Authors: Juno Hwang, Yong-Hyun Park, Junghyo Jo
Diffusion models generate high-resolution images through iterative stochastic processes. In particular, the denoising method is one of the most popular approaches that predicts the noise in samples and denoises it at each time step. It has been commonly observed that the resolution of generated samples changes over time, starting off blurry and coarse, and becoming sharper and finer. In this paper, we introduce "resolution chromatography" that indicates the signal generation rate of each resolution, which is very helpful concept to mathematically explain this coarse-to-fine behavior in generation process, to understand the role of noise schedule, and to design time-dependent modulation. Using resolution chromatography, we determine which resolution level becomes dominant at a specific time step, and experimentally verify our theory with text-to-image diffusion models. We also propose some direct applications utilizing the concept: upscaling pre-trained models to higher resolutions and time-dependent prompt composing. Our theory not only enables a better understanding of numerous pre-existing techniques for manipulating image generation, but also suggests the potential for designing better noise schedules.
Authors: Jiawei Jiang, Yinwei Li, Shaowen Luo, Ping Li, Yiming Zhu
Video synthetic aperture radar (SAR) is attracting more attention in recent years due to its abilities of high resolution, high frame rate and advantages in continuous observation. Generally, the polar format algorithm (PFA) is an efficient algorithm for spotlight mode video SAR. However, in the process of PFA, the wavefront curvature error (WCE) limits the imaging scene size and the 2-D interpolation affects the efficiency. To solve the aforementioned problems, a beam-segmenting PFA based on principle of chirp scaling (PCS), called BS-PCS-PFA, is proposed for video SAR imaging, which has the capability of persistent imaging for different carrier frequencies video SAR. Firstly, an improved PCS applicable to video SAR PFA is proposed to replace the 2-D interpolation and the coarse image in the ground output coordinate system (GOCS) is obtained. As for the distortion or defocus existing in the coarse image, a novel sub-block imaging method based on beam-segmenting fast filtering is proposed to segment the image into multiple sub-beam data, whose distortion and defocus can be ignored when the equivalent size of sub-block is smaller than the distortion negligible region. Through processing the sub-beam data and mosaicking the refocused subimages, the full image in GOCS without distortion and defocus is obtained. Moreover, a three-step MoCo method is applied to the algorithm for the adaptability to the actual irregular trajectories. The proposed method can significantly expand the effective scene size of PFA, and the better operational efficiency makes it more suitable for video SAR imaging. The feasibility of the algorithm is verified by the experimental data.
Authors: Vahid Ahmadi Kalkhorani, Qingquan Zhang, Guanqun Song, Ting Zhu
Video smmarization is a crucial method to reduce the time of videos which reduces the spent time to watch/review a long video. This apporach has became more important as the amount of publisehed video is increasing everyday. A single or multiple videos can be summarized into a relatively short video using various of techniques from multimodal audio-visual techniques, to natural language processing approaches. Audiovisual techniques may be used to recognize significant visual events and pick the most important parts, while NLP techniques can be used to evaluate the audio transcript and extract the main sentences (timestamps) and corresponding video frames from the original video. Another approach is to use the best of both domain. Meaning that we can use audio-visual cues as well as video transcript to extract and summarize the video. In this paper, we combine a variety of NLP techniques (extractive and contect-based summarizers) with video processing techniques to convert a long video into a single relatively short video. We design this toll in a way that user can specify the relative length of the summarized video. We have also explored ways of summarizing and concatenating multiple videos into a single short video which will help having most important concepts from the same subject in a single short video. Out approach shows that video summarizing is a difficult but significant work, with substantial potential for further research and development, and it is possible thanks to the development of NLP models.
