the type-2 fuzzy identifier. The second stage eliminates corrupted pixels by computing the appropriate weight using GMF with the mean and variance of the uncorrupted pixels in the filter window. Simulation results vividly show that the obtained denoised images preserve image features, i.e., edges, corners, and other sharp structures, compared with different filtering methods.
Authors: Olaya Álvarez-Tuñón, Yury Brodskiy, Erdal Kayacan
This paper overviews different pose representations and metric functions in visual odometry (VO) networks. The performance of VO networks heavily relies on how their architecture encodes the information. The choice of pose representation and loss function significantly impacts network convergence and generalization. We investigate these factors in the VO network DeepVO by implementing loss functions based on Euler, quaternion, and chordal distance and analyzing their influence on performance. The results of this study provide insights into how loss functions affect the designing of efficient and accurate VO networks for camera motion estimation. The experiments illustrate that a distance that complies with the mathematical requirements of a metric, such as the chordal distance, provides better generalization and faster convergence. The code for the experiments can be found at https://github.com/remaro-network/Loss_VO_right
Authors: Xisheng Li, Wei Li, Pinhao Song, Mingjun Zhang, Jie Zhou
The inherent characteristics and light fluctuations of water bodies give rise to the huge difference between different layers and regions in underwater environments. When the test set is collected in a different marine area from the training set, the issue of domain shift emerges, significantly compromising the model's ability to generalize. The Domain Adversarial Learning (DAL) training strategy has been previously utilized to tackle such challenges. However, DAL heavily depends on manually one-hot domain labels, which implies no difference among the samples in the same domain. Such an assumption results in the instability of DAL. This paper introduces the concept of Domain Similarity-Perceived Label Assignment (DSP). The domain label for each image is regarded as its similarity to the specified domains. Through domain-specific data augmentation techniques, we achieved state-of-the-art results on the underwater cross-domain object detection benchmark S-UODAC2020. Furthermore, we validated the effectiveness of our method in the Cityscapes dataset.
Authors: Tresor Y. Koffi, Youssef Mourchid, Mohammed Hindawi, Yohan Dupuis
Falls among individuals, especially the elderly population, can lead to serious injuries and complications. Detecting impact moments within a fall event is crucial for providing timely assistance and minimizing the negative consequences. In this work, we aim to address this challenge by applying thorough preprocessing techniques to the multisensor dataset, the goal is to eliminate noise and improve data quality. Furthermore, we employ a feature selection process to identify the most relevant features derived from the multisensor UP-FALL dataset, which in turn will enhance the performance and efficiency of machine learning models. We then evaluate the efficiency of various machine learning models in detecting the impact moment using the resulting data information from multiple sensors. Through extensive experimentation, we assess the accuracy of our approach using various evaluation metrics. Our results achieve high accuracy rates in impact detection, showcasing the power of leveraging multisensor data for fall detection tasks. This highlights the potential of our approach to enhance fall detection systems and improve the overall safety and well-being of individuals at risk of falls.
Authors: Xueyuan Yang, Chao Yao, Xiaojuan Ban
Leveraging wearable devices for motion reconstruction has emerged as an economical and viable technique. Certain methodologies employ sparse Inertial Measurement Units (IMUs) on the human body and harness data-driven strategies to model human poses. However, the reconstruction of motion based solely on sparse IMUs data is inherently fraught with ambiguity, a consequence of numerous identical IMU readings corresponding to different poses. In this paper, we explore the spatial importance of multiple sensors, supervised by text that describes specific actions. Specifically, uncertainty is introduced to derive weighted features for each IMU. We also design a Hierarchical Temporal Transformer (HTT) and apply contrastive learning to achieve precise temporal and feature alignment of sensor data with textual semantics. Experimental results demonstrate our proposed approach achieves significant improvements in multiple metrics compared to existing methods. Notably, with textual supervision, our method not only differentiates between ambiguous actions such as sitting and standing but also produces more precise and natural motion.
Authors: Lin Zhang, Linghan Xu, Saman Motamed, Shayok Chakraborty, Fernando De la Torre
Unsupervised domain adaptation (UDA) for image classification has made remarkable progress in transferring classification knowledge from a labeled source domain to an unlabeled target domain, thanks to effective domain alignment techniques. Recently, in order to further improve performance on a target domain, many Single-Target Active Domain Adaptation (ST-ADA) methods have been proposed to identify and annotate the salient and exemplar target samples. However, it requires one model to be trained and deployed for each target domain and the domain label associated with each test sample. This largely restricts its application in the ubiquitous scenarios with multiple target domains. Therefore, we propose a Multi-Target Active Domain Adaptation (MT-ADA) framework for image classification, named D3GU, to simultaneously align different domains and actively select samples from them for annotation. This is the first research effort in this field to our best knowledge. D3GU applies Decomposed Domain Discrimination (D3) during training to achieve both source-target and target-target domain alignments. Then during active sampling, a Gradient Utility (GU) score is designed to weight every unlabeled target image by its contribution towards classification and domain alignment tasks, and is further combined with KMeans clustering to form GU-KMeans for diverse image sampling. Extensive experiments on three benchmark datasets, Office31, OfficeHome, and DomainNet, have been conducted to validate consistently superior performance of D3GU for MT-ADA.
Authors: Siddharth Tiwari
The segmentation of medical images is important for the improvement and creation of healthcare systems, particularly for early disease detection and treatment planning. In recent years, the use of convolutional neural networks (CNNs) and other state-of-the-art methods has greatly advanced medical image segmentation. However, CNNs have been found to struggle with learning long-range dependencies and capturing global context due to the limitations of convolution operations. In this paper, we explore the use of transformers and CNNs for medical image segmentation and propose a hybrid architecture that combines the ability of transformers to capture global dependencies with the ability of CNNs to capture low-level spatial details. We compare various architectures and configurations and conduct multiple experiments to evaluate their effectiveness.
Authors: GeonU Kim, Kim Youwang, Tae-Hyun Oh
We present FPRF, a feed-forward photorealistic style transfer method for large-scale 3D neural radiance fields. FPRF stylizes large-scale 3D scenes with arbitrary, multiple style reference images without additional optimization while preserving multi-view appearance consistency. Prior arts required tedious per-style/-scene optimization and were limited to small-scale 3D scenes. FPRF efficiently stylizes large-scale 3D scenes by introducing a style-decomposed 3D neural radiance field, which inherits AdaIN's feed-forward stylization machinery, supporting arbitrary style reference images. Furthermore, FPRF supports multi-reference stylization with the semantic correspondence matching and local AdaIN, which adds diverse user control for 3D scene styles. FPRF also preserves multi-view consistency by applying semantic matching and style transfer processes directly onto queried features in 3D space. In experiments, we demonstrate that FPRF achieves favorable photorealistic quality 3D scene stylization for large-scale scenes with diverse reference images. Project page: https://kim-geonu.github.io/FPRF/
Authors: Marcellus Amadeus, William Alberto Cruz Castañeda, André Felipe Zanella, Felipe Rodrigues Perche Mahlow
Generative AI has become pervasive in society, witnessing significant advancements in various domains. Particularly in the realm of Text-to-Image (TTI) models, Latent Diffusion Models (LDMs), showcase remarkable capabilities in generating visual content based on textual prompts. This paper addresses the potential of LDMs in representing local cultural concepts, historical figures, and endangered species. In this study, we use the cultural heritage of Rio Grande do Sul (RS), Brazil, as an illustrative case. Our objective is to contribute to the broader understanding of how generative models can help to capture and preserve the cultural and historical identity of regions. The paper outlines the methodology, including subject selection, dataset creation, and the fine-tuning process. The results showcase the images generated, alongside the challenges and feasibility of each concept. In conclusion, this work shows the power of these models to represent and preserve unique aspects of diverse regions and communities.
Authors: Erik Isai Valle Salgado, Chen Li, Yaqi Han, Linchao Shi, Xinghui Li
Ensemble methods exploit the availability of a given number of classifiers or detectors trained in single or multiple source domains and tasks to address machine learning problems such as domain adaptation or multi-source transfer learning. Existing research measures the domain distance between the sources and the target dataset, trains multiple networks on the same data with different samples per class, or combines predictions from models trained under varied hyperparameters and settings. Their solutions enhanced the performance on small or tail categories but hurt the rest. To this end, we propose a modified consensus focus for semi-supervised and long-tailed object detection. We introduce a voting system based on source confidence that spots the contribution of each model in a consensus, lets the user choose the relevance of each class in the target label space so that it relaxes minority bounding boxes suppression, and combines multiple models' results without discarding the poisonous networks. Our tests on synthetic driving datasets retrieved higher confidence and more accurate bounding boxes than the NMS, soft-NMS, and WBF.
Authors: Kevin Van Vorst, Li Shen
Self-supervised learning has become a popular way to pretrain a deep learning model and then transfer it to perform downstream tasks. However, most of these methods are developed on large-scale image datasets that contain natural objects with clear textures, outlines, and distinct color contrasts. It remains uncertain whether these methods are equally effective for medical imaging, where the regions of interest often blend subtly and indistinctly with the surrounding tissues. In this study, we propose an alternative method that uses contralateral mammograms to train a neural network to encode similar embeddings when a pair contains both normal images and different embeddings when a pair contains normal and abnormal images. Our approach leverages the natural symmetry of human body as weak labels to learn to distinguish abnormal lesions from background tissues in a fully unsupervised manner. Our findings suggest that it's feasible by incorporating soft labels derived from the Euclidean distances between the embeddings of the image pairs into the Siamese network loss. Our method demonstrates superior performance in mammogram patch classification compared to existing self-supervised learning methods. This approach not only leverages a vast amount of image data effectively but also minimizes reliance on costly labels, a significant advantage particularly in the field of medical imaging.
Authors: Chenbin Pan, Burhaneddin Yaman, Tommaso Nesti, Abhirup Mallik, Alessandro G Allievi, Senem Velipasalar, Liu Ren
Autonomous driving is a complex and challenging task that aims at safe motion planning through scene understanding and reasoning. While vision-only autonomous driving methods have recently achieved notable performance, through enhanced scene understanding, several key issues, including lack of reasoning, low generalization performance and long-tail scenarios, still need to be addressed. In this paper, we present VLP, a novel Vision-Language-Planning framework that exploits language models to bridge the gap between linguistic understanding and autonomous driving. VLP enhances autonomous driving systems by strengthening both the source memory foundation and the self-driving car's contextual understanding. VLP achieves state-of-the-art end-to-end planning performance on the challenging NuScenes dataset by achieving 35.9\% and 60.5\% reduction in terms of average L2 error and collision rates, respectively, compared to the previous best method. Moreover, VLP shows improved performance in challenging long-tail scenarios and strong generalization capabilities when faced with new urban environments.
Authors: Chaoyang Wang, Peiye Zhuang, Aliaksandr Siarohin, Junli Cao, Guocheng Qian, Hsin-Ying Lee, Sergey Tulyakov
Dynamic novel view synthesis aims to capture the temporal evolution of visual content within videos. Existing methods struggle to distinguishing between motion and structure, particularly in scenarios where camera poses are either unknown or constrained compared to object motion. Furthermore, with information solely from reference images, it is extremely challenging to hallucinate unseen regions that are occluded or partially observed in the given videos. To address these issues, we first finetune a pretrained RGB-D diffusion model on the video frames using a customization technique. Subsequently, we distill the knowledge from the finetuned model to a 4D representations encompassing both dynamic and static Neural Radiance Fields (NeRF) components. The proposed pipeline achieves geometric consistency while preserving the scene identity. We perform thorough experiments to evaluate the efficacy of the proposed method qualitatively and quantitatively. Our results demonstrate the robustness and utility of our approach in challenging cases, further advancing dynamic novel view synthesis.
Authors: Edison Guo, Maruf Ahmed, Yue Sun, Rahul Mahendru, Rui Yang, Harrison Cook, Tennessee Leeuwenburg, Ben Evans
Recently, the FourCastNet Neural Earth System Model (NESM) has shown impressive results on predicting various atmospheric variables, trained on the ERA5 reanalysis dataset. While FourCastNet enjoys quasi-linear time and memory complexity in sequence length compared to quadratic complexity in vanilla transformers, training FourCastNet on ERA5 from scratch still requires large amount of compute resources, which is expensive or even inaccessible to most researchers. In this work, we will show improved methods that can train FourCastNet using only 1% of the compute required by the baseline, while maintaining model performance or par or even better than the baseline.
Authors: Adrian Xuan Wei Lim, Lynnette Hui Xian Ng, Conor Griffin, Nicholas Kyger, Faraz Baghernezhad
We present Reverse Projection, a novel projective texture mapping technique for painting a decal directly to the texture of a 3D object. Designed to be used in games, this technique works in real-time. By using projection techniques that are computed in local space textures and outward-looking, users using low-end android devices to high-end gaming desktops are able to enjoy the personalization of their assets. We believe our proposed pipeline is a step in improving the speed and versatility of model painting.
Authors: Prakash Mallick, Feras Dayoub, Jamie Sherrah
This paper addresses the significant challenge in open-set object detection (OSOD): the tendency of state-of-the-art detectors to erroneously classify unknown objects as known categories with high confidence. We present a novel approach that effectively identifies unknown objects by distinguishing between high and low-density regions in latent space. Our method builds upon the Open-Det (OD) framework, introducing two new elements to the loss function. These elements enhance the known embedding space's clustering and expand the unknown space's low-density regions. The first addition is the Class Wasserstein Anchor (CWA), a new function that refines the classification boundaries. The second is a spectral normalisation step, improving the robustness of the model. Together, these augmentations to the existing Contrastive Feature Learner (CFL) and Unknown Probability Learner (UPL) loss functions significantly improve OSOD performance. Our proposed OpenDet-CWA (OD-CWA) method demonstrates: a) a reduction in open-set errors by approximately 17%-22%, b) an enhancement in novelty detection capability by 1.5%-16%, and c) a decrease in the wilderness index by 2%-20% across various open-set scenarios. These results represent a substantial advancement in the field, showcasing the potential of our approach in managing the complexities of open-set object detection.
Authors: Lucas W. Remedios, Shunxing Bao, Samuel W. Remedios, Ho Hin Lee, Leon Y. Cai, Thomas Li, Ruining Deng, Can Cui, Jia Li, Qi Liu, Ken S. Lau, Joseph T. Roland, Mary K. Washington, Lori A. Coburn, Keith T. Wilson, Yuankai Huo, Bennett A. Landman
Understanding the way cells communicate, co-locate, and interrelate is essential to understanding human physiology. Hematoxylin and eosin (H&E) staining is ubiquitously available both for clinical studies and research. The Colon Nucleus Identification and Classification (CoNIC) Challenge has recently innovated on robust artificial intelligence labeling of six cell types on H&E stains of the colon. However, this is a very small fraction of the number of potential cell classification types. Specifically, the CoNIC Challenge is unable to classify epithelial subtypes (progenitor, endocrine, goblet), lymphocyte subtypes (B, helper T, cytotoxic T), or connective subtypes (fibroblasts, stromal). In this paper, we propose to use inter-modality learning to label previously un-labelable cell types on virtual H&E. We leveraged multiplexed immunofluorescence (MxIF) histology imaging to identify 14 subclasses of cell types. We performed style transfer to synthesize virtual H&E from MxIF and transferred the higher density labels from MxIF to these virtual H&E images. We then evaluated the efficacy of learning in this approach. We identified helper T and progenitor nuclei with positive predictive values of $0.34 \pm 0.15$ (prevalence $0.03 \pm 0.01$) and $0.47 \pm 0.1$ (prevalence $0.07 \pm 0.02$) respectively on virtual H&E. This approach represents a promising step towards automating annotation in digital pathology.
Authors: Andrew Gritsevskiy, Arjun Panickssery, Aaron Kirtland, Derik Kauffman, Hans Gundlach, Irina Gritsevskaya, Joe Cavanagh, Jonathan Chiang, Lydia La Roux, Michelle Hung
We propose a new benchmark evaluating the performance of multimodal large language models on rebus puzzles. The dataset covers 333 original examples of image-based wordplay, cluing 13 categories such as movies, composers, major cities, and food. To achieve good performance on the benchmark of identifying the clued word or phrase, models must combine image recognition and string manipulation with hypothesis testing, multi-step reasoning, and an understanding of human cognition, making for a complex, multimodal evaluation of capabilities. We find that proprietary models such as GPT-4V and Gemini Pro significantly outperform all other tested models. However, even the best model has a final accuracy of just 24%, highlighting the need for substantial improvements in reasoning. Further, models rarely understand all parts of a puzzle, and are almost always incapable of retroactively explaining the correct answer. Our benchmark can therefore be used to identify major shortcomings in the knowledge and reasoning of multimodal large language models.
Authors: Juni Kim, Zhikang Dong, Pawel Polak
We introduce a novel method that combines differential geometry, kernels smoothing, and spectral analysis to quantify facial muscle activity from widely accessible video recordings, such as those captured on personal smartphones. Our approach emphasizes practicality and accessibility. It has significant potential for applications in national security and plastic surgery. Additionally, it offers remote diagnosis and monitoring for medical conditions such as stroke, Bell's palsy, and acoustic neuroma. Moreover, it is adept at detecting and classifying emotions, from the overt to the subtle. The proposed face muscle analysis technique is an explainable alternative to deep learning methods and a non-invasive substitute to facial electromyography (fEMG).
Authors: Gang Wu, Junjun Jiang, Junpeng Jiang, Xianming Liu
Recent progress in single-image super-resolution (SISR) has achieved remarkable performance, yet the computational costs of these methods remain a challenge for deployment on resource-constrained devices. Especially for transformer-based methods, the self-attention mechanism in such models brings great breakthroughs while incurring substantial computational costs. To tackle this issue, we introduce the Convolutional Transformer layer (ConvFormer) and the ConvFormer-based Super-Resolution network (CFSR), which offer an effective and efficient solution for lightweight image super-resolution tasks. In detail, CFSR leverages the large kernel convolution as the feature mixer to replace the self-attention module, efficiently modeling long-range dependencies and extensive receptive fields with a slight computational cost. Furthermore, we propose an edge-preserving feed-forward network, simplified as EFN, to obtain local feature aggregation and simultaneously preserve more high-frequency information. Extensive experiments demonstrate that CFSR can achieve an advanced trade-off between computational cost and performance when compared to existing lightweight SR methods. Compared to state-of-the-art methods, e.g. ShuffleMixer, the proposed CFSR achieves 0.39 dB gains on Urban100 dataset for x2 SR task while containing 26% and 31% fewer parameters and FLOPs, respectively. Code and pre-trained models are available at https://github.com/Aitical/CFSR.
Authors: Changtai Li, Xu Han, Chao Yao, Xiaojuan Ban
Accurate and efficient extraction of microstructures in microscopic images of materials plays a critical role in the exploration of structure-property relationships and the optimization of process parameters. Deep learning-based image segmentation techniques that rely on manual annotation are time-consuming and labor-intensive and hardly meet the demand for model transferability and generalization. Segment Anything Model (SAM), a large visual model with powerful deep feature representation and zero-shot generalization capabilities, has provided new solutions for image segmentation. However, directly applying SAM to segmenting microstructures in microscopic images of materials without human annotation cannot achieve the expected results, as the difficulty of adapting its native prompt engineering to the dense and dispersed characteristics of key microstructures in materials microscopy images. In this paper, we propose MatSAM, a general and efficient microstructure extraction solution based on SAM. A new point-based prompts generation strategy is designed, grounded on the distribution and shape of materials microstructures. It generates prompts for different microscopic images, fuses the prompts of the region of interest (ROI) key points and grid key points, and integrates post-processing methods for quantitative characterization of materials microstructures. For common microstructures including grain boundary and phase, MatSAM achieves superior segmentation performance to conventional methods and is even preferable to supervised learning methods evaluated on 18 materials microstructures imaged by the optical microscope (OM) and scanning electron microscope (SEM). We believe that MatSAM can significantly reduce the cost of quantitative characterization of materials microstructures and accelerate the design of new materials.
Authors: Chunlei Peng, Boyu Wang, Decheng Liu, Nannan Wang, Ruimin Hu, Xinbo Gao
Cloth-changing person re-identification (CC-ReID) aims to match persons who change clothes over long periods. The key challenge in CC-ReID is to extract clothing-independent features, such as face, hairstyle, body shape, and gait. Current research mainly focuses on modeling body shape using multi-modal biological features (such as silhouettes and sketches). However, it does not fully leverage the personal description information hidden in the original RGB image. Considering that there are certain attribute descriptions which remain unchanged after the changing of cloth, we propose a Masked Attribute Description Embedding (MADE) method that unifies personal visual appearance and attribute description for CC-ReID. Specifically, handling variable clothing-sensitive information, such as color and type, is challenging for effective modeling. To address this, we mask the clothing and color information in the personal attribute description extracted through an attribute detection model. The masked attribute description is then connected and embedded into Transformer blocks at various levels, fusing it with the low-level to high-level features of the image. This approach compels the model to discard clothing information. Experiments are conducted on several CC-ReID benchmarks, including PRCC, LTCC, Celeb-reID-light, and LaST. Results demonstrate that MADE effectively utilizes attribute description, enhancing cloth-changing person re-identification performance, and compares favorably with state-of-the-art methods. The code is available at https://github.com/moon-wh/MADE.
Authors: Seung Hyun Lee, Yinxiao Li, Junjie Ke, Innfarn Yoo, Han Zhang, Jiahui Yu, Qifei Wang, Fei Deng, Glenn Entis, Junfeng He, Gang Li, Sangpil Kim, Irfan Essa, Feng Yang
Recent works demonstrate that using reinforcement learning (RL) with quality rewards can enhance the quality of generated images in text-to-image (T2I) generation. However, a simple aggregation of multiple rewards may cause over-optimization in certain metrics and degradation in others, and it is challenging to manually find the optimal weights. An effective strategy to jointly optimize multiple rewards in RL for T2I generation is highly desirable. This paper introduces Parrot, a novel multi-reward RL framework for T2I generation. Through the use of the batch-wise Pareto optimal selection, Parrot automatically identifies the optimal trade-off among different rewards during the RL optimization of the T2I generation. Additionally, Parrot employs a joint optimization approach for the T2I model and the prompt expansion network, facilitating the generation of quality-aware text prompts, thus further enhancing the final image quality. To counteract the potential catastrophic forgetting of the original user prompt due to prompt expansion, we introduce original prompt centered guidance at inference time, ensuring that the generated image remains faithful to the user input. Extensive experiments and a user study demonstrate that Parrot outperforms several baseline methods across various quality criteria, including aesthetics, human preference, image sentiment, and text-image alignment.
Authors: Weibo Jiang, Weihong Ren, Jiandong Tian, Liangqiong Qu, Zhiyong Wang, Honghai Liu
Human-Object Interaction (HOI) detection plays a vital role in scene understanding, which aims to predict the HOI triplet in the form of <human, object, action>. Existing methods mainly extract multi-modal features (e.g., appearance, object semantics, human pose) and then fuse them together to directly predict HOI triplets. However, most of these methods focus on seeking for self-triplet aggregation, but ignore the potential cross-triplet dependencies, resulting in ambiguity of action prediction. In this work, we propose to explore Self- and Cross-Triplet Correlations (SCTC) for HOI detection. Specifically, we regard each triplet proposal as a graph where Human, Object represent nodes and Action indicates edge, to aggregate self-triplet correlation. Also, we try to explore cross-triplet dependencies by jointly considering instance-level, semantic-level, and layout-level relations. Besides, we leverage the CLIP model to assist our SCTC obtain interaction-aware feature by knowledge distillation, which provides useful action clues for HOI detection. Extensive experiments on HICO-DET and V-COCO datasets verify the effectiveness of our proposed SCTC.
Authors: Blaise Appolinary, Alex Deaconu, Sophia Yang
In this paper, we present a novel method for dynamically expanding Convolutional Neural Networks (CNNs) during training, aimed at meeting the increasing demand for efficient and sustainable deep learning models. Our approach, drawing from the seminal work on Self-Expanding Neural Networks (SENN), employs a natural expansion score as an expansion criteria to address the common issue of over-parameterization in deep convolutional neural networks, thereby ensuring that the model's complexity is finely tuned to the task's specific needs. A significant benefit of this method is its eco-friendly nature, as it obviates the necessity of training multiple models of different sizes. We employ a strategy where a single model is dynamically expanded, facilitating the extraction of checkpoints at various complexity levels, effectively reducing computational resource use and energy consumption while also expediting the development cycle by offering diverse model complexities from a single training session. We evaluate our method on the CIFAR-10 dataset and our experimental results validate this approach, demonstrating that dynamically adding layers not only maintains but also improves CNN performance, underscoring the effectiveness of our expansion criteria. This approach marks a considerable advancement in developing adaptive, scalable, and environmentally considerate neural network architectures, addressing key challenges in the field of deep learning.
Authors: Licai Sun, Zheng Lian, Bin Liu, Jianhua Tao
Audio-Visual Emotion Recognition (AVER) has garnered increasing attention in recent years for its critical role in creating emotion-ware intelligent machines. Previous efforts in this area are dominated by the supervised learning paradigm. Despite significant progress, supervised learning is meeting its bottleneck due to the longstanding data scarcity issue in AVER. Motivated by recent advances in self-supervised learning, we propose Hierarchical Contrastive Masked Autoencoder (HiCMAE), a novel self-supervised framework that leverages large-scale self-supervised pre-training on vast unlabeled audio-visual data to promote the advancement of AVER. Following prior arts in self-supervised audio-visual representation learning, HiCMAE adopts two primary forms of self-supervision for pre-training, namely masked data modeling and contrastive learning. Unlike them which focus exclusively on top-layer representations while neglecting explicit guidance of intermediate layers, HiCMAE develops a three-pronged strategy to foster hierarchical audio-visual feature learning and improve the overall quality of learned representations. To verify the effectiveness of HiCMAE, we conduct extensive experiments on 9 datasets covering both categorical and dimensional AVER tasks. Experimental results show that our method significantly outperforms state-of-the-art supervised and self-supervised audio-visual methods, which indicates that HiCMAE is a powerful audio-visual emotion representation learner. Codes and models will be publicly available at https://github.com/sunlicai/HiCMAE.
Authors: Hui Lv, Qianru Sun
Video Anomaly Detection (VAD) aims to localize abnormal events on the timeline of long-range surveillance videos. Anomaly-scoring-based methods have been prevailing for years but suffer from the high complexity of thresholding and low explanability of detection results. In this paper, we conduct pioneer research on equipping video-based large language models (VLLMs) in the framework of VAD, making the VAD model free from thresholds and able to explain the reasons for the detected anomalies. We introduce a novel network module Long-Term Context (LTC) to mitigate the incapability of VLLMs in long-range context modeling. We design a three-phase training method to improve the efficiency of fine-tuning VLLMs by substantially minimizing the requirements for VAD data and lowering the costs of annotating instruction-tuning data. Our trained model achieves the top performance on the anomaly videos of the UCF-Crime and TAD benchmarks, with the AUC improvements of +3.86\% and +4.96\%, respectively. More impressively, our approach can provide textual explanations for detected anomalies.
Authors: Jaeill Kim, Duhun Hwang, Eunjung Lee, Jangwon Suh, Jimyeong Kim, Wonjong Rhee
In the past few years, contrastive learning has played a central role for the success of visual unsupervised representation learning. Around the same time, high-performance non-contrastive learning methods have been developed as well. While most of the works utilize only two views, we carefully review the existing multi-view methods and propose a general multi-view strategy that can improve learning speed and performance of any contrastive or non-contrastive method. We first analyze CMC's full-graph paradigm and empirically show that the learning speed of $K$-views can be increased by $_{K}\mathrm{C}_{2}$ times for small learning rate and early training. Then, we upgrade CMC's full-graph by mixing views created by a crop-only augmentation, adopting small-size views as in SwAV multi-crop, and modifying the negative sampling. The resulting multi-view strategy is called ECPP (Efficient Combinatorial Positive Pairing). We investigate the effectiveness of ECPP by applying it to SimCLR and assessing the linear evaluation performance for CIFAR-10 and ImageNet-100. For each benchmark, we achieve a state-of-the-art performance. In case of ImageNet-100, ECPP boosted SimCLR outperforms supervised learning.
Authors: Kumara Kahatapitiya, Adil Karjauv, Davide Abati, Fatih Porikli, Yuki M. Asano, Amirhossein Habibian
Diffusion-based video editing have reached impressive quality and can transform either the global style, local structure, and attributes of given video inputs, following textual edit prompts. However, such solutions typically incur heavy memory and computational costs to generate temporally-coherent frames, either in the form of diffusion inversion and/or cross-frame attention. In this paper, we conduct an analysis of such inefficiencies, and suggest simple yet effective modifications that allow significant speed-ups whilst maintaining quality. Moreover, we introduce Object-Centric Diffusion, coined as OCD, to further reduce latency by allocating computations more towards foreground edited regions that are arguably more important for perceptual quality. We achieve this by two novel proposals: i) Object-Centric Sampling, decoupling the diffusion steps spent on salient regions or background, allocating most of the model capacity to the former, and ii) Object-Centric 3D Token Merging, which reduces cost of cross-frame attention by fusing redundant tokens in unimportant background regions. Both techniques are readily applicable to a given video editing model \textit{without} retraining, and can drastically reduce its memory and computational cost. We evaluate our proposals on inversion-based and control-signal-based editing pipelines, and show a latency reduction up to 10x for a comparable synthesis quality.
Authors: Chenghao Li, Boheng Zeng, Yi Lu, Pengbo Shi, Qingzi Chen, Jirui Liu, Lingyun Zhu
We revisit the relationship between attention mechanisms and large kernel ConvNets in visual transformers and propose a new spatial attention named Large Kernel Convolutional Attention (LKCA). It simplifies the attention operation by replacing it with a single large kernel convolution. LKCA combines the advantages of convolutional neural networks and visual transformers, possessing a large receptive field, locality, and parameter sharing. We explained the superiority of LKCA from both convolution and attention perspectives, providing equivalent code implementations for each view. Experiments confirm that LKCA implemented from both the convolutional and attention perspectives exhibit equivalent performance. We extensively experimented with the LKCA variant of ViT in both classification and segmentation tasks. The experiments demonstrated that LKCA exhibits competitive performance in visual tasks. Our code will be made publicly available at https://github.com/CatworldLee/LKCA.
Authors: Barry Shichen Hu, Siyun Liang, Johannes Paetzold, Huy H. Nguyen, Isao Echizen, Jiapeng Tang
We propose the use of a Transformer to accurately predict normals from point clouds with noise and density variations. Previous learning-based methods utilize PointNet variants to explicitly extract multi-scale features at different input scales, then focus on a surface fitting method by which local point cloud neighborhoods are fitted to a geometric surface approximated by either a polynomial function or a multi-layer perceptron (MLP). However, fitting surfaces to fixed-order polynomial functions can suffer from overfitting or underfitting, and learning MLP-represented hyper-surfaces requires pre-generated per-point weights. To avoid these limitations, we first unify the design choices in previous works and then propose a simplified Transformer-based model to extract richer and more robust geometric features for the surface normal estimation task. Through extensive experiments, we demonstrate that our Transformer-based method achieves state-of-the-art performance on both the synthetic shape dataset PCPNet, and the real-world indoor scene dataset SceneNN, exhibiting more noise-resilient behavior and significantly faster inference. Most importantly, we demonstrate that the sophisticated hand-designed modules in existing works are not necessary to excel at the task of surface normal estimation.
Authors: Peng Dai, Feitong Tan, Xin Yu, Yinda Zhang, Xiaojuan Qi
Despite advances in 3D generation, the direct creation of 3D objects within an existing 3D scene represented as NeRF remains underexplored. This process requires not only high-quality 3D object generation but also seamless composition of the generated 3D content into the existing NeRF. To this end, we propose a new method, GO-NeRF, capable of utilizing scene context for high-quality and harmonious 3D object generation within an existing NeRF. Our method employs a compositional rendering formulation that allows the generated 3D objects to be seamlessly composited into the scene utilizing learned 3D-aware opacity maps without introducing unintended scene modification. Moreover, we also develop tailored optimization objectives and training strategies to enhance the model's ability to exploit scene context and mitigate artifacts, such as floaters, originating from 3D object generation within a scene. Extensive experiments on both feed-forward and $360^o$ scenes show the superior performance of our proposed GO-NeRF in generating objects harmoniously composited with surrounding scenes and synthesizing high-quality novel view images. Project page at {\url{https://daipengwa.github.io/GO-NeRF/}.