Authors: Yuteng Liu, Haowen Li, Haishan Zou, Jing Lu, Zhibin Lin
Active headrests can reduce low-frequency noise around ears based on active noise control (ANC) system. Both the control system using fixed control filters and the remote microphone-based adaptive control system provide good noise reduction performance when the head is in the original position. However, their performance degrades significantly when the head is in motion. In this paper, a human ear-positioning system based on the depth camera is introduced to address this problem. The system uses RTMpose model to estimate the two-dimensional (2D) positions of the ears in the color frame, and then derives the corresponding three-dimensional (3D) coordinates in the depth frame with a depth camera. Experimental results show that the ear-positioning system can effectively track the movement of ears, and the broadband noise reduction performance of the active headrest combined with the system is significantly improved when the human head is translating or rotating.
Authors: Xiang Li, Kevin M. Short
This paper explores the null space properties of neural networks. We extend the null space definition from linear to nonlinear maps and discuss the presence of a null space in neural networks. The null space of a given neural network can tell us the part of the input data that makes no contribution to the final prediction so that we can use it to trick the neural network. This reveals an inherent weakness in neural networks that can be exploited. One application described here leads to a method of image steganography. Through experiments on image datasets such as MNIST, we show that we can use null space components to force the neural network to choose a selected hidden image class, even though the overall image can be made to look like a completely different image. We conclude by showing comparisons between what a human viewer would see, and the part of the image that the neural network is actually using to make predictions and, hence, show that what the neural network ``sees'' is completely different than what we would expect.
Authors: Yikang Wei, Yahong Han
Federated Domain Generalization aims to learn a domain-invariant model from multiple decentralized source domains for deployment on unseen target domain. Due to privacy concerns, the data from different source domains are kept isolated, which poses challenges in bridging the domain gap. To address this issue, we propose a Multi-source Collaborative Gradient Discrepancy Minimization (MCGDM) method for federated domain generalization. Specifically, we propose intra-domain gradient matching between the original images and augmented images to avoid overfitting the domain-specific information within isolated domains. Additionally, we propose inter-domain gradient matching with the collaboration of other domains, which can further reduce the domain shift across decentralized domains. Combining intra-domain and inter-domain gradient matching, our method enables the learned model to generalize well on unseen domains. Furthermore, our method can be extended to the federated domain adaptation task by fine-tuning the target model on the pseudo-labeled target domain. The extensive experiments on federated domain generalization and adaptation indicate that our method outperforms the state-of-the-art methods significantly.
Authors: Hongjun Zhang
Through examples of coordinate and probability transformation between different distributions, the basic principle of normalizing flow is introduced in a simple and concise manner. From the perspective of the distribution of random variable function, the essence of probability transformation is explained, and the scaling factor Jacobian determinant of probability transformation is introduced. Treating the dataset as a sample from the population, obtaining normalizing flow is essentially through sampling surveys to statistically infer the numerical features of the population, and then the loss function is established by using the maximum likelihood estimation method. This article introduces how normalizing flow cleverly solves the two major application challenges of high-dimensional matrix determinant calculation and neural network reversible transformation. Using symmetric structured image dataset of three-span beam bridge, arch bridge, cable-stayed bridge and suspension bridge, constructing and training normalizing flow based on the Glow API in the TensorFlow Probability library. The model can smoothly transform the complex distribution of the bridge dataset into a standard normal distribution, and from the obtained latent space sampling, it can generate new bridge types that are different from the training dataset.
Authors: Yang Sui, Miao Yin, Yu Gong, Jinqi Xiao, Huy Phan, Bo Yuan
Low-rank compression, a popular model compression technique that produces compact convolutional neural networks (CNNs) with low rankness, has been well-studied in the literature. On the other hand, low-rank training, as an alternative way to train low-rank CNNs from scratch, has been exploited little yet. Unlike low-rank compression, low-rank training does not need pre-trained full-rank models, and the entire training phase is always performed on the low-rank structure, bringing attractive benefits for practical applications. However, the existing low-rank training solutions still face several challenges, such as a considerable accuracy drop and/or still needing to update full-size models during the training. In this paper, we perform a systematic investigation on low-rank CNN training. By identifying the proper low-rank format and performance-improving strategy, we propose ELRT, an efficient low-rank training solution for high-accuracy, high-compactness, low-rank CNN models. Our extensive evaluation results for training various CNNs on different datasets demonstrate the effectiveness of ELRT.