Authors: Na Wang, Lei Qi, Jintao Guo, Yinghuan Shi, Yang Gao
Domain generalization (DG) intends to train a model on multiple source domains to ensure that it can generalize well to an arbitrary unseen target domain. The acquisition of domain-invariant representations is pivotal for DG as they possess the ability to capture the inherent semantic information of the data, mitigate the influence of domain shift, and enhance the generalization capability of the model. Adopting multiple perspectives, such as the sample and the feature, proves to be effective. The sample perspective facilitates data augmentation through data manipulation techniques, whereas the feature perspective enables the extraction of meaningful generalization features. In this paper, we focus on improving the generalization ability of the model by compelling it to acquire domain-invariant representations from both the sample and feature perspectives by disentangling spurious correlations and enhancing potential correlations. 1) From the sample perspective, we develop a frequency restriction module, guiding the model to focus on the relevant correlations between object features and labels, thereby disentangling spurious correlations. 2) From the feature perspective, the simple Tail Interaction module implicitly enhances potential correlations among all samples from all source domains, facilitating the acquisition of domain-invariant representations across multiple domains for the model. The experimental results show that Convolutional Neural Networks (CNNs) or Multi-Layer Perceptrons (MLPs) with a strong baseline embedded with these two modules can achieve superior results, e.g., an average accuracy of 92.30% on Digits-DG.
Authors: Adrian Gheorghiu, Iulian-Marius Tăiatu, Dumitru-Clementin Cercel, Iuliana Marin, Florin Pop
The detection and classification of diseases in Robusta coffee leaves are essential to ensure that plants are healthy and the crop yield is kept high. However, this job requires extensive botanical knowledge and much wasted time. Therefore, this task and others similar to it have been extensively researched subjects in image classification. Regarding leaf disease classification, most approaches have used the more popular PlantVillage dataset while completely disregarding other datasets, like the Robusta Coffee Leaf (RoCoLe) dataset. As the RoCoLe dataset is imbalanced and does not have many samples, fine-tuning of pre-trained models and multiple augmentation techniques need to be used. The current paper uses the RoCoLe dataset and approaches based on deep learning for classifying coffee leaf diseases from images, incorporating the pix2pix model for segmentation and cycle-generative adversarial network (CycleGAN) for augmentation. Our study demonstrates the effectiveness of Transformer-based models, online augmentations, and CycleGAN augmentation in improving leaf disease classification. While synthetic data has limitations, it complements real data, enhancing model performance. These findings contribute to developing robust techniques for plant disease detection and classification.
Authors: Kunpeng Qiu, Zhiying Zhou, Yongxin Guo
Accurate lesion classification in Wireless Capsule Endoscopy (WCE) images is vital for early diagnosis and treatment of gastrointestinal (GI) cancers. However, this task is confronted with challenges like tiny lesions and background interference. Additionally, WCE images exhibit higher intra-class variance and inter-class similarities, adding complexity. To tackle these challenges, we propose Decoupled Supervised Contrastive Learning for WCE image classification, learning robust representations from zoomed-in WCE images generated by Saliency Augmentor. Specifically, We use uniformly down-sampled WCE images as anchors and WCE images from the same class, especially their zoomed-in images, as positives. This approach empowers the Feature Extractor to capture rich representations from various views of the same image, facilitated by Decoupled Supervised Contrastive Learning. Training a linear Classifier on these representations within 10 epochs yields an impressive 92.01% overall accuracy, surpassing the prior state-of-the-art (SOTA) by 0.72% on a blend of two publicly accessible WCE datasets. Code is available at: https://github.com/Qiukunpeng/DSCL.
Authors: Wujie Sun, Defang Chen, Jiawei Chen, Yan Feng, Chun Chen, Can Wang
Deep learning has witnessed significant advancements in recent years at the cost of increasing training, inference, and model storage overhead. While existing model compression methods strive to reduce the number of model parameters while maintaining high accuracy, they inevitably necessitate the re-training of the compressed model or impose architectural constraints. To overcome these limitations, this paper presents a novel framework, termed \textbf{K}nowledge \textbf{T}ranslation (KT), wherein a ``translation'' model is trained to receive the parameters of a larger model and generate compressed parameters. The concept of KT draws inspiration from language translation, which effectively employs neural networks to convert different languages, maintaining identical meaning. Accordingly, we explore the potential of neural networks to convert models of disparate sizes, while preserving their functionality. We propose a comprehensive framework for KT, introduce data augmentation strategies to enhance model performance despite restricted training data, and successfully demonstrate the feasibility of KT on the MNIST dataset. Code is available at \url{https://github.com/zju-SWJ/KT}.
Authors: Jing Wu, Trung Le, Munawar Hayat, Mehrtash Harandi
In response to data protection regulations and the ``right to be forgotten'', in this work, we introduce an unlearning algorithm for diffusion models. Our algorithm equips a diffusion model with a mechanism to mitigate the concerns related to data memorization. To achieve this, we formulate the unlearning problem as a bi-level optimization problem, wherein the outer objective is to preserve the utility of the diffusion model on the remaining data. The inner objective aims to scrub the information associated with forgetting data by deviating the learnable generative process from the ground-truth denoising procedure. To solve the resulting bi-level problem, we adopt a first-order method, having superior practical performance while being vigilant about the diffusion process and solving a bi-level problem therein. Empirically, we demonstrate that our algorithm can preserve the model utility, effectiveness, and efficiency while removing across two widely-used diffusion models and in both conditional and unconditional image generation scenarios. In our experiments, we demonstrate the unlearning of classes, attributes, and even a race from face and object datasets such as UTKFace, CelebA, CelebA-HQ, and CIFAR10.
Authors: Xiaoyan Yu, Neng Dong, Liehuang Zhu, Hao Peng, Dapeng Tao
Visible-infrared person re-identification (VIReID) primarily deals with matching identities across person images from different modalities. Due to the modality gap between visible and infrared images, cross-modality identity matching poses significant challenges. Recognizing that high-level semantics of pedestrian appearance, such as gender, shape, and clothing style, remain consistent across modalities, this paper intends to bridge the modality gap by infusing visual features with high-level semantics. Given the capability of CLIP to sense high-level semantic information corresponding to visual representations, we explore the application of CLIP within the domain of VIReID. Consequently, we propose a CLIP-Driven Semantic Discovery Network (CSDN) that consists of Modality-specific Prompt Learner, Semantic Information Integration (SII), and High-level Semantic Embedding (HSE). Specifically, considering the diversity stemming from modality discrepancies in language descriptions, we devise bimodal learnable text tokens to capture modality-private semantic information for visible and infrared images, respectively. Additionally, acknowledging the complementary nature of semantic details across different modalities, we integrate text features from the bimodal language descriptions to achieve comprehensive semantics. Finally, we establish a connection between the integrated text features and the visual features across modalities. This process embed rich high-level semantic information into visual representations, thereby promoting the modality invariance of visual representations. The effectiveness and superiority of our proposed CSDN over existing methods have been substantiated through experimental evaluations on multiple widely used benchmarks. The code will be released at \url{https://github.com/nengdong96/CSDN}.
Authors: Alejandro Cobo, Roberto Valle, José M. Buenaposada, Luis Baumela
Head pose estimation (HPE) is a problem of interest in computer vision to improve the performance of face processing tasks in semi-frontal or profile settings. Recent applications require the analysis of faces in the full 360{\deg} rotation range. Traditional approaches to solve the semi-frontal and profile cases are not directly amenable for the full rotation case. In this paper we analyze the methodology for short- and wide-range HPE and discuss which representations and metrics are adequate for each case. We show that the popular Euler angles representation is a good choice for short-range HPE, but not at extreme rotations. However, the Euler angles' gimbal lock problem prevents them from being used as a valid metric in any setting. We also revisit the current cross-data set evaluation methodology and note that the lack of alignment between the reference systems of the training and test data sets negatively biases the results of all articles in the literature. We introduce a procedure to quantify this misalignment and a new methodology for cross-data set HPE that establishes new, more accurate, SOTA for the 300W-LP|Biwi benchmark. We also propose a generalization of the geodesic angular distance metric that enables the construction of a loss that controls the contribution of each training sample to the optimization of the model. Finally, we introduce a wide range HPE benchmark based on the CMU Panoptic data set.
Authors: Yannick Emonds, Kai Xi, Holger Fröning
Resistive memory is a promising alternative to SRAM, but is also an inherently unstable device that requires substantial effort to ensure correct read and write operations. To avoid the associated costs in terms of area, time and energy, the present work is concerned with exploring how much noise in memory operations can be tolerated by image classification tasks based on neural networks. We introduce a special noisy operator that mimics the noise in an exemplary resistive memory unit, explore the resilience of convolutional neural networks on the CIFAR-10 classification task, and discuss a couple of countermeasures to improve this resilience.
Authors: Jinge Wu, Yunsoo Kim, Honghan Wu
The recent success of large language and vision models on vision question answering (VQA), particularly their applications in medicine (Med-VQA), has shown a great potential of realizing effective visual assistants for healthcare. However, these models are not extensively tested on the hallucination phenomenon in clinical settings. Here, we created a hallucination benchmark of medical images paired with question-answer sets and conducted a comprehensive evaluation of the state-of-the-art models. The study provides an in-depth analysis of current models limitations and reveals the effectiveness of various prompting strategies.
Authors: Hanzhang Wang, Haoran Wang, Jinze Yang, Zhongrui Yu, Zeke Xie, Lei Tian, Xinyan Xiao, Junjun Jiang, Xianming Liu, Mingming Sun
The goal of Arbitrary Style Transfer (AST) is injecting the artistic features of a style reference into a given image/video. Existing methods usually focus on pursuing the balance between style and content, whereas ignoring the significant demand for flexible and customized stylization results and thereby limiting their practical application. To address this critical issue, a novel AST approach namely HiCAST is proposed, which is capable of explicitly customizing the stylization results according to various source of semantic clues. In the specific, our model is constructed based on Latent Diffusion Model (LDM) and elaborately designed to absorb content and style instance as conditions of LDM. It is characterized by introducing of \textit{Style Adapter}, which allows user to flexibly manipulate the output results by aligning multi-level style information and intrinsic knowledge in LDM. Lastly, we further extend our model to perform video AST. A novel learning objective is leveraged for video diffusion model training, which significantly improve cross-frame temporal consistency in the premise of maintaining stylization strength. Qualitative and quantitative comparisons as well as comprehensive user studies demonstrate that our HiCAST outperforms the existing SoTA methods in generating visually plausible stylization results.
Authors: Yu Jing, Tan Yujuan, Ren Ao, Liu Duo
Occlusions pose a significant challenge to optical flow algorithms that even rely on global evidences. We consider an occluded point to be one that is imaged in the reference frame but not in the next. Estimating the motion of these points is extremely difficult, particularly in the two-frame setting. Previous work only used the current frame as the only input, which could not guarantee providing correct global reference information for occluded points, and had problems such as long calculation time and poor accuracy in predicting optical flow at occluded points. To enable both high accuracy and efficiency, We fully mine and utilize the spatiotemporal information provided by the frame pair, design a loopback judgment algorithm to ensure that correct global reference information is obtained, mine multiple necessary global information, and design an efficient refinement module that fuses these global information. Specifically, we propose a YOIO framework, which consists of three main components: an initial flow estimator, a multiple global information extraction module, and a unified refinement module. We demonstrate that optical flow estimates in the occluded regions can be significantly improved in only one iteration without damaging the performance in non-occluded regions. Compared with GMA, the optical flow prediction accuracy of this method in the occluded area is improved by more than 10%, and the occ_out area exceeds 15%, while the calculation time is 27% shorter. This approach, running up to 18.9fps with 436*1024 image resolution, obtains new state-of-the-art results on the challenging Sintel dataset among all published and unpublished approaches that can run in real-time, suggesting a new paradigm for accurate and efficient optical flow estimation.
Authors: Ion Ciobotari, Adriana Príncipe, Maria Alexandra Oliveira, João Nuno Silva
The collection of ecological data in the field is essential to diagnose, monitor and manage ecosystems in a sustainable way. Since acquisition of this information through traditional methods are generally time-consuming, due to the capability of recording large volumes of data in short time periods, automation of data acquisition sees a growing trend. Terrestrial laser scanners (TLS), particularly LiDAR sensors, have been used in ecology, allowing to reconstruct the 3D structure of vegetation, and thus, infer ecosystem characteristics based on the spatial variation of the density of points. However, the low amount of information obtained per beam, lack of data analysis tools and the high cost of the equipment limit their use. This way, a low-cost TLS (<10k$) was developed along with data acquisition and processing mechanisms applicable in two case studies: an urban garden and a target area for ecological restoration. The orientation of LiDAR was modified to make observations in the vertical plane and a motor was integrated for its rotation, enabling the acquisition of 360 degree data with high resolution. Motion and location sensors were also integrated for automatic error correction and georeferencing. From the data generated, histograms of point density variation along the vegetation height were created, where shrub stratum was easily distinguishable from tree stratum, and maximum tree height and shrub cover were calculated. These results agreed with the field data, whereby the developed TLS has proved to be effective in calculating metrics of structural complexity of vegetation.
Authors: David Rivas-Villar, Álvaro S. Hervella, José Rouco, Jorge Novo
Retinal image registration is of utmost importance due to its wide applications in medical practice. In this context, we propose ConKeD, a novel deep learning approach to learn descriptors for retinal image registration. In contrast to current registration methods, our approach employs a novel multi-positive multi-negative contrastive learning strategy that enables the utilization of additional information from the available training samples. This makes it possible to learn high quality descriptors from limited training data. To train and evaluate ConKeD, we combine these descriptors with domain-specific keypoints, particularly blood vessel bifurcations and crossovers, that are detected using a deep neural network. Our experimental results demonstrate the benefits of the novel multi-positive multi-negative strategy, as it outperforms the widely used triplet loss technique (single-positive and single-negative) as well as the single-positive multi-negative alternative. Additionally, the combination of ConKeD with the domain-specific keypoints produces comparable results to the state-of-the-art methods for retinal image registration, while offering important advantages such as avoiding pre-processing, utilizing fewer training samples, and requiring fewer detected keypoints, among others. Therefore, ConKeD shows a promising potential towards facilitating the development and application of deep learning-based methods for retinal image registration.
Authors: Hyunjin Kim, Minhyuk Sung
We introduce PartSTAD, a method designed for the task adaptation of 2D-to-3D segmentation lifting. Recent studies have highlighted the advantages of utilizing 2D segmentation models to achieve high-quality 3D segmentation through few-shot adaptation. However, previous approaches have focused on adapting 2D segmentation models for domain shift to rendered images and synthetic text descriptions, rather than optimizing the model specifically for 3D segmentation. Our proposed task adaptation method finetunes a 2D bounding box prediction model with an objective function for 3D segmentation. We introduce weights for 2D bounding boxes for adaptive merging and learn the weights using a small additional neural network. Additionally, we incorporate SAM, a foreground segmentation model on a bounding box, to improve the boundaries of 2D segments and consequently those of 3D segmentation. Our experiments on the PartNet-Mobility dataset show significant improvements with our task adaptation approach, achieving a 7.0%p increase in mIoU and a 5.2%p improvement in mAP_50 for semantic and instance segmentation compared to the SotA few-shot 3D segmentation model.
Authors: Kang Chen, Yuanjie Liu
This article introduces a sliding window model for defocus deblurring that achieves the best performance to date with extremely low memory usage. Named Swintormer, the method utilizes a diffusion model to generate latent prior features that assist in restoring more detailed images. It also extends the sliding window strategy to specialized Transformer blocks for efficient inference. Additionally, we have further optimized Multiply-Accumulate operations (Macs). Compared to the currently top-performing GRL method, our Swintormer model drastically reduces computational complexity from 140.35 GMACs to 8.02 GMacs, while also improving the Signal-to-Noise Ratio (SNR) for defocus deblurring from 27.04 dB to 27.07 dB. This new method allows for the processing of higher resolution images on devices with limited memory, significantly expanding potential application scenarios. The article concludes with an ablation study that provides an in-depth analysis of the impact of each network module on final performance. The source code and model will be available at the following website: https://github.com/bnm6900030/swintormer.
Authors: Bin Dou, Tianyu Zhang, Yongjia Ma, Zhaohui Wang, Zejian Yuan
We propose Compact and Swift Segmenting 3D Gaussians(CoSSegGaussians), a method for compact 3D-consistent scene segmentation at fast rendering speed with only RGB images input. Previous NeRF-based 3D segmentation methods have relied on implicit or voxel neural scene representation and ray-marching volume rendering which are time consuming. Recent 3D Gaussian Splatting significantly improves the rendering speed, however, existing Gaussians-based segmentation methods(eg: Gaussian Grouping) fail to provide compact segmentation masks especially in zero-shot segmentation, which is mainly caused by the lack of robustness and compactness for straightforwardly assigning learnable parameters to each Gaussian when encountering inconsistent 2D machine-generated labels. Our method aims to achieve compact and reliable zero-shot scene segmentation swiftly by mapping fused spatial and semantically meaningful features for each Gaussian point with a shallow decoding network. Specifically, our method firstly optimizes Gaussian points' position, convariance and color attributes under the supervision of RGB images. After Gaussian Locating, we distill multi-scale DINO features extracted from images through unprojection to each Gaussian, which is then incorporated with spatial features from the fast point features processing network, i.e. RandLA-Net. Then the shallow decoding MLP is applied to the multi-scale fused features to obtain compact segmentation. Experimental results show that our model can perform high-quality zero-shot scene segmentation, as our model outperforms other segmentation methods on both semantic and panoptic segmentation task, meanwhile consumes approximately only 10% segmenting time compared to NeRF-based segmentation. Code and more results will be available at https://David-Dou.github.io/CoSSegGaussians
Authors: Hongjun Zhang
Try to generate new bridge types using generative artificial intelligence technology. Using symmetric structured image dataset of three-span beam bridge, arch bridge, cable-stayed bridge and suspension bridge , based on Python programming language, TensorFlow and Keras deep learning platform framework , PixelCNN is constructed and trained. The model can capture the statistical structure of the images and calculate the probability distribution of the next pixel when the previous pixels are given. From the obtained latent space sampling, new bridge types different from the training dataset can be generated. PixelCNN can organically combine different structural components on the basis of human original bridge types, creating new bridge types that have a certain degree of human original ability. Autoregressive models cannot understand the meaning of the sequence, while multimodal models combine regression and autoregressive models to understand the sequence. Multimodal models should be the way to achieve artificial general intelligence in the future.
Authors: Yashwardhan Chaudhuri, Ankit Kumar, Orchid Chetia Phukan, Arun Balaji Buduru
Crowd counting finds direct applications in real-world situations, making computational efficiency and performance crucial. However, most of the previous methods rely on a heavy backbone and a complex downstream architecture that restricts the deployment. To address this challenge and enhance the versatility of crowd-counting models, we introduce two lightweight models. These models maintain the same downstream architecture while incorporating two distinct backbones: MobileNet and MobileViT. We leverage Adjacent Feature Fusion to extract diverse scale features from a Pre-Trained Model (PTM) and subsequently combine these features seamlessly. This approach empowers our models to achieve improved performance while maintaining a compact and efficient design. With the comparison of our proposed models with previously available state-of-the-art (SOTA) methods on ShanghaiTech-A ShanghaiTech-B and UCF-CC-50 dataset, it achieves comparable results while being the most computationally efficient model. Finally, we present a comparative study, an extensive ablation study, along with pruning to show the effectiveness of our models.
Authors: Rouwan Wu, Xiaoya Cheng, Juelin Zhu, Xuxiang Liu, Maojun Zhang, Shen Yan
Despite significant progress in global localization of Unmanned Aerial Vehicles (UAVs) in GPS-denied environments, existing methods remain constrained by the availability of datasets. Current datasets often focus on small-scale scenes and lack viewpoint variability, accurate ground truth (GT) pose, and UAV build-in sensor data. To address these limitations, we introduce a large-scale 6-DoF UAV dataset for localization (UAVD4L) and develop a two-stage 6-DoF localization pipeline (UAVLoc), which consists of offline synthetic data generation and online visual localization. Additionally, based on the 6-DoF estimator, we design a hierarchical system for tracking ground target in 3D space. Experimental results on the new dataset demonstrate the effectiveness of the proposed approach. Code and dataset are available at https://github.com/RingoWRW/UAVD4L
Authors: Qian Gong, Jieyang Chen, Ben Whitney, Xin Liang, Viktor Reshniak, Tania Banerjee, Jaemoon Lee, Anand Rangarajan, Lipeng Wan, Nicolas Vidal, Qing Liu, Ana Gainaru, Norbert Podhorszki, Richard Archibald, Sanjay Ranka, Scott Klasky
We describe MGARD, a software providing MultiGrid Adaptive Reduction for floating-point scientific data on structured and unstructured grids. With exceptional data compression capability and precise error control, MGARD addresses a wide range of requirements, including storage reduction, high-performance I/O, and in-situ data analysis. It features a unified application programming interface (API) that seamlessly operates across diverse computing architectures. MGARD has been optimized with highly-tuned GPU kernels and efficient memory and device management mechanisms, ensuring scalable and rapid operations.
Authors: Sizhen Bian, Mengxi Liu, Bo Zhou, Paul Lukowicz, Michele Magno
Due to the fact that roughly sixty percent of the human body is essentially composed of water, the human body is inherently a conductive object, being able to, firstly, form an inherent electric field from the body to the surroundings and secondly, deform the distribution of an existing electric field near the body. Body-area capacitive sensing, also called body-area electric field sensing, is becoming a promising alternative for wearable devices to accomplish certain tasks in human activity recognition and human-computer interaction. Over the last decade, researchers have explored plentiful novel sensing systems backed by the body-area electric field. On the other hand, despite the pervasive exploration of the body-area electric field, a comprehensive survey does not exist for an enlightening guideline. Moreover, the various hardware implementations, applied algorithms, and targeted applications result in a challenging task to achieve a systematic overview of the subject. This paper aims to fill in the gap by comprehensively summarizing the existing works on body-area capacitive sensing so that researchers can have a better view of the current exploration status. To this end, we first sorted the explorations into three domains according to the involved body forms: body-part electric field, whole-body electric field, and body-to-body electric field, and enumerated the state-of-art works in the domains with a detailed survey of the backed sensing tricks and targeted applications. We then summarized the three types of sensing frontends in circuit design, which is the most critical part in body-area capacitive sensing, and analyzed the data processing pipeline categorized into three kinds of approaches. Finally, we described the challenges and outlooks of body-area electric sensing.
Authors: Linus Franke, Darius Rückert, Laura Fink, Marc Stamminger
Point-based radiance field rendering has demonstrated impressive results for novel view synthesis, offering a compelling blend of rendering quality and computational efficiency. However, also latest approaches in this domain are not without their shortcomings. 3D Gaussian Splatting [Kerbl and Kopanas et al. 2023] struggles when tasked with rendering highly detailed scenes, due to blurring and cloudy artifacts. On the other hand, ADOP [R\"uckert et al. 2022] can accommodate crisper images, but the neural reconstruction network decreases performance, it grapples with temporal instability and it is unable to effectively address large gaps in the point cloud.
In this paper, we present TRIPS (Trilinear Point Splatting), an approach that combines ideas from both Gaussian Splatting and ADOP. The fundamental concept behind our novel technique involves rasterizing points into a screen-space image pyramid, with the selection of the pyramid layer determined by the projected point size. This approach allows rendering arbitrarily large points using a single trilinear write. A lightweight neural network is then used to reconstruct a hole-free image including detail beyond splat resolution. Importantly, our render pipeline is entirely differentiable, allowing for automatic optimization of both point sizes and positions.
Our evaluation demonstrate that TRIPS surpasses existing state-of-the-art methods in terms of rendering quality while maintaining a real-time frame rate of 60 frames per second on readily available hardware. This performance extends to challenging scenarios, such as scenes featuring intricate geometry, expansive landscapes, and auto-exposed footage.
Authors: Benjamin Peters, James J. DiCarlo, Todd Gureckis, Ralf Haefner, Leyla Isik, Joshua Tenenbaum, Talia Konkle, Thomas Naselaris, Kimberly Stachenfeld, Zenna Tavares, Doris Tsao, Ilker Yildirim, Nikolaus Kriegeskorte
Vision is widely understood as an inference problem. However, two contrasting conceptions of the inference process have each been influential in research on biological vision as well as the engineering of machine vision. The first emphasizes bottom-up signal flow, describing vision as a largely feedforward, discriminative inference process that filters and transforms the visual information to remove irrelevant variation and represent behaviorally relevant information in a format suitable for downstream functions of cognition and behavioral control. In this conception, vision is driven by the sensory data, and perception is direct because the processing proceeds from the data to the latent variables of interest. The notion of "inference" in this conception is that of the engineering literature on neural networks, where feedforward convolutional neural networks processing images are said to perform inference. The alternative conception is that of vision as an inference process in Helmholtz's sense, where the sensory evidence is evaluated in the context of a generative model of the causal processes giving rise to it. In this conception, vision inverts a generative model through an interrogation of the evidence in a process often thought to involve top-down predictions of sensory data to evaluate the likelihood of alternative hypotheses. The authors include scientists rooted in roughly equal numbers in each of the conceptions and motivated to overcome what might be a false dichotomy between them and engage the other perspective in the realm of theory and experiment. The primate brain employs an unknown algorithm that may combine the advantages of both conceptions. We explain and clarify the terminology, review the key empirical evidence, and propose an empirical research program that transcends the dichotomy and sets the stage for revealing the mysterious hybrid algorithm of primate vision.
Authors: Martin S J Rogers, Maria Fox, Andrew Fleming, Louisa van Zeeland, Jeremy Wilkinson, J. Scott Hosking
Synthetic Aperture Radar (SAR) imagery is the primary data type used for sea ice mapping due to its spatio-temporal coverage and the ability to detect sea ice independent of cloud and lighting conditions. Automatic sea ice detection using SAR imagery remains problematic due to the presence of ambiguous signal and noise within the image. Conversely, ice and water are easily distinguishable using multispectral imagery (MSI), but in the polar regions the ocean's surface is often occluded by cloud or the sun may not appear above the horizon for many months. To address some of these limitations, this paper proposes a new tool trained using concurrent multispectral Visible and SAR imagery for sea Ice Detection (ViSual\_IceD). ViSual\_IceD is a convolution neural network (CNN) that builds on the classic U-Net architecture by containing two parallel encoder stages, enabling the fusion and concatenation of MSI and SAR imagery containing different spatial resolutions. The performance of ViSual\_IceD is compared with U-Net models trained using concatenated MSI and SAR imagery as well as models trained exclusively on MSI or SAR imagery. ViSual\_IceD outperforms the other networks, with a F1 score 1.60\% points higher than the next best network, and results indicate that ViSual\_IceD is selective in the image type it uses during image segmentation. Outputs from ViSual\_IceD are compared to sea ice concentration products derived from the AMSR2 Passive Microwave (PMW) sensor. Results highlight how ViSual\_IceD is a useful tool to use in conjunction with PMW data, particularly in coastal regions. As the spatial-temporal coverage of MSI and SAR imagery continues to increase, ViSual\_IceD provides a new opportunity for robust, accurate sea ice coverage detection in polar regions.
Authors: Rocío del Amor, Julio Silva-Rodríguez, Adrián Colomer, Valery Naranjo
The development of computer vision solutions for gigapixel images in digital pathology is hampered by significant computational limitations due to the large size of whole slide images. In particular, digitizing biopsies at high resolutions is a time-consuming process, which is necessary due to the worsening results from the decrease in image detail. To alleviate this issue, recent literature has proposed using knowledge distillation to enhance the model performance at reduced image resolutions. In particular, soft labels and features extracted at the highest magnification level are distilled into a model that takes lower-magnification images as input. However, this approach fails to transfer knowledge about the most discriminative image regions in the classification process, which may be lost when the resolution is decreased. In this work, we propose to distill this information by incorporating attention maps during training. In particular, our formulation leverages saliency maps of the target class via grad-CAMs, which guides the lower-resolution Student model to match the Teacher distribution by minimizing the l2 distance between them. Comprehensive experiments on prostate histology image grading demonstrate that the proposed approach substantially improves the model performance across different image resolutions compared to previous literature.
Authors: Cui Beilei, Islam Mobarakol, Bai Long, Ren Hongliang
Purpose: Depth estimation in robotic surgery is vital in 3D reconstruction, surgical navigation and augmented reality visualization. Although the foundation model exhibits outstanding performance in many vision tasks, including depth estimation (e.g., DINOv2), recent works observed its limitations in medical and surgical domain-specific applications. This work presents a low-ranked adaptation (LoRA) of the foundation model for surgical depth estimation. Methods: We design a foundation model-based depth estimation method, referred to as Surgical-DINO, a low-rank adaptation of the DINOv2 for depth estimation in endoscopic surgery. We build LoRA layers and integrate them into DINO to adapt with surgery-specific domain knowledge instead of conventional fine-tuning. During training, we freeze the DINO image encoder, which shows excellent visual representation capacity, and only optimize the LoRA layers and depth decoder to integrate features from the surgical scene. Results: Our model is extensively validated on a MICCAI challenge dataset of SCARED, which is collected from da Vinci Xi endoscope surgery. We empirically show that Surgical-DINO significantly outperforms all the state-of-the-art models in endoscopic depth estimation tasks. The analysis with ablation studies has shown evidence of the remarkable effect of our LoRA layers and adaptation. Conclusion: Surgical-DINO shed some light on the successful adaptation of the foundation models into the surgical domain for depth estimation. There is clear evidence in the results that zero-shot prediction on pre-trained weights in computer vision datasets or naive fine-tuning is not sufficient to use the foundation model in the surgical domain directly. Code is available at https://github.com/BeileiCui/SurgicalDINO.
Authors: Pablo Alonso, Jon Ander Iñiguez de Gordoa, Juan Diego Ortega, Sara García, Francisco Javier Iriarte, Marcos Nieto
Runway and taxiway pavements are exposed to high stress during their projected lifetime, which inevitably leads to a decrease in their condition over time. To make sure airport pavement condition ensure uninterrupted and resilient operations, it is of utmost importance to monitor their condition and conduct regular inspections. UAV-based inspection is recently gaining importance due to its wide range monitoring capabilities and reduced cost. In this work, we propose a vision-based approach to automatically identify pavement distress using images captured by UAVs. The proposed method is based on Deep Learning (DL) to segment defects in the image. The DL architecture leverages the low computational capacities of embedded systems in UAVs by using an optimised implementation of EfficientNet feature extraction and Feature Pyramid Network segmentation. To deal with the lack of annotated data for training we have developed a synthetic dataset generation methodology to extend available distress datasets. We demonstrate that the use of a mixed dataset composed of synthetic and real training images yields better results when testing the training models in real application scenarios.
Authors: Zhiyu Zhu, Huaming Chen, Xinyi Wang, Jiayu Zhang, Zhibo Jin, Kim-Kwang Raymond Choo
Adversarial generative models, such as Generative Adversarial Networks (GANs), are widely applied for generating various types of data, i.e., images, text, and audio. Accordingly, its promising performance has led to the GAN-based adversarial attack methods in the white-box and black-box attack scenarios. The importance of transferable black-box attacks lies in their ability to be effective across different models and settings, more closely aligning with real-world applications. However, it remains challenging to retain the performance in terms of transferable adversarial examples for such methods. Meanwhile, we observe that some enhanced gradient-based transferable adversarial attack algorithms require prolonged time for adversarial sample generation. Thus, in this work, we propose a novel algorithm named GE-AdvGAN to enhance the transferability of adversarial samples whilst improving the algorithm's efficiency. The main approach is via optimising the training process of the generator parameters. With the functional and characteristic similarity analysis, we introduce a novel gradient editing (GE) mechanism and verify its feasibility in generating transferable samples on various models. Moreover, by exploring the frequency domain information to determine the gradient editing direction, GE-AdvGAN can generate highly transferable adversarial samples while minimizing the execution time in comparison to the state-of-the-art transferable adversarial attack algorithms. The performance of GE-AdvGAN is comprehensively evaluated by large-scale experiments on different datasets, which results demonstrate the superiority of our algorithm. The code for our algorithm is available at: https://github.com/LMBTough/GE-advGAN
Authors: Partha Ghosh, Soubhik Sanyal, Cordelia Schmid, Bernhard Schölkopf
We present a novel unconditional video generative model designed to address long-term spatial and temporal dependencies. To capture these dependencies, our approach incorporates a hybrid explicit-implicit tri-plane representation inspired by 3D-aware generative frameworks developed for three-dimensional object representation and employs a singular latent code to model an entire video sequence. Individual video frames are then synthesized from an intermediate tri-plane representation, which itself is derived from the primary latent code. This novel strategy reduces computational complexity by a factor of $2$ as measured in FLOPs. Consequently, our approach facilitates the efficient and temporally coherent generation of videos. Moreover, our joint frame modeling approach, in contrast to autoregressive methods, mitigates the generation of visual artifacts. We further enhance the model's capabilities by integrating an optical flow-based module within our Generative Adversarial Network (GAN) based generator architecture, thereby compensating for the constraints imposed by a smaller generator size. As a result, our model is capable of synthesizing high-fidelity video clips at a resolution of $256\times256$ pixels, with durations extending to more than $5$ seconds at a frame rate of 30 fps. The efficacy and versatility of our approach are empirically validated through qualitative and quantitative assessments across three different datasets comprising both synthetic and real video clips.