Authors: Vandan Gorade, Sparsh Mittal, Debesh Jha, Rekha Singhal, Ulas Bagci
Deep learning has demonstrated remarkable achievements in medical image segmentation. However, prevailing deep learning models struggle with poor generalization due to (i) intra-class variations, where the same class appears differently in different samples, and (ii) inter-class independence, resulting in difficulties capturing intricate relationships between distinct objects, leading to higher false negative cases. This paper presents a novel approach that synergies spatial and spectral representations to enhance domain-generalized medical image segmentation. We introduce the innovative Spectral Correlation Coefficient objective to improve the model's capacity to capture middle-order features and contextual long-range dependencies. This objective complements traditional spatial objectives by incorporating valuable spectral information. Extensive experiments reveal that optimizing this objective with existing architectures like UNet and TransUNet significantly enhances generalization, interpretability, and noise robustness, producing more confident predictions. For instance, in cardiac segmentation, we observe a 0.81 pp and 1.63 pp (pp = percentage point) improvement in DSC over UNet and TransUNet, respectively. Our interpretability study demonstrates that, in most tasks, objectives optimized with UNet outperform even TransUNet by introducing global contextual information alongside local details. These findings underscore the versatility and effectiveness of our proposed method across diverse imaging modalities and medical domains.
Authors: Chetan M Badgujar, Alwin Poulose, Hao Gan
Vision is a major component in several digital technologies and tools used in agriculture. The object detector, You Look Only Once (YOLO), has gained popularity in agriculture in a relatively short span due to its state-of-the-art performance. YOLO offers real-time detection with good accuracy and is implemented in various agricultural tasks, including monitoring, surveillance, sensing, automation, and robotics. The research and application of YOLO in agriculture are accelerating rapidly but are fragmented and multidisciplinary. Moreover, the performance characteristics (i.e., accuracy, speed, computation) of the object detector influence the rate of technology implementation and adoption in agriculture. Thus, the study aims to collect extensive literature to document and critically evaluate the advances and application of YOLO for agricultural object recognition. First, we conducted a bibliometric review of 257 articles to understand the scholarly landscape of YOLO in agricultural domain. Secondly, we conducted a systematic review of 30 articles to identify current knowledge, gaps, and modifications in YOLO for specific agricultural tasks. The study critically assesses and summarizes the information on YOLO's end-to-end learning approach, including data acquisition, processing, network modification, integration, and deployment. We also discussed task-specific YOLO algorithm modification and integration to meet the agricultural object or environment-specific challenges. In general, YOLO-integrated digital tools and technologies show the potential for real-time, automated monitoring, surveillance, and object handling to reduce labor, production cost, and environmental impact while maximizing resource efficiency. The study provides detailed documentation and significantly advances the existing knowledge on applying YOLO in agriculture, which can greatly benefit the scientific community.
Authors: Dewant Katare, David Solans Noguero, Souneil Park, Nicolas Kourtellis, Marijn Janssen, Aaron Yi Ding
The accuracy and fairness of perception systems in autonomous driving are crucial, particularly for vulnerable road users. Mainstream research has looked into improving the performance metrics for classification accuracy. However, the hidden traits of bias inheritance in the AI models, class imbalances and disparities in the datasets are often overlooked. In this context, our study examines the class imbalances for vulnerable road users by focusing on class distribution analysis, performance evaluation, and bias impact assessment. We identify the concern of imbalances in class representation, leading to potential biases in detection accuracy. Utilizing popular CNN models and Vision Transformers (ViTs) with the nuScenes dataset, our performance evaluation reveals detection disparities for underrepresented classes. We propose a methodology for model optimization and bias mitigation, which includes data augmentation, resampling, and metric-specific learning. Using the proposed mitigation approaches, we see improvement in IoU(%) and NDS(%) metrics from 71.3 to 75.6 and 80.6 to 83.7 respectively, for the CNN model. Similarly, for ViT, we observe improvement in IoU and NDS metrics from 74.9 to 79.2 and 83.8 to 87.1 respectively. This research contributes to developing more reliable models and datasets, enhancing inclusiveness for minority classes.