Authors: Guanjun Wu, Taoran Yi, Jiemin Fang, Wenyu Liu, Xinggang Wang
Neural Radiances Fields (NeRF) and their extensions have shown great success in representing 3D scenes and synthesizing novel-view images. However, most NeRF methods take in low-dynamic-range (LDR) images, which may lose details, especially with nonuniform illumination. Some previous NeRF methods attempt to introduce high-dynamic-range (HDR) techniques but mainly target static scenes. To extend HDR NeRF methods to wider applications, we propose a dynamic HDR NeRF framework, named HDR-HexPlane, which can learn 3D scenes from dynamic 2D images captured with various exposures. A learnable exposure mapping function is constructed to obtain adaptive exposure values for each image. Based on the monotonically increasing prior, a camera response function is designed for stable learning. With the proposed model, high-quality novel-view images at any time point can be rendered with any desired exposure. We further construct a dataset containing multiple dynamic scenes captured with diverse exposures for evaluation. All the datasets and code are available at \url{https://guanjunwu.github.io/HDR-HexPlane/}.
Authors: Giuseppe Vecchio, Valentin Deschaintre
We introduce MatSynth, a dataset of $4,000+$ CC0 ultra-high resolution PBR materials. Materials are crucial components of virtual relightable assets, defining the interaction of light at the surface of geometries. Given their importance, significant research effort was dedicated to their representation, creation and acquisition. However, in the past 6 years, most research in material acquisiton or generation relied either on the same unique dataset, or on company-owned huge library of procedural materials. With this dataset we propose a significantly larger, more diverse, and higher resolution set of materials than previously publicly available. We carefully discuss the data collection process and demonstrate the benefits of this dataset on material acquisition and generation applications. The complete data further contains metadata with each material's origin, license, category, tags, creation method and, when available, descriptions and physical size, as well as 3M+ renderings of the augmented materials, in 1K, under various environment lightings. The MatSynth dataset is released through the project page at: https://www.gvecchio.com/matsynth.
Authors: Zhaowei Li, Qi Xu, Dong Zhang, Hang Song, Yiqing Cai, Qi Qi, Ran Zhou, Junting Pan, Zefeng Li, Van Tu Vu, Zhida Huang, Tao Wang
Multi-modal large language models have demonstrated impressive performance across various tasks in different modalities. However, existing multi-modal models primarily emphasize capturing global information within each modality while neglecting the importance of perceiving local information across modalities. Consequently, these models lack the ability to effectively understand the fine-grained details of input data, limiting their performance in tasks that require a more nuanced understanding. To address this limitation, there is a compelling need to develop models that enable fine-grained understanding across multiple modalities, thereby enhancing their applicability to a wide range of tasks. In this paper, we propose LEGO, a language enhanced multi-modal grounding model. Beyond capturing global information like other multi-modal models, our proposed model excels at tasks demanding a detailed understanding of local information within the input. It demonstrates precise identification and localization of specific regions in images or moments in videos. To achieve this objective, we design a diversified dataset construction pipeline, resulting in a multi-modal, multi-granularity dataset for model training. The code, dataset, and demo of our model can be found at https: //github.com/lzw-lzw/LEGO.
Authors: Moab Arar, Andrey Voynov, Amir Hertz, Omri Avrahami, Shlomi Fruchter, Yael Pritch, Daniel Cohen-Or, Ariel Shamir
Content creators often aim to create personalized images using personal subjects that go beyond the capabilities of conventional text-to-image models. Additionally, they may want the resulting image to encompass a specific location, style, ambiance, and more. Existing personalization methods may compromise personalization ability or the alignment to complex textual prompts. This trade-off can impede the fulfillment of user prompts and subject fidelity. We propose a new approach focusing on personalization methods for a \emph{single} prompt to address this issue. We term our approach prompt-aligned personalization. While this may seem restrictive, our method excels in improving text alignment, enabling the creation of images with complex and intricate prompts, which may pose a challenge for current techniques. In particular, our method keeps the personalized model aligned with a target prompt using an additional score distillation sampling term. We demonstrate the versatility of our method in multi- and single-shot settings and further show that it can compose multiple subjects or use inspiration from reference images, such as artworks. We compare our approach quantitatively and qualitatively with existing baselines and state-of-the-art techniques.
Authors: Luis Bolanos, Shih-Yang Su, Helge Rhodin
Neural character models can now reconstruct detailed geometry and texture from video, but they lack explicit shadows and shading, leading to artifacts when generating novel views and poses or during relighting. It is particularly difficult to include shadows as they are a global effect and the required casting of secondary rays is costly. We propose a new shadow model using a Gaussian density proxy that replaces sampling with a simple analytic formula. It supports dynamic motion and is tailored for shadow computation, thereby avoiding the affine projection approximation and sorting required by the closely related Gaussian splatting. Combined with a deferred neural rendering model, our Gaussian shadows enable Lambertian shading and shadow casting with minimal overhead. We demonstrate improved reconstructions, with better separation of albedo, shading, and shadows in challenging outdoor scenes with direct sun light and hard shadows. Our method is able to optimize the light direction without any input from the user. As a result, novel poses have fewer shadow artifacts and relighting in novel scenes is more realistic compared to the state-of-the-art methods, providing new ways to pose neural characters in novel environments, increasing their applicability.
Authors: Dilyara Bareeva, Marina M.-C. Höhne, Alexander Warnecke, Lukas Pirch, Klaus-Robert Müller, Konrad Rieck, Kirill Bykov
Deep Neural Networks (DNNs) are capable of learning complex and versatile representations, however, the semantic nature of the learned concepts remains unknown. A common method used to explain the concepts learned by DNNs is Activation Maximization (AM), which generates a synthetic input signal that maximally activates a particular neuron in the network. In this paper, we investigate the vulnerability of this approach to adversarial model manipulations and introduce a novel method for manipulating feature visualization without altering the model architecture or significantly impacting the model's decision-making process. We evaluate the effectiveness of our method on several neural network models and demonstrate its capabilities to hide the functionality of specific neurons by masking the original explanations of neurons with chosen target explanations during model auditing. As a remedy, we propose a protective measure against such manipulations and provide quantitative evidence which substantiates our findings.
Authors: Jack Saunders, Vinay Namboodiri
Visual dubbing is the process of generating lip motions of an actor in a video to synchronise with given audio. Recent advances have made progress towards this goal but have not been able to produce an approach suitable for mass adoption. Existing methods are split into either person-generic or person-specific models. Person-specific models produce results almost indistinguishable from reality but rely on long training times using large single-person datasets. Person-generic works have allowed for the visual dubbing of any video to any audio without further training, but these fail to capture the person-specific nuances and often suffer from visual artefacts. Our method, based on data-efficient neural rendering priors, overcomes the limitations of existing approaches. Our pipeline consists of learning a deferred neural rendering prior network and actor-specific adaptation using neural textures. This method allows for $\textbf{high-quality visual dubbing with just a few seconds of data}$, that enables video dubbing for any actor - from A-list celebrities to background actors. We show that we achieve state-of-the-art in terms of $\textbf{visual quality}$ and $\textbf{recognisability}$ both quantitatively, and qualitatively through two user studies. Our prior learning and adaptation method $\textbf{generalises to limited data}$ better and is more $\textbf{scalable}$ than existing person-specific models. Our experiments on real-world, limited data scenarios find that our model is preferred over all others. The project page may be found at https://dubbingforeveryone.github.io/
Authors: Yifan Gong, Zheng Zhan, Qing Jin, Yanyu Li, Yerlan Idelbayev, Xian Liu, Andrey Zharkov, Kfir Aberman, Sergey Tulyakov, Yanzhi Wang, Jian Ren
One highly promising direction for enabling flexible real-time on-device image editing is utilizing data distillation by leveraging large-scale text-to-image diffusion models, such as Stable Diffusion, to generate paired datasets used for training generative adversarial networks (GANs). This approach notably alleviates the stringent requirements typically imposed by high-end commercial GPUs for performing image editing with diffusion models. However, unlike text-to-image diffusion models, each distilled GAN is specialized for a specific image editing task, necessitating costly training efforts to obtain models for various concepts. In this work, we introduce and address a novel research direction: can the process of distilling GANs from diffusion models be made significantly more efficient? To achieve this goal, we propose a series of innovative techniques. First, we construct a base GAN model with generalized features, adaptable to different concepts through fine-tuning, eliminating the need for training from scratch. Second, we identify crucial layers within the base GAN model and employ Low-Rank Adaptation (LoRA) with a simple yet effective rank search process, rather than fine-tuning the entire base model. Third, we investigate the minimal amount of data necessary for fine-tuning, further reducing the overall training time. Extensive experiments show that we can efficiently empower GANs with the ability to perform real-time high-quality image editing on mobile devices with remarkable reduced training cost and storage for each concept.
Authors: Yue Zhao, Long Zhao, Xingyi Zhou, Jialin Wu, Chun-Te Chu, Hui Miao, Florian Schroff, Hartwig Adam, Ting Liu, Boqing Gong, Philipp Krähenbühl, Liangzhe Yuan
The recent advance in vision-language models is largely attributed to the abundance of image-text data. We aim to replicate this success for video-language models, but there simply is not enough human-curated video-text data available. We thus resort to fine-tuning a video-language model from a strong image-language baseline with synthesized instructional data. The resulting video-language model is then used to auto-label millions of videos to generate high-quality captions. We show the adapted video-language model performs well on a wide range of video-language benchmarks. For instance, it surpasses the best prior result on open-ended NExT-QA by 2.8%. Besides, our model generates detailed descriptions for previously unseen videos, which provide better textual supervision than existing methods. Experiments show that a video-language dual-encoder model contrastively trained on these auto-generated captions is 3.8% better than the strongest baseline that also leverages vision-language models. Our best model outperforms state-of-the-art methods on MSR-VTT zero-shot text-to-video retrieval by 6%.
Authors: Pedro R. A. S. Bassi, Sergio S. J. Dertkigil, Andrea Cavalli
Features in images' backgrounds can spuriously correlate with the images' classes, representing background bias. They can influence the classifier's decisions, causing shortcut learning (Clever Hans effect). The phenomenon generates deep neural networks (DNNs) that perform well on standard evaluation datasets but generalize poorly to real-world data. Layer-wise Relevance Propagation (LRP) explains DNNs' decisions. Here, we show that the optimization of LRP heatmaps can minimize the background bias influence on deep classifiers, hindering shortcut learning. By not increasing run-time computational cost, the approach is light and fast. Furthermore, it applies to virtually any classification architecture. After injecting synthetic bias in images' backgrounds, we compared our approach (dubbed ISNet) to eight state-of-the-art DNNs, quantitatively demonstrating its superior robustness to background bias. Mixed datasets are common for COVID-19 and tuberculosis classification with chest X-rays, fostering background bias. By focusing on the lungs, the ISNet reduced shortcut learning. Thus, its generalization performance on external (out-of-distribution) test databases significantly surpassed all implemented benchmark models.
Authors: Xianda Guo, Zheng Zhu, Tian Yang, Beibei Lin, Junjie Huang, Jiankang Deng, Guan Huang, Jie Zhou, Jiwen Lu
Gait benchmarks empower the research community to train and evaluate high-performance gait recognition systems. Even though growing efforts have been devoted to cross-view recognition, academia is restricted by current existing databases captured in the controlled environment. In this paper, we contribute a new benchmark and strong baseline for Gait REcognition in the Wild (GREW). The GREW dataset is constructed from natural videos, which contain hundreds of cameras and thousands of hours of streams in open systems. With tremendous manual annotations, the GREW consists of 26K identities and 128K sequences with rich attributes for unconstrained gait recognition. Moreover, we add a distractor set of over 233K sequences, making it more suitable for real-world applications. Compared with prevailing predefined cross-view datasets, the GREW has diverse and practical view variations, as well as more naturally challenging factors. To the best of our knowledge, this is the first large-scale dataset for gait recognition in the wild. Equipped with this benchmark, we dissect the unconstrained gait recognition problem, where representative appearance-based and model-based methods are explored. The proposed GREW benchmark proves to be essential for both training and evaluating gait recognizers in unconstrained scenarios. In addition, we propose the Single Path One-Shot neural architecture search with uniform sampling for Gait recognition, named SPOSGait, which is the first NAS-based gait recognition model. In experiments, SPOSGait achieves state-of-the-art performance on the CASIA-B, OU-MVLP, Gait3D, and GREW benchmarks, outperforming existing approaches by a large margin. The code will be released at https://github.com/XiandaGuo/SPOSGait.
Authors: Rima Alaifari, Giovanni S. Alberti, Tandri Gauksson
As interest in deep neural networks (DNNs) for image reconstruction tasks grows, their reliability has been called into question (Antun et al., 2020; Gottschling et al., 2020). However, recent work has shown that, compared to total variation (TV) minimization, when appropriately regularized, DNNs show similar robustness to adversarial noise in terms of $\ell^2$-reconstruction error (Genzel et al., 2022). We consider a different notion of robustness, using the $\ell^\infty$-norm, and argue that localized reconstruction artifacts are a more relevant defect than the $\ell^2$-error. We create adversarial perturbations to undersampled magnetic resonance imaging measurements (in the frequency domain) which induce severe localized artifacts in the TV-regularized reconstruction. Notably, the same attack method is not as effective against DNN based reconstruction. Finally, we show that this phenomenon is inherent to reconstruction methods for which exact recovery can be guaranteed, as with compressed sensing reconstructions with $\ell^1$- or TV-minimization.
Authors: Yuanchao Bai, Xianming Liu, Kai Wang, Xiangyang Ji, Xiaolin Wu, Wen Gao
Lossless and near-lossless image compression is of paramount importance to professional users in many technical fields, such as medicine, remote sensing, precision engineering and scientific research. But despite rapidly growing research interests in learning-based image compression, no published method offers both lossless and near-lossless modes. In this paper, we propose a unified and powerful deep lossy plus residual (DLPR) coding framework for both lossless and near-lossless image compression. In the lossless mode, the DLPR coding system first performs lossy compression and then lossless coding of residuals. We solve the joint lossy and residual compression problem in the approach of VAEs, and add autoregressive context modeling of the residuals to enhance lossless compression performance. In the near-lossless mode, we quantize the original residuals to satisfy a given $\ell_\infty$ error bound, and propose a scalable near-lossless compression scheme that works for variable $\ell_\infty$ bounds instead of training multiple networks. To expedite the DLPR coding, we increase the degree of algorithm parallelization by a novel design of coding context, and accelerate the entropy coding with adaptive residual interval. Experimental results demonstrate that the DLPR coding system achieves both the state-of-the-art lossless and near-lossless image compression performance with competitive coding speed.
Authors: An Tao, Yueqi Duan, Yingqi Wang, Jiwen Lu, Jie Zhou
In this paper, we investigate the dynamics-aware adversarial attack problem of adaptive neural networks. Most existing adversarial attack algorithms are designed under a basic assumption -- the network architecture is fixed throughout the attack process. However, this assumption does not hold for many recently proposed adaptive neural networks, which adaptively deactivate unnecessary execution units based on inputs to improve computational efficiency. It results in a serious issue of lagged gradient, making the learned attack at the current step ineffective due to the architecture change afterward. To address this issue, we propose a Leaded Gradient Method (LGM) and show the significant effects of the lagged gradient. More specifically, we reformulate the gradients to be aware of the potential dynamic changes of network architectures, so that the learned attack better "leads" the next step than the dynamics-unaware methods when network architecture changes dynamically. Extensive experiments on representative types of adaptive neural networks for both 2D images and 3D point clouds show that our LGM achieves impressive adversarial attack performance compared with the dynamic-unaware attack methods. Code is available at https://github.com/antao97/LGM.
Authors: Debaditya Roy, Ramanathan Rajendiran, Basura Fernando
Human-object interaction is one of the most important visual cues and we propose a novel way to represent human-object interactions for egocentric action anticipation. We propose a novel transformer variant to model interactions by computing the change in the appearance of objects and human hands due to the execution of the actions and use those changes to refine the video representation. Specifically, we model interactions between hands and objects using Spatial Cross-Attention (SCA) and further infuse contextual information using Trajectory Cross-Attention to obtain environment-refined interaction tokens. Using these tokens, we construct an interaction-centric video representation for action anticipation. We term our model InAViT which achieves state-of-the-art action anticipation performance on large-scale egocentric datasets EPICKTICHENS100 (EK100) and EGTEA Gaze+. InAViT outperforms other visual transformer-based methods including object-centric video representation. On the EK100 evaluation server, InAViT is the top-performing method on the public leaderboard (at the time of submission) where it outperforms the second-best model by 3.3% on mean-top5 recall.
Authors: Ronglai Zuo, Brian Mak
Most deep-learning-based continuous sign language recognition (CSLR) models share a similar backbone consisting of a visual module, a sequential module, and an alignment module. However, due to limited training samples, a connectionist temporal classification loss may not train such CSLR backbones sufficiently. In this work, we propose three auxiliary tasks to enhance the CSLR backbones. The first task enhances the visual module, which is sensitive to the insufficient training problem, from the perspective of consistency. Specifically, since the information of sign languages is mainly included in signers' facial expressions and hand movements, a keypoint-guided spatial attention module is developed to enforce the visual module to focus on informative regions, i.e., spatial attention consistency. Second, noticing that both the output features of the visual and sequential modules represent the same sentence, to better exploit the backbone's power, a sentence embedding consistency constraint is imposed between the visual and sequential modules to enhance the representation power of both features. We name the CSLR model trained with the above auxiliary tasks as consistency-enhanced CSLR, which performs well on signer-dependent datasets in which all signers appear during both training and testing. To make it more robust for the signer-independent setting, a signer removal module based on feature disentanglement is further proposed to remove signer information from the backbone. Extensive ablation studies are conducted to validate the effectiveness of these auxiliary tasks. More remarkably, with a transformer-based backbone, our model achieves state-of-the-art or competitive performance on five benchmarks, PHOENIX-2014, PHOENIX-2014-T, PHOENIX-2014-SI, CSL, and CSL-Daily. Code and Models are available at https://github.com/2000ZRL/LCSA_C2SLR_SRM.
Authors: Hao Feng, Keyi Zhou, Wengang Zhou, Yufei Yin, Jiajun Deng, Houqiang Li
Contour-based instance segmentation has been actively studied, thanks to its flexibility and elegance in processing visual objects within complex backgrounds. In this work, we propose a novel deep network architecture, i.e., PolySnake, for generic contour-based instance segmentation. Motivated by the classic Snake algorithm, the proposed PolySnake achieves superior and robust segmentation performance with an iterative and progressive contour refinement strategy. Technically, PolySnake introduces a recurrent update operator to estimate the object contour iteratively. It maintains a single estimate of the contour that is progressively deformed toward the object boundary. At each iteration, PolySnake builds a semantic-rich representation for the current contour and feeds it to the recurrent operator for further contour adjustment. Through the iterative refinements, the contour progressively converges to a stable status that tightly encloses the object instance. Beyond the scope of general instance segmentation, extensive experiments are conducted to validate the effectiveness and generalizability of our PolySnake in two additional specific task scenarios, including scene text detection and lane detection. The results demonstrate that the proposed PolySnake outperforms the existing advanced methods on several multiple prevalent benchmarks across the three tasks. The codes and pre-trained models are available at https://github.com/fh2019ustc/PolySnake
Authors: Deepika Bablani, Jeffrey L. Mckinstry, Steven K. Esser, Rathinakumar Appuswamy, Dharmendra S. Modha
For efficient neural network inference, it is desirable to achieve state-of-the-art accuracy with the simplest networks requiring the least computation, memory, and power. Quantizing networks to lower precision is a powerful technique for simplifying networks. As each layer of a network may have different sensitivity to quantization, mixed precision quantization methods selectively tune the precision of individual layers to achieve a minimum drop in task performance (e.g., accuracy). To estimate the impact of layer precision choice on task performance, two methods are introduced: i) Entropy Approximation Guided Layer selection (EAGL) is fast and uses the entropy of the weight distribution, and ii) Accuracy-aware Layer Precision Selection (ALPS) is straightforward and relies on single epoch fine-tuning after layer precision reduction. Using EAGL and ALPS for layer precision selection, full-precision accuracy is recovered with a mix of 4-bit and 2-bit layers for ResNet-50, ResNet-101 and BERT-base transformer networks, demonstrating enhanced performance across the entire accuracy-throughput frontier. The techniques demonstrate better performance than existing techniques in several commensurate comparisons. Notably, this is accomplished with significantly lesser computational time required to reach a solution.
Authors: Matthew Trager, Pramuditha Perera, Luca Zancato, Alessandro Achille, Parminder Bhatia, Stefano Soatto
We investigate compositional structures in data embeddings from pre-trained vision-language models (VLMs). Traditionally, compositionality has been associated with algebraic operations on embeddings of words from a pre-existing vocabulary. In contrast, we seek to approximate representations from an encoder as combinations of a smaller set of vectors in the embedding space. These vectors can be seen as "ideal words" for generating concepts directly within the embedding space of the model. We first present a framework for understanding compositional structures from a geometric perspective. We then explain what these compositional structures entail probabilistically in the case of VLM embeddings, providing intuitions for why they arise in practice. Finally, we empirically explore these structures in CLIP's embeddings and we evaluate their usefulness for solving different vision-language tasks such as classification, debiasing, and retrieval. Our results show that simple linear algebraic operations on embedding vectors can be used as compositional and interpretable methods for regulating the behavior of VLMs.
Authors: Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie
Protecting personal data against exploitation of machine learning models is crucial. Recently, availability attacks have shown great promise to provide an extra layer of protection against the unauthorized use of data to train neural networks. These methods aim to add imperceptible noise to clean data so that the neural networks cannot extract meaningful patterns from the protected data, claiming that they can make personal data "unexploitable." This paper provides a strong countermeasure against such approaches, showing that unexploitable data might only be an illusion. In particular, we leverage the power of diffusion models and show that a carefully designed denoising process can counteract the effectiveness of the data-protecting perturbations. We rigorously analyze our algorithm, and theoretically prove that the amount of required denoising is directly related to the magnitude of the data-protecting perturbations. Our approach, called AVATAR, delivers state-of-the-art performance against a suite of recent availability attacks in various scenarios, outperforming adversarial training even under distribution mismatch between the diffusion model and the protected data. Our findings call for more research into making personal data unexploitable, showing that this goal is far from over. Our implementation is available at this repository: https://github.com/hmdolatabadi/AVATAR.
Authors: Leida Zhang, Zhengda Lu, Kai Liu, Yiqun Wang
Learning-based point cloud registration methods can handle clean point clouds well, while it is still challenging to generalize to noisy, partial, and density-varying point clouds. To this end, we propose a novel point cloud registration framework for these imperfect point clouds. By introducing a neural implicit representation, we replace the problem of rigid registration between point clouds with a registration problem between the point cloud and the neural implicit function. We then propose to alternately optimize the implicit function and the registration between the implicit function and point cloud. In this way, point cloud registration can be performed in a coarse-to-fine manner. By fully capitalizing on the capabilities of the neural implicit function without computing point correspondences, our method showcases remarkable robustness in the face of challenges such as noise, incompleteness, and density changes of point clouds.
Authors: Tin Barisin, Katja Schladitz, Claudia Redenbach
Scale invariance of an algorithm refers to its ability to treat objects equally independently of their size. For neural networks, scale invariance is typically achieved by data augmentation. However, when presented with a scale far outside the range covered by the training set, neural networks may fail to generalize.
Here, we introduce the Riesz network, a novel scale invariant neural network. Instead of standard 2d or 3d convolutions for combining spatial information, the Riesz network is based on the Riesz transform which is a scale equivariant operation. As a consequence, this network naturally generalizes to unseen or even arbitrary scales in a single forward pass. As an application example, we consider detecting and segmenting cracks in tomographic images of concrete. In this context, 'scale' refers to the crack thickness which may vary strongly even within the same sample. To prove its scale invariance, the Riesz network is trained on one fixed crack width. We then validate its performance in segmenting simulated and real tomographic images featuring a wide range of crack widths. An additional experiment is carried out on the MNIST Large Scale data set.
Authors: Johannes F. Kunz, Stefan Ruschke, Reinhard Heckel
Cardiac magnetic resonance imaging (MRI) requires reconstructing a real-time video of a beating heart from continuous highly under-sampled measurements. This task is challenging since the object to be reconstructed (the heart) is continuously changing during signal acquisition. In this paper, we propose a reconstruction approach based on representing the beating heart with an implicit neural network and fitting the network so that the representation of the heart is consistent with the measurements. The network in the form of a multi-layer perceptron with Fourier-feature inputs acts as an effective signal prior and enables adjusting the regularization strength in both the spatial and temporal dimensions of the signal. We study the proposed approach for 2D free-breathing cardiac real-time MRI in different operating regimes, i.e., for different image resolutions, slice thicknesses, and acquisition lengths. Our method achieves reconstruction quality on par with or slightly better than state-of-the-art untrained convolutional neural networks and superior image quality compared to a recent method that fits an implicit representation directly to Fourier-domain measurements. However, this comes at a relatively high computational cost. Our approach does not require any additional patient data or biosensors including electrocardiography, making it potentially applicable in a wide range of clinical scenarios.
Authors: Zaibin Zhang, Yuanhang Zhang, Lijun Wang, Yifan Wang, Huchuan Lu
A popular approach for constructing bird's-eye-view (BEV) representation in 3D detection is to lift 2D image features onto the viewing frustum space based on explicitly predicted depth distribution. However, depth distribution can only characterize the 3D geometry of visible object surfaces but fails to capture their internal space and overall geometric structure, leading to sparse and unsatisfactory 3D representations. To mitigate this issue, we present BEV-IO, a new 3D detection paradigm to enhance BEV representation with instance occupancy information. At the core of our method is the newly-designed instance occupancy prediction (IOP) module, which aims to infer point-level occupancy status for each instance in the frustum space. To ensure training efficiency while maintaining representational flexibility, it is trained using the combination of both explicit and implicit supervision. With the predicted occupancy, we further design a geometry-aware feature propagation mechanism (GFP), which performs self-attention based on occupancy distribution along each ray in frustum and is able to enforce instance-level feature consistency. By integrating the IOP module with GFP mechanism, our BEV-IO detector is able to render highly informative 3D scene structures with more comprehensive BEV representations. Experimental results demonstrate that BEV-IO can outperform state-of-the-art methods while only adding a negligible increase in parameters (0.2%) and computational overhead (0.24%in GFLOPs).
Authors: Chih-Yang Lin, Chia-Yu Lin, Yu-Tso Liu, Timothy K. Shih
The utilization of Wi-Fi based human activity recognition has gained considerable interest in recent times, primarily owing to its applications in various domains such as healthcare for monitoring breath and heart rate, security, elderly care. These Wi-Fi-based methods exhibit several advantages over conventional state-of-the-art techniques that rely on cameras and sensors, including lower costs and ease of deployment. However, a significant challenge associated with Wi-Fi-based HAR is the significant decline in performance when the scene or subject changes. To mitigate this issue, it is imperative to train the model using an extensive dataset. In recent studies, the utilization of CNN-based models or sequence-to-sequence models such as LSTM, GRU, or Transformer has become prevalent. While sequence-to-sequence models can be more precise, they are also more computationally intensive and require a larger amount of training data. To tackle these limitations, we propose a novel approach that leverages a temporal convolution network with augmentations and attention, referred to as TCN-AA. Our proposed method is computationally efficient and exhibits improved accuracy even when the data size is increased threefold through our augmentation techniques. Our experiments on a publicly available dataset indicate that our approach outperforms existing state-of-the-art methods, with a final accuracy of 99.42%.
Authors: Xianqiang Lyu, Junhui Hou
The high-dimensional nature of the 4-D light field (LF) poses great challenges in achieving efficient and effective feature embedding, that severely impacts the performance of downstream tasks. To tackle this crucial issue, in contrast to existing methods with empirically-designed architectures, we propose a probabilistic-based feature embedding (PFE), which learns a feature embedding architecture by assembling various low-dimensional convolution patterns in a probability space for fully capturing spatial-angular information. Building upon the proposed PFE, we then leverage the intrinsic linear imaging model of the coded aperture camera to construct a cycle-consistent 4-D LF reconstruction network from coded measurements. Moreover, we incorporate PFE into an iterative optimization framework for 4-D LF denoising. Our extensive experiments demonstrate the significant superiority of our methods on both real-world and synthetic 4-D LF images, both quantitatively and qualitatively, when compared with state-of-the-art methods. The source code will be publicly available at https://github.com/lyuxianqiang/LFCA-CR-NET.
Authors: Fengheng Li, An Liu, Wei Feng, Honghe Zhu, Yaoyu Li, Zheng Zhang, Jingjing Lv, Xin Zhu, Junjie Shen, Zhangang Lin, Jingping Shao
Poster layout is a crucial aspect of poster design. Prior methods primarily focus on the correlation between visual content and graphic elements. However, a pleasant layout should also consider the relationship between visual and textual contents and the relationship between elements. In this study, we introduce a relation-aware diffusion model for poster layout generation that incorporates these two relationships in the generation process. Firstly, we devise a visual-textual relation-aware module that aligns the visual and textual representations across modalities, thereby enhancing the layout's efficacy in conveying textual information. Subsequently, we propose a geometry relation-aware module that learns the geometry relationship between elements by comprehensively considering contextual information. Additionally, the proposed method can generate diverse layouts based on user constraints. To advance research in this field, we have constructed a poster layout dataset named CGL-Dataset V2. Our proposed method outperforms state-of-the-art methods on CGL-Dataset V2. The data and code will be available at https://github.com/liuan0803/RADM.
Authors: Xiang Li, Varun Belagali, Jinghuan Shang, Michael S. Ryoo
Sequence modeling approaches have shown promising results in robot imitation learning. Recently, diffusion models have been adopted for behavioral cloning in a sequence modeling fashion, benefiting from their exceptional capabilities in modeling complex data distributions. The standard diffusion-based policy iteratively generates action sequences from random noise conditioned on the input states. Nonetheless, the model for diffusion policy can be further improved in terms of visual representations. In this work, we propose Crossway Diffusion, a simple yet effective method to enhance diffusion-based visuomotor policy learning via a carefully designed state decoder and an auxiliary self-supervised learning (SSL) objective. The state decoder reconstructs raw image pixels and other state information from the intermediate representations of the reverse diffusion process. The whole model is jointly optimized by the SSL objective and the original diffusion loss. Our experiments demonstrate the effectiveness of Crossway Diffusion in various simulated and real-world robot tasks, confirming its consistent advantages over the standard diffusion-based policy and substantial improvements over the baselines.
Authors: Sheng-Lan Liu, Yu-Ning Ding, Gang Yan, Si-Fan Zhang, Jin-Rong Zhang, Wen-Yue Chen, Ning Zhou, Xue-Hai Xu, Hao Liu
The fine-grained action analysis of the existing action datasets is challenged by insufficient action categories, low fine granularities, limited modalities, and tasks. In this paper, we propose a Multi-modality and Multi-task dataset of Figure Skating (MMFS) which was collected from the World Figure Skating Championships. MMFS, which possesses action recognition and action quality assessment, captures RGB, skeleton, and is collected the score of actions from 11671 clips with 256 categories including spatial and temporal labels. The key contributions of our dataset fall into three aspects as follows. (1) Independently spatial and temporal categories are first proposed to further explore fine-grained action recognition and quality assessment. (2) MMFS first introduces the skeleton modality for complex fine-grained action quality assessment. (3) Our multi-modality and multi-task dataset encourage more action analysis models. To benchmark our dataset, we adopt RGB-based and skeleton-based baseline methods for action recognition and action quality assessment.
Authors: Chenyu Zhang, Zhaozheng Yin, Ruwen Qin
Efficiently monitoring the condition of civil infrastructure requires automating the structural condition assessment in visual inspection. This paper proposes an Attention-Enhanced Co-Interactive Fusion Network (AECIF-Net) for automatic structural condition assessment in visual bridge inspection. AECIF-Net can simultaneously parse structural elements and segment surface defects on the elements in inspection images. It integrates two task-specific relearning subnets to extract task-specific features from an overall feature embedding. A co-interactive feature fusion module further captures the spatial correlation and facilitates information sharing between tasks. Experimental results demonstrate that the proposed AECIF-Net outperforms the current state-of-the-art approaches, achieving promising performance with 92.11% mIoU for element segmentation and 87.16% mIoU for corrosion segmentation on the test set of the new benchmark dataset Steel Bridge Condition Inspection Visual (SBCIV). An ablation study verifies the merits of the designs for AECIF-Net, and a case study demonstrates its capability to automate structural condition assessment.
Authors: Tin Barisin, Jesus Angulo, Katja Schladitz, Claudia Redenbach
Scattering networks yield powerful and robust hierarchical image descriptors which do not require lengthy training and which work well with very few training data. However, they rely on sampling the scale dimension. Hence, they become sensitive to scale variations and are unable to generalize to unseen scales. In this work, we define an alternative feature representation based on the Riesz transform. We detail and analyze the mathematical foundations behind this representation. In particular, it inherits scale equivariance from the Riesz transform and completely avoids sampling of the scale dimension. Additionally, the number of features in the representation is reduced by a factor four compared to scattering networks. Nevertheless, our representation performs comparably well for texture classification with an interesting addition: scale equivariance. Our method yields superior performance when dealing with scales outside of those covered by the training dataset. The usefulness of the equivariance property is demonstrated on the digit classification task, where accuracy remains stable even for scales four times larger than the one chosen for training. As a second example, we consider classification of textures.