Authors: Yongchen Zhou, Richard Jiang
In the domain of computer vision, the restoration of missing information in video frames is a critical challenge, particularly in applications such as autonomous driving and surveillance systems. This paper introduces the Siamese Masked Conditional Variational Autoencoder (SiamMCVAE), leveraging a siamese architecture with twin encoders based on vision transformers. This innovative design enhances the model's ability to comprehend lost content by capturing intrinsic similarities between paired frames. SiamMCVAE proficiently reconstructs missing elements in masked frames, effectively addressing issues arising from camera malfunctions through variational inferences. Experimental results robustly demonstrate the model's effectiveness in restoring missing information, thus enhancing the resilience of computer vision systems. The incorporation of Siamese Vision Transformer (SiamViT) encoders in SiamMCVAE exemplifies promising potential for addressing real-world challenges in computer vision, reinforcing the adaptability of autonomous systems in dynamic environments.
Authors: Xin Yuan, Jinoo Baek, Keyang Xu, Omer Tov, Hongliang Fei
We propose an efficient diffusion-based text-to-video super-resolution (SR) tuning approach that leverages the readily learned capacity of pixel level image diffusion model to capture spatial information for video generation. To accomplish this goal, we design an efficient architecture by inflating the weightings of the text-to-image SR model into our video generation framework. Additionally, we incorporate a temporal adapter to ensure temporal coherence across video frames. We investigate different tuning approaches based on our inflated architecture and report trade-offs between computational costs and super-resolution quality. Empirical evaluation, both quantitative and qualitative, on the Shutterstock video dataset, demonstrates that our approach is able to perform text-to-video SR generation with good visual quality and temporal consistency. To evaluate temporal coherence, we also present visualizations in video format in https://drive.google.com/drive/folders/1YVc-KMSJqOrEUdQWVaI-Yfu8Vsfu_1aO?usp=sharing .
Authors: Jinli Duan
DataViz3D is an innovative online software that transforms complex datasets into interactive 3D spatial models using holographic technology. This tool enables users to generate scatter plot within a 3D space, accurately mapped to the XYZ coordinates of the dataset, providing a vivid and intuitive understanding of the spatial relationships inherent in the data. DataViz3D's user friendly interface makes advanced 3D modeling and holographic visualization accessible to a wide range of users, fostering new opportunities for collaborative research and education across various disciplines.
Authors: Juwita juwita, Ghulam Mubashar Hassan, Naveed Akhtar, Amitava Datta
Segmenting organs in CT scan images is a necessary process for multiple downstream medical image analysis tasks. Currently, manual CT scan segmentation by radiologists is prevalent, especially for organs like the pancreas, which requires a high level of domain expertise for reliable segmentation due to factors like small organ size, occlusion, and varying shapes. When resorting to automated pancreas segmentation, these factors translate to limited reliable labeled data to train effective segmentation models. Consequently, the performance of contemporary pancreas segmentation models is still not within acceptable ranges. To improve that, we propose M3BUNet, a fusion of MobileNet and U-Net neural networks, equipped with a novel Mean-Max (MM) attention that operates in two stages to gradually segment pancreas CT images from coarse to fine with mask guidance for object detection. This approach empowers the network to surpass segmentation performance achieved by similar network architectures and achieve results that are on par with complex state-of-the-art methods, all while maintaining a low parameter count. Additionally, we introduce external contour segmentation as a preprocessing step for the coarse stage to assist in the segmentation process through image standardization. For the fine segmentation stage, we found that applying a wavelet decomposition filter to create multi-input images enhances pancreas segmentation performance. We extensively evaluate our approach on the widely known NIH pancreas dataset and MSD pancreas dataset. Our approach demonstrates a considerable performance improvement, achieving an average Dice Similarity Coefficient (DSC) value of up to 89.53% and an Intersection Over Union (IOU) score of up to 81.16 for the NIH pancreas dataset, and 88.60% DSC and 79.90% IOU for the MSD Pancreas dataset.