Authors: Seyed Rohollah Hosseyni, Sanaz Seyedin, Hasan Taheri
Understanding the relationship between different parts of an image is crucial in a variety of applications, including object recognition, scene understanding, and image classification. Despite the fact that Convolutional Neural Networks (CNNs) have demonstrated impressive results in classifying and detecting objects, they lack the capability to extract the relationship between different parts of an image, which is a crucial factor in Human Action Recognition (HAR). To address this problem, this paper proposes a new module that functions like a convolutional layer that uses Vision Transformer (ViT). In the proposed model, the Vision Transformer can complement a convolutional neural network in a variety of tasks by helping it to effectively extract the relationship among various parts of an image. It is shown that the proposed model, compared to a simple CNN, can extract meaningful parts of an image and suppress the misleading parts. The proposed model has been evaluated on the Stanford40 and PASCAL VOC 2012 action datasets and has achieved 95.5% mean Average Precision (mAP) and 91.5% mAP results, respectively, which are promising compared to other state-of-the-art methods.
Authors: Yongxin Shao, Aihong Tan, Binrui Wang, Tianhong Yan, Zhetao Sun, Yiyang Zhang, Jiaxin Liu
LiDAR point clouds can effectively depict the motion and posture of objects in three-dimensional space. Many studies accomplish the 3D object detection by voxelizing point clouds. However, in autonomous driving scenarios, the sparsity and hollowness of point clouds create some difficulties for voxel-based methods. The sparsity of point clouds makes it challenging to describe the geometric features of objects. The hollowness of point clouds poses difficulties for the aggregation of 3D features. We propose a two-stage 3D object detection framework, called MS23D. (1) We propose a method using voxel feature points from multi-branch to construct the 3D feature layer. Using voxel feature points from different branches, we construct a relatively compact 3D feature layer with rich semantic features. Additionally, we propose a distance-weighted sampling method, reducing the loss of foreground points caused by downsampling and allowing the 3D feature layer to retain more foreground points. (2) In response to the hollowness of point clouds, we predict the offsets between deep-level feature points and the object's centroid, making them as close as possible to the object's centroid. This enables the aggregation of these feature points with abundant semantic features. For feature points from shallow-level, we retain them on the object's surface to describe the geometric features of the object. To validate our approach, we evaluated its effectiveness on both the KITTI and ONCE datasets.
Authors: Neil D. Dizon, Jeffrey A. Hogan
Recently, novel quaternion-valued wavelets on the plane were constructed using an optimisation approach. These wavelets are compactly supported, smooth, orthonormal, non-separable and truly quaternionic. However, they have not been tested in application. In this paper, we introduce a methodology for decomposing and reconstructing colour images using quaternionic wavelet filters associated to recently developed quaternion-valued wavelets on the plane. We investigate its applicability in compression, enhancement, segmentation, and denoising of colour images. Our results demonstrate these wavelets as promising tools for an end-to-end quaternion processing of colour images.
Authors: Gabriel Mejia, Paula Cárdenas, Daniela Ruiz, Angela Castillo, Pablo Arbeláez
Spatial transcriptomics is an emerging technology that aligns histopathology images with spatially resolved gene expression profiling. It holds the potential for understanding many diseases but faces significant bottlenecks such as specialized equipment and domain expertise. In this work, we present SEPAL, a new model for predicting genetic profiles from visual tissue appearance. Our method exploits the biological biases of the problem by directly supervising relative differences with respect to mean expression, and leverages local visual context at every coordinate to make predictions using a graph neural network. This approach closes the gap between complete locality and complete globality in current methods. In addition, we propose a novel benchmark that aims to better define the task by following current best practices in transcriptomics and restricting the prediction variables to only those with clear spatial patterns. Our extensive evaluation in two different human breast cancer datasets indicates that SEPAL outperforms previous state-of-the-art methods and other mechanisms of including spatial context.
Authors: Ziming Wang, Shumin Han, Xiaodi Wang, Jing Hao, Xianbin Cao, Baochang Zhang
Small CNN-based models usually require transferring knowledge from a large model before they are deployed in computationally resource-limited edge devices. Masked image modeling (MIM) methods achieve great success in various visual tasks but remain largely unexplored in knowledge distillation for heterogeneous deep models. The reason is mainly due to the significant discrepancy between the Transformer-based large model and the CNN-based small network. In this paper, we develop the first Heterogeneous Generative Knowledge Distillation (H-GKD) based on MIM, which can efficiently transfer knowledge from large Transformer models to small CNN-based models in a generative self-supervised fashion. Our method builds a bridge between Transformer-based models and CNNs by training a UNet-style student with sparse convolution, which can effectively mimic the visual representation inferred by a teacher over masked modeling. Our method is a simple yet effective learning paradigm to learn the visual representation and distribution of data from heterogeneous teacher models, which can be pre-trained using advanced generative methods. Extensive experiments show that it adapts well to various models and sizes, consistently achieving state-of-the-art performance in image classification, object detection, and semantic segmentation tasks. For example, in the Imagenet 1K dataset, H-GKD improves the accuracy of Resnet50 (sparse) from 76.98% to 80.01%.
Authors: Yiping Wei, Kunyu Peng, Alina Roitberg, Jiaming Zhang, Junwei Zheng, Ruiping Liu, Yufan Chen, Kailun Yang, Rainer Stiefelhagen
Self-supervised representation learning for human action recognition has developed rapidly in recent years. Most of the existing works are based on skeleton data while using a multi-modality setup. These works overlooked the differences in performance among modalities, which led to the propagation of erroneous knowledge between modalities while only three fundamental modalities, i.e., joints, bones, and motions are used, hence no additional modalities are explored.
In this work, we first propose an Implicit Knowledge Exchange Module (IKEM) which alleviates the propagation of erroneous knowledge between low-performance modalities. Then, we further propose three new modalities to enrich the complementary information between modalities. Finally, to maintain efficiency when introducing new modalities, we propose a novel teacher-student framework to distill the knowledge from the secondary modalities into the mandatory modalities considering the relationship constrained by anchors, positives, and negatives, named relational cross-modality knowledge distillation. The experimental results demonstrate the effectiveness of our approach, unlocking the efficient use of skeleton-based multi-modality data. Source code will be made publicly available at https://github.com/desehuileng0o0/IKEM.
Authors: Ningli Xu, Rongjun Qin, Debao Huang, Fabio Remondino
Neural Radiance Fields (NeRF) offer the potential to benefit 3D reconstruction tasks, including aerial photogrammetry. However, the scalability and accuracy of the inferred geometry are not well-documented for large-scale aerial assets,since such datasets usually result in very high memory consumption and slow convergence.. In this paper, we aim to scale the NeRF on large-scael aerial datasets and provide a thorough geometry assessment of NeRF. Specifically, we introduce a location-specific sampling technique as well as a multi-camera tiling (MCT) strategy to reduce memory consumption during image loading for RAM, representation training for GPU memory, and increase the convergence rate within tiles. MCT decomposes a large-frame image into multiple tiled images with different camera models, allowing these small-frame images to be fed into the training process as needed for specific locations without a loss of accuracy. We implement our method on a representative approach, Mip-NeRF, and compare its geometry performance with threephotgrammetric MVS pipelines on two typical aerial datasets against LiDAR reference data. Both qualitative and quantitative results suggest that the proposed NeRF approach produces better completeness and object details than traditional approaches, although as of now, it still falls short in terms of accuracy.
Authors: Noah Wiederhold, Ava Megyeri, DiMaggio Paris, Sean Banerjee, Natasha Kholgade Banerjee
We present the HOH (Human-Object-Human) Handover Dataset, a large object count dataset with 136 objects, to accelerate data-driven research on handover studies, human-robot handover implementation, and artificial intelligence (AI) on handover parameter estimation from 2D and 3D data of person interactions. HOH contains multi-view RGB and depth data, skeletons, fused point clouds, grasp type and handedness labels, object, giver hand, and receiver hand 2D and 3D segmentations, giver and receiver comfort ratings, and paired object metadata and aligned 3D models for 2,720 handover interactions spanning 136 objects and 20 giver-receiver pairs-40 with role-reversal-organized from 40 participants. We also show experimental results of neural networks trained using HOH to perform grasp, orientation, and trajectory prediction. As the only fully markerless handover capture dataset, HOH represents natural human-human handover interactions, overcoming challenges with markered datasets that require specific suiting for body tracking, and lack high-resolution hand tracking. To date, HOH is the largest handover dataset in number of objects, participants, pairs with role reversal accounted for, and total interactions captured.
Authors: Jianhong Bai, Zuozhu Liu, Hualiang Wang, Ruizhe Chen, Lianrui Mu, Xiaomeng Li, Joey Tianyi Zhou, Yang Feng, Jian Wu, Haoji Hu
Data imbalance and open-ended distribution are two intrinsic characteristics of the real visual world. Though encouraging progress has been made in tackling each challenge separately, few works dedicated to combining them towards real-world scenarios. While several previous works have focused on classifying close-set samples and detecting open-set samples during testing, it's still essential to be able to classify unknown subjects as human beings. In this paper, we formally define a more realistic task as distribution-agnostic generalized category discovery (DA-GCD): generating fine-grained predictions for both close- and open-set classes in a long-tailed open-world setting. To tackle the challenging problem, we propose a Self-Balanced Co-Advice contrastive framework (BaCon), which consists of a contrastive-learning branch and a pseudo-labeling branch, working collaboratively to provide interactive supervision to resolve the DA-GCD task. In particular, the contrastive-learning branch provides reliable distribution estimation to regularize the predictions of the pseudo-labeling branch, which in turn guides contrastive learning through self-balanced knowledge transfer and a proposed novel contrastive loss. We compare BaCon with state-of-the-art methods from two closely related fields: imbalanced semi-supervised learning and generalized category discovery. The effectiveness of BaCon is demonstrated with superior performance over all baselines and comprehensive analysis across various datasets. Our code is publicly available.
Authors: Pronob Kumar Barman
Nowadays, detecting aberrant health issues is a difficult process. Falling, especially among the elderly, is a severe concern worldwide. Falls can result in deadly consequences, including unconsciousness, internal bleeding, and often times, death. A practical and optimal, smart approach of detecting falling is currently a concern. The use of vision-based fall monitoring is becoming more common among scientists as it enables senior citizens and those with other health conditions to live independently. For tracking, surveillance, and rescue, unmanned aerial vehicles use video or image segmentation and object detection methods. The Tello drone is equipped with a camera and with this device we determined normal and abnormal behaviors among our participants. The autonomous falling objects are classified using a convolutional neural network (CNN) classifier. The results demonstrate that the systems can identify falling objects with a precision of 0.9948.
Authors: Jianwei Li, Weizhi Gao, Qi Lei, Dongkuan Xu
It is widely acknowledged that large and sparse models have higher accuracy than small and dense models under the same model size constraints. This motivates us to train a large model and then remove its redundant neurons or weights by pruning. Most existing works pruned the networks in a deterministic way, the performance of which solely depends on a single pruning criterion and thus lacks variety. Instead, in this paper, we propose a model pruning strategy that first generates several pruning masks in a designed random way. Subsequently, along with an effective mask-selection rule, the optimal mask is chosen from the pool of mask candidates. To further enhance efficiency, we introduce an early mask evaluation strategy, mitigating the overhead associated with training multiple masks. Our extensive experiments demonstrate that this approach achieves state-of-the-art performance across eight datasets from GLUE, particularly excelling at high levels of sparsity.
Authors: Yusuke Akamatsu, Terumi Umematsu, Hitoshi Imaoka
Video-based heart and respiratory rate measurements using facial videos are more useful and user-friendly than traditional contact-based sensors. However, most of the current deep learning approaches require ground-truth pulse and respiratory waves for model training, which are expensive to collect. In this paper, we propose CalibrationPhys, a self-supervised video-based heart and respiratory rate measurement method that calibrates between multiple cameras. CalibrationPhys trains deep learning models without supervised labels by using facial videos captured simultaneously by multiple cameras. Contrastive learning is performed so that the pulse and respiratory waves predicted from the synchronized videos using multiple cameras are positive and those from different videos are negative. CalibrationPhys also improves the robustness of the models by means of a data augmentation technique and successfully leverages a pre-trained model for a particular camera. Experimental results utilizing two datasets demonstrate that CalibrationPhys outperforms state-of-the-art heart and respiratory rate measurement methods. Since we optimize camera-specific models using only videos from multiple cameras, our approach makes it easy to use arbitrary cameras for heart and respiratory rate measurements.
Authors: Jameel Hassan, Hanan Gani, Noor Hussein, Muhammad Uzair Khattak, Muzammal Naseer, Fahad Shahbaz Khan, Salman Khan
The promising zero-shot generalization of vision-language models such as CLIP has led to their adoption using prompt learning for numerous downstream tasks. Previous works have shown test-time prompt tuning using entropy minimization to adapt text prompts for unseen domains. While effective, this overlooks the key cause for performance degradation to unseen domains -- distribution shift. In this work, we explicitly handle this problem by aligning the out-of-distribution (OOD) test sample statistics to those of the source data using prompt tuning. We use a single test sample to adapt multi-modal prompts at test time by minimizing the feature distribution shift to bridge the gap in the test domain. Evaluating against the domain generalization benchmark, our method improves zero-shot top- 1 accuracy beyond existing prompt-learning techniques, with a 3.08% improvement over the baseline MaPLe. In cross-dataset generalization with unseen categories across 10 datasets, our method improves consistently across all datasets compared to the existing state-of-the-art. Our source code and models are available at https://jameelhassan.github.io/promptalign.
Authors: Cheng Cheng, Lin Song, Ruoyi Xue, Hang Wang, Hongbin Sun, Yixiao Ge, Ying Shan
The contrastive vision-language pre-training, known as CLIP, demonstrates remarkable potential in perceiving open-world visual concepts, enabling effective zero-shot image recognition. Nevertheless, few-shot learning methods based on CLIP typically require offline fine-tuning of the parameters on few-shot samples, resulting in longer inference time and the risk of over-fitting in certain domains. To tackle these challenges, we propose the Meta-Adapter, a lightweight residual-style adapter, to refine the CLIP features guided by the few-shot samples in an online manner. With a few training samples, our method can enable effective few-shot learning capabilities and generalize to unseen data or tasks without additional fine-tuning, achieving competitive performance and high efficiency. Without bells and whistles, our approach outperforms the state-of-the-art online few-shot learning method by an average of 3.6\% on eight image classification datasets with higher inference speed. Furthermore, our model is simple and flexible, serving as a plug-and-play module directly applicable to downstream tasks. Without further fine-tuning, Meta-Adapter obtains notable performance improvements in open-vocabulary object detection and segmentation tasks.
Authors: Zhengcong Fei, Mingyuan Fan, Junshi Huang
This paper presents that the masked-modeling principle driving the success of large foundational vision models can be effectively applied to audio by making predictions in a latent space. We introduce Audio-based Joint-Embedding Predictive Architecture (A-JEPA), a simple extension method for self-supervised learning from the audio spectrum. Following the design of I-JEPA, our A-JEPA encodes visible audio spectrogram patches with a curriculum masking strategy via context encoder, and predicts the representations of regions sampled at well-designed locations. The target representations of those regions are extracted by the exponential moving average of context encoder, \emph{i.e.}, target encoder, on the whole spectrogram. We find it beneficial to transfer random block masking into time-frequency aware masking in a curriculum manner, considering the complexity of highly correlated in local time and frequency in audio spectrograms. To enhance contextual semantic understanding and robustness, we fine-tune the encoder with a regularized masking on target datasets, instead of input dropping or zero. Empirically, when built with Vision Transformers structure, we find A-JEPA to be highly scalable and sets new state-of-the-art performance on multiple audio and speech classification tasks, outperforming other recent models that use externally supervised pre-training.