Authors: Borui Zhang, Wenzhao Zheng, Jie Zhou, Jiwen Lu
Rigorousness and clarity are both essential for interpretations of DNNs to engender human trust. Path methods are commonly employed to generate rigorous attributions that satisfy three axioms. However, the meaning of attributions remains ambiguous due to distinct path choices. To address the ambiguity, we introduce \textbf{Concentration Principle}, which centrally allocates high attributions to indispensable features, thereby endowing aesthetic and sparsity. We then present \textbf{SAMP}, a model-agnostic interpreter, which efficiently searches the near-optimal path from a pre-defined set of manipulation paths. Moreover, we propose the infinitesimal constraint (IC) and momentum strategy (MS) to improve the rigorousness and optimality. Visualizations show that SAMP can precisely reveal DNNs by pinpointing salient image pixels. We also perform quantitative experiments and observe that our method significantly outperforms the counterparts. Code: https://github.com/zbr17/SAMP.
Authors: Liwen Hu, Ziluo Ding, Mianzhi Liu, Lei Ma, Tiejun Huang
As a neuromorphic sensor with high temporal resolution, spike camera can generate continuous binary spike streams to capture per-pixel light intensity. We can use reconstruction methods to restore scene details in high-speed scenarios. However, due to limited information in spike streams, low-light scenes are difficult to effectively reconstruct. In this paper, we propose a bidirectional recurrent-based reconstruction framework, including a Light-Robust Representation (LR-Rep) and a fusion module, to better handle such extreme conditions. LR-Rep is designed to aggregate temporal information in spike streams, and a fusion module is utilized to extract temporal features. Additionally, we have developed a reconstruction benchmark for high-speed low-light scenes. Light sources in the scenes are carefully aligned to real-world conditions. Experimental results demonstrate the superiority of our method, which also generalizes well to real spike streams. Related codes and proposed datasets will be released after publication.
Authors: Hanxun Huang, Ricardo J. G. B. Campello, Sarah Monazam Erfani, Xingjun Ma, Michael E. Houle, James Bailey
Representations learned via self-supervised learning (SSL) can be susceptible to dimensional collapse, where the learned representation subspace is of extremely low dimensionality and thus fails to represent the full data distribution and modalities. Dimensional collapse also known as the "underfilling" phenomenon is one of the major causes of degraded performance on downstream tasks. Previous work has investigated the dimensional collapse problem of SSL at a global level. In this paper, we demonstrate that representations can span over high dimensional space globally, but collapse locally. To address this, we propose a method called $\textit{local dimensionality regularization (LDReg)}$. Our formulation is based on the derivation of the Fisher-Rao metric to compare and optimize local distance distributions at an asymptotically small radius for each data point. By increasing the local intrinsic dimensionality, we demonstrate through a range of experiments that LDReg improves the representation quality of SSL. The results also show that LDReg can regularize dimensionality at both local and global levels.
Authors: Xiangshuo Qiao, Xianxin Li, Xiaozhe Qu, Jie Zhang, Yang Liu, Yu Luo, Cihang Jin, Jin Ma
Vision-Language Models pre-trained on large-scale image-text datasets have shown superior performance in downstream tasks such as image retrieval. Most of the images for pre-training are presented in the form of open domain common-sense visual elements. Differently, video covers in short video search scenarios are presented as user-originated contents that provide important visual summaries of videos. In addition, a portion of the video covers come with manually designed cover texts that provide semantic complements. In order to fill in the gaps in short video cover data, we establish the first large-scale cover-text benchmark for Chinese short video search scenarios. Specifically, we release two large-scale datasets CBVS-5M/10M to provide short video covers, and the manual fine-labeling dataset CBVS-20K to provide real user queries, which serves as an image-text benchmark test in the Chinese short video search field. To integrate the semantics of cover text in the case of modality missing, we propose UniCLIP where cover texts play a guiding role during training, however are not relied upon by inference. Extensive evaluation on CBVS-20K demonstrates the excellent performance of our proposal. UniCLIP has been deployed to Tencent's online video search systems with hundreds of millions of visits and achieved significant gains. The complete dataset, code and checkpoints will be available upon release.
Authors: Mingjian Li, Mingyuan Meng, Michael Fulham, David Dagan Feng, Lei Bi, Jinman Kim
Medical image representations can be learned through medical vision-language contrastive learning (mVLCL) where medical imaging reports are used as weak supervision through image-text alignment. These learned image representations can be transferred to and benefit various downstream medical vision tasks such as disease classification and segmentation. Recent mVLCL methods attempt to align image sub-regions and the report keywords as local-matchings. However, these methods aggregate all local-matchings via simple pooling operations while ignoring the inherent relations between them. These methods therefore fail to reason between local-matchings that are semantically related, e.g., local-matchings that correspond to the disease word and the location word (semantic-relations), and also fail to differentiate such clinically important local-matchings from others that correspond to less meaningful words, e.g., conjunction words (importance-relations). Hence, we propose a mVLCL method that models the inter-matching relations between local-matchings via a relation-enhanced contrastive learning framework (RECLF). In RECLF, we introduce a semantic-relation reasoning module (SRM) and an importance-relation reasoning module (IRM) to enable more fine-grained report supervision for image representation learning. We evaluated our method using four public benchmark datasets on four downstream tasks, including segmentation, zero-shot classification, supervised classification, and cross-modal retrieval. Our results demonstrated the superiority of our RECLF over the state-of-the-art mVLCL methods with consistent improvements across single-modal and cross-modal tasks. These results suggest that our RECLF, by modelling the inter-matching relations, can learn improved medical image representations with better generalization capabilities.
Authors: Zewen Chen, Juan Wang, Bing Li, Chunfeng Yuan, Weiming Hu, Junxian Liu, Peng Li, Yan Wang, Youqun Zhang, Congxuan Zhang
Due to the subjective nature of image quality assessment (IQA), assessing which image has better quality among a sequence of images is more reliable than assigning an absolute mean opinion score for an image. Thus, IQA models are evaluated by global correlation consistency (GCC) metrics like PLCC and SROCC, rather than mean opinion consistency (MOC) metrics like MAE and MSE. However, most existing methods adopt MOC metrics to define their loss functions, due to the infeasible computation of GCC metrics during training. In this work, we construct a novel loss function and network to exploit Global-correlation and Mean-opinion Consistency, forming a GMC-IQA framework. Specifically, we propose a novel GCC loss by defining a pairwise preference-based rank estimation to solve the non-differentiable problem of SROCC and introducing a queue mechanism to reserve previous data to approximate the global results of the whole data. Moreover, we propose a mean-opinion network, which integrates diverse opinion features to alleviate the randomness of weight learning and enhance the model robustness. Experiments indicate that our method outperforms SOTA methods on multiple authentic datasets with higher accuracy and generalization. We also adapt the proposed loss to various networks, which brings better performance and more stable training.
Authors: Yunpeng Gong, Jiaquan Li, Lifei Chen, Min Jiang
In the field of computer vision, the persistent presence of color bias, resulting from fluctuations in real-world lighting and camera conditions, presents a substantial challenge to the robustness of models. This issue is particularly pronounced in complex wide-area surveillance scenarios, such as person re-identification and industrial dust segmentation, where models often experience a decline in performance due to overfitting on color information during training, given the presence of environmental variations. Consequently, there is a need to effectively adapt models to cope with the complexities of camera conditions. To address this challenge, this study introduces a learning strategy named Random Color Erasing, which draws inspiration from ensemble learning. This strategy selectively erases partial or complete color information in the training data without disrupting the original image structure, thereby achieving a balanced weighting of color features and other features within the neural network. This approach mitigates the risk of overfitting and enhances the model's ability to handle color variation, thereby improving its overall robustness. The approach we propose serves as an ensemble learning strategy, characterized by robust interpretability. A comprehensive analysis of this methodology is presented in this paper. Across various tasks such as person re-identification and semantic segmentation, our approach consistently improves strong baseline methods. Notably, in comparison to existing methods that prioritize color robustness, our strategy significantly enhances performance in cross-domain scenarios. The code available at \url{https://github.com/layumi/Person\_reID\_baseline\_pytorch/blob/master/random\_erasing.py} or \url{https://github.com/finger-monkey/Data-Augmentation}.
Authors: Hao Zhang, Shuaijie Zhang
Bounding box regression plays a crucial role in the field of object detection, and the positioning accuracy of object detection largely depends on the loss function of bounding box regression. Existing researchs improve regression performance by utilizing the geometric relationship between bounding boxes, while ignoring the impact of difficult and easy sample distribution on bounding box regression. In this article, we analyzed the impact of difficult and easy sample distribution on regression results, and then proposed Focaler-IoU, which can improve detector performance in different detection tasks by focusing on different regression samples. Finally, comparative experiments were conducted using existing advanced detectors and regression methods for different detection tasks, and the detection performance was further improved by using the method proposed in this paper.Code is available at \url{https://github.com/malagoutou/Focaler-IoU}.
Authors: Yeongtak Oh, Saehyung Lee, Uiwon Hwang, Sungroh Yoon
Large-scale language-vision pre-training models, such as CLIP, have achieved remarkable text-guided image morphing results by leveraging several unconditional generative models. However, existing CLIP-guided image morphing methods encounter difficulties when morphing photorealistic images. Specifically, existing guidance fails to provide detailed explanations of the morphing regions within the image, leading to misguidance. In this paper, we observed that such misguidance could be effectively mitigated by simply using a proper regularization loss. Our approach comprises two key components: 1) a geodesic cosine similarity loss that minimizes inter-modality features (i.e., image and text) on a projected subspace of CLIP space, and 2) a latent regularization loss that minimizes intra-modality features (i.e., image and image) on the image manifold. By replacing the na\"ive directional CLIP loss in a drop-in replacement manner, our method achieves superior morphing results on both images and videos for various benchmarks, including CLIP-inversion.
Authors: Xiyao Wang, Yuhang Zhou, Xiaoyu Liu, Hongjin Lu, Yuancheng Xu, Feihong He, Jaehong Yoon, Taixi Lu, Gedas Bertasius, Mohit Bansal, Huaxiu Yao, Furong Huang
Multimodal Large Language Models (MLLMs) have demonstrated proficiency in handling a variety of visual-language tasks. However, current MLLM benchmarks are predominantly designed to evaluate reasoning based on static information about a single image, and the ability of modern MLLMs to extrapolate from image sequences, which is essential for understanding our ever-changing world, has been less investigated. To address this challenge, this paper introduces Mementos, a new benchmark designed to assess MLLMs' sequential image reasoning abilities. Mementos features 4,761 diverse image sequences with varying lengths. We also employ a GPT-4 assisted method to evaluate MLLM reasoning performance. Through a careful evaluation of nine recent MLLMs on Mementos, including GPT-4V and Gemini, we find that they struggle to accurately describe dynamic information about given image sequences, often leading to hallucinations/misrepresentations of objects and their corresponding behaviors. Our quantitative analysis and case studies identify three key factors impacting MLLMs' sequential image reasoning: the correlation between object and behavioral hallucinations, the influence of cooccurring behaviors, and the compounding impact of behavioral hallucinations. Our dataset is available at https://github.com/umd-huang-lab/Mementos.
Authors: Junyu Gao, Liangliang Zhao, Xuelong Li
Object counting is a hot topic in computer vision, which aims to estimate the number of objects in a given image. However, most methods only count objects of a single category for an image, which cannot be applied to scenes that need to count objects with multiple categories simultaneously, especially in aerial scenes. To this end, this paper introduces a Multi-category Object Counting (MOC) task to estimate the numbers of different objects (cars, buildings, ships, etc.) in an aerial image. Considering the absence of a dataset for this task, a large-scale Dataset (NWPU-MOC) is collected, consisting of 3,416 scenes with a resolution of 1024 $\times$ 1024 pixels, and well-annotated using 14 fine-grained object categories. Besides, each scene