Authors: Salvatore Contino, Luca Cruciata, Orazio Gambino, Roberto Pirrone
Background and Objective: In recent years, Artificial Intelligence (AI) and in particular Deep Neural Networks (DNN) became a relevant research topic in biomedical image segmentation due to the availability of more and more data sets along with the establishment of well known competitions. Despite the popularity of DNN based segmentation on the research side, these techniques are almost unused in the daily clinical practice even if they could support effectively the physician during the diagnostic process. Apart from the issues related to the explainability of the predictions of a neural model, such systems are not integrated in the diagnostic workflow, and a standardization of their use is needed to achieve this goal. Methods: This paper presents IODeep a new DICOM Information Object Definition (IOD) aimed at storing both the weights and the architecture of a DNN already trained on a particular image dataset that is labeled as regards the acquisition modality, the anatomical region, and the disease under investigation. Results: The IOD architecture is presented along with a DNN selection algorithm from the PACS server based on the labels outlined above, and a simple PACS viewer purposely designed for demonstrating the effectiveness of the DICOM integration, while no modifications are required on the PACS server side. Also a service based architecture in support of the entire workflow has been implemented. Conclusion: IODeep ensures full integration of a trained AI model in a DICOM infrastructure, and it is also enables a scenario where a trained model can be either fine-tuned with hospital data or trained in a federated learning scheme shared by different hospitals. In this way AI models can be tailored to the real data produced by a Radiology ward thus improving the physician decision making process. Source code is freely available at https://github.com/CHILab1/IODeep.git
Authors: Cheng Chen, Jingkuan Song, LianLi Gao, Heng Tao Shen
Catastrophic Forgetting (CF) is a prominent issue in continual learning. Parameter isolation addresses this challenge by masking a sub-network for each task to mitigate interference with old tasks. However, these sub-networks are constructed relying on weight magnitude, which does not necessarily correspond to the importance of weights, resulting in maintaining unimportant weights and constructing redundant sub-networks. To overcome this limitation, inspired by information bottleneck, which removes redundancy between adjacent network layers, we propose \textbf{\underline{I}nformation \underline{B}ottleneck \underline{M}asked sub-network (IBM)} to eliminate redundancy within sub-networks. Specifically, IBM accumulates valuable information into essential weights to construct redundancy-free sub-networks, not only effectively mitigating CF by freezing the sub-networks but also facilitating new tasks training through the transfer of valuable knowledge. Additionally, IBM decomposes hidden representations to automate the construction process and make it flexible. Extensive experiments demonstrate that IBM consistently outperforms state-of-the-art methods. Notably, IBM surpasses the state-of-the-art parameter isolation method with a 70\% reduction in the number of parameters within sub-networks and an 80\% decrease in training time.
Authors: Amir Hertz, Andrey Voynov, Shlomi Fruchter, Daniel Cohen-Or
Large-scale Text-to-Image (T2I) models have rapidly gained prominence across creative fields, generating visually compelling outputs from textual prompts. However, controlling these models to ensure consistent style remains challenging, with existing methods necessitating fine-tuning and manual intervention to disentangle content and style. In this paper, we introduce StyleAligned, a novel technique designed to establish style alignment among a series of generated images. By employing minimal `attention sharing' during the diffusion process, our method maintains style consistency across images within T2I models. This approach allows for the creation of style-consistent images using a reference style through a straightforward inversion operation. Our method's evaluation across diverse styles and text prompts demonstrates high-quality synthesis and fidelity, underscoring its efficacy in achieving consistent style across various inputs.
Authors: Xulu Zhang, Xiao-Yong Wei, Jinlin Wu, Tianyi Zhang, Zhaoxiang Zhang, Zhen Lei, Qing Li
Inversion methods, such as Textual Inversion, generate personalized images by incorporating concepts of interest provided by user images. However, existing methods often suffer from overfitting issues, where the dominant presence of inverted concepts leads to the absence of other desired concepts. It stems from the fact that during inversion, the irrelevant semantics in the user images are also encoded, forcing the inverted concepts to occupy locations far from the core distribution in the embedding space. To address this issue, we propose a method that guides the inversion process towards the core distribution for compositional embeddings. Additionally, we introduce a spatial regularization approach to balance the attention on the concepts being composed. Our method is designed as a post-training approach and can be seamlessly integrated with other inversion methods. Experimental results demonstrate the effectiveness of our proposed approach in mitigating the overfitting problem and generating more diverse and balanced compositions of concepts in the synthesized images. The source code is available at https://github.com/zhangxulu1996/Compositional-Inversion.
Authors: Ethan Thomas, Salman Aslam
Cardiovascular diseases, particularly heart failure, are a leading cause of death globally. The early detection of heart failure through routine echocardiogram screenings is often impeded by the high cost and labor-intensive nature of these procedures, a barrier that can mean the difference between life and death. This paper presents ConFormer, a novel deep learning model designed to automate the estimation of Ejection Fraction (EF) and Left Ventricular Wall Thickness from echocardiograms. The implementation of ConFormer has the potential to enhance preventative cardiology by enabling cost-effective, accessible, and comprehensive heart health monitoring, thereby saving countless lives. The source code is available at https://github.com/Aether111/ConFormer.
Authors: Kefu Yi, Kai Luo, Xiaolei Luo, Jiangui Huang, Hao Wu, Rongdong Hu, Wei Hao
Multi-object tracking (MOT) in video sequences remains a challenging task, especially in scenarios with significant camera movements. This is because targets can drift considerably on the image plane, leading to erroneous tracking outcomes. Addressing such challenges typically requires supplementary appearance cues or Camera Motion Compensation (CMC). While these strategies are effective, they also introduce a considerable computational burden, posing challenges for real-time MOT. In response to this, we introduce UCMCTrack, a novel motion model-based tracker robust to camera movements. Unlike conventional CMC that computes compensation parameters frame-by-frame, UCMCTrack consistently applies the same compensation parameters throughout a video sequence. It employs a Kalman filter on the ground plane and introduces the Mapped Mahalanobis Distance (MMD) as an alternative to the traditional Intersection over Union (IoU) distance measure. By leveraging projected probability distributions on the ground plane, our approach efficiently captures motion patterns and adeptly manages uncertainties introduced by homography projections. Remarkably, UCMCTrack, relying solely on motion cues, achieves state-of-the-art performance across a variety of challenging datasets, including MOT17, MOT20, DanceTrack and KITTI. More details and code are available at https://github.com/corfyi/UCMCTrack
Authors: Huy Le, Tung Kieu, Anh Nguyen, Ngan Le
Text-video retrieval, a prominent sub-field within the domain of multimodal information retrieval, has witnessed remarkable growth in recent years. However, existing methods assume video scenes are consistent with unbiased descriptions. These limitations fail to align with real-world scenarios since descriptions can be influenced by annotator biases, diverse writing styles, and varying textual perspectives. To overcome the aforementioned problems, we introduce $\texttt{WAVER}$, a cross-domain knowledge distillation framework via vision-language models through open-vocabulary knowledge designed to tackle the challenge of handling different writing styles in video descriptions. $\texttt{WAVER}$ capitalizes on the open-vocabulary properties that lie in pre-trained vision-language models and employs an implicit knowledge distillation approach to transfer text-based knowledge from a teacher model to a vision-based student. Empirical studies conducted across four standard benchmark datasets, encompassing various settings, provide compelling evidence that $\texttt{WAVER}$ can achieve state-of-the-art performance in text-video retrieval task while handling writing-style variations. The code is available at: https://github.com/Fsoft-AIC/WAVER
Authors: Tianxiang Hao, Mengyao Lyu, Hui Chen, Sicheng Zhao, Jungong Han, Guiguang Ding
With the development of large pre-trained vision-language models, how to effectively transfer the knowledge of such foundational models to downstream tasks becomes a hot topic, especially in a data-deficient scenario. Recently, prompt tuning has become a popular solution. When adapting the vision-language models, researchers freeze the parameters in the backbone and only design and tune the prompts. On the one hand, the delicate design of prompt tuning exhibits strong performance. On the other hand, complicated structures and update rules largely increase the computation and storage cost. Motivated by the observation that the evolution pattern of the generalization capability in visual-language models aligns harmoniously with the trend of rank variations in the prompt matrix during adaptation, we design a new type of prompt, Re-parameterized Low-rank Prompt (RLP), for both efficient and effective adaptation. Our method could largely reduce the number of tunable parameters and storage space, which is quite beneficial in resource-limited scenarios. Extensive experiments further demonstrate the superiority of RLP. In particular, RLP shows comparable or even stronger performance than the latest state-of-the-art methods with an extremely small number of parameters. On a series of tasks over 11 datasets, RLP significantly increases the average downstream accuracy of classic prompt tuning by up to 5.25% using merely 0.5K parameters.
Authors: Xiaolong Shen, Jianxin Ma, Chang Zhou, Zongxin Yang
Generating photorealistic 3D faces from given conditions is a challenging task. Existing methods often rely on time-consuming one-by-one optimization approaches, which are not efficient for modeling the same distribution content, e.g., faces. Additionally, an ideal controllable 3D face generation model should consider both facial attributes and expressions. Thus we propose a novel approach called TEx-Face(TExt & Expression-to-Face) that addresses these challenges by dividing the task into three components, i.e., 3D GAN Inversion, Conditional Style Code Diffusion, and 3D Face Decoding. For 3D GAN inversion, we introduce two methods which aim to enhance the representation of style codes and alleviate 3D inconsistencies. Furthermore, we design a style code denoiser to incorporate multiple conditions into the style code and propose a data augmentation strategy to address the issue of insufficient paired visual-language data. Extensive experiments conducted on FFHQ, CelebA-HQ, and CelebA-Dialog demonstrate the promising performance of our TEx-Face in achieving the efficient and controllable generation of photorealistic 3D faces. The code will be available at https://github.com/sxl142/TEx-Face.
Authors: Layton Aho, Mark Winter, Marc DeCarlo, Agne Frismantiene, Yannick Blum, Paolo Armando Gagliardi, Olivier Pertz, Andrew R. Cohen
Live cell microscopy captures 5-D $(x,y,z,channel,time)$ movies that display patterns of cellular motion and signaling dynamics. We present here an approach to finding spatiotemporal patterns of cell signaling dynamics in 5-D live cell microscopy movies unique in requiring no a priori knowledge of expected pattern dynamics, and no training data. The proposed cell signaling structure function (SSF) is a Kolmogorov structure function that optimally measures cell signaling state as nuclear intensity w.r.t. surrounding cytoplasm, a significant improvement compared to the current state-of-the-art cytonuclear ratio. SSF kymographs store at each spatiotemporal cell centroid the SSF value, or a functional output such as velocity. Patterns of similarity are identified via the metric normalized compression distance (NCD). The NCD is a reproducing kernel for a Hilbert space that represents the input SSF kymographs as points in a low dimensional embedding that optimally captures the pattern similarity identified by the NCD throughout the space. The only parameter is the expected cell radii ($\mu m$). A new formulation of the cluster structure function optimally estimates how meaningful an embedding from the RKHS representation. Results are presented quantifying the impact of ERK and AKT signaling between different oncogenic mutations, and by the relation between ERK signaling and cellular velocity patterns for movies of 2-D monolayers of human breast epithelial (MCF10A) cells, 3-D MCF10A spheroids under optogenetic manipulation of ERK, and human induced pluripotent stem cells .
Authors: Seyed Mohammad Hossein Hashemi, Leila Safari, Amirhossein Dadashzade Taromi
In the field of medical sciences, reliable detection and classification of brain tumors from images remains a formidable challenge due to the rarity of tumors within the population of patients. Therefore, the ability to detect tumors in anomaly scenarios is paramount for ensuring timely interventions and improved patient outcomes. This study addresses the issue by leveraging deep learning (DL) techniques to detect and classify brain tumors in challenging situations. The curated data set from the National Brain Mapping Lab (NBML) comprises 81 patients, including 30 Tumor cases and 51 Normal cases. The detection and classification pipelines are separated into two consecutive tasks. The detection phase involved comprehensive data analysis and pre-processing to modify the number of image samples and the number of patients of each class to anomaly distribution (9 Normal per 1 Tumor) to comply with real world scenarios. Next, in addition to common evaluation metrics for the testing, we employed a novel performance evaluation method called Patient to Patient (PTP), focusing on the realistic evaluation of the model. In the detection phase, we fine-tuned a YOLOv8n detection model to detect the tumor region. Subsequent testing and evaluation yielded competitive performance both in Common Evaluation Metrics and PTP metrics. Furthermore, using the Data Efficient Image Transformer (DeiT) module, we distilled a Vision Transformer (ViT) model from a fine-tuned ResNet152 as a teacher in the classification phase. This approach demonstrates promising strides in reliable tumor detection and classification, offering potential advancements in tumor diagnosis for real-world medical imaging scenarios.
Authors: Hanem Ellethy, Shekhar S. Chandra, Viktor Vegh
Traumatic Brain Injury (TBI) poses a significant global public health challenge, contributing to high morbidity and mortality rates and placing a substantial economic burden on healthcare systems worldwide. The diagnosis of TBI relies on clinical information along with Computed Tomography (CT) scans. Addressing the multifaceted challenges posed by TBI has seen the development of innovative, data-driven approaches, for this complex condition. Particularly noteworthy is the prevalence of mild TBI (mTBI), which constitutes the majority of TBI cases where conventional methods often fall short. As such, we review the state-of-the-art Machine Learning (ML) techniques applied to clinical information and CT scans in TBI, with a particular focus on mTBI. We categorize ML applications based on their data sources, and there is a spectrum of ML techniques used to date. Most of these techniques have primarily focused on diagnosis, with relatively few attempts at predicting the prognosis. This review may serve as a source of inspiration for future research studies aimed at improving the diagnosis of TBI using data-driven approaches and standard diagnostic data.
Authors: Jiquan Yuan, Xinyan Cao, Jinming Che, Qinyuan Wang, Sen Liang, Wei Ren, Jinlong Lin, Xixin Cao
Recently, AIGC image quality assessment (AIGCIQA), which aims to assess the quality of AI-generated images (AIGIs) from a human perception perspective, has emerged as a new topic in computer vision. Unlike common image quality assessment tasks where images are derived from original ones distorted by noise, blur, and compression, \textit{etc.}, in AIGCIQA tasks, images are typically generated by generative models using text prompts. Considerable efforts have been made in the past years to advance AIGCIQA. However, most existing AIGCIQA methods regress predicted scores directly from individual generated images, overlooking the information contained in the text prompts of these images. This oversight partially limits the performance of these AIGCIQA methods. To address this issue, we propose a text-image encoder-based regression (TIER) framework. Specifically, we process the generated images and their corresponding text prompts as inputs, utilizing a text encoder and an image encoder to extract features from these text prompts and generated images, respectively. To demonstrate the effectiveness of our proposed TIER method, we conduct extensive experiments on several mainstream AIGCIQA databases, including AGIQA-1K, AGIQA-3K, and AIGCIQA2023. The experimental results indicate that our proposed TIER method generally demonstrates superior performance compared to baseline in most cases.
Authors: Xiaojuan Wang, Taesung Park, Yang Zhou, Eli Shechtman, Richard Zhang
A jump cut offers an abrupt, sometimes unwanted change in the viewing experience. We present a novel framework for smoothing these jump cuts, in the context of talking head videos. We leverage the appearance of the subject from the other source frames in the video, fusing it with a mid-level representation driven by DensePose keypoints and face landmarks. To achieve motion, we interpolate the keypoints and landmarks between the end frames around the cut. We then use an image translation network from the keypoints and source frames, to synthesize pixels. Because keypoints can contain errors, we propose a cross-modal attention scheme to select and pick the most appropriate source amongst multiple options for each key point. By leveraging this mid-level representation, our method can achieve stronger results than a strong video interpolation baseline. We demonstrate our method on various jump cuts in the talking head videos, such as cutting filler words, pauses, and even random cuts. Our experiments show that we can achieve seamless transitions, even in the challenging cases where the talking head rotates or moves drastically in the jump cut.
Authors: Mateusz Krubiński, Stefan Matcovici, Diana Grigore, Daniel Voinea, Alin-Ionut Popa
Watermark text spotting in document images can offer access to an often unexplored source of information, providing crucial evidence about a record's scope, audience and sometimes even authenticity. Stemming from the problem of text spotting, detecting and understanding watermarks in documents inherits the same hardships - in the wild, writing can come in various fonts, sizes and forms, making generic recognition a very difficult problem. To address the lack of resources in this field and propel further research, we propose a novel benchmark (K-Watermark) containing 65,447 data samples generated using Wrender, a watermark text patterns rendering procedure. A validity study using humans raters yields an authenticity score of 0.51 against pre-generated watermarked documents. To prove the usefulness of the dataset and rendering technique, we developed an end-to-end solution (Wextract) for detecting the bounding box instances of watermark text, while predicting the depicted text. To deal with this specific task, we introduce a variance minimization loss and a hierarchical self-attention mechanism. To the best of our knowledge, we are the first to propose an evaluation benchmark and a complete solution for retrieving watermarks from documents surpassing baselines by 5 AP points in detection and 4 points in character accuracy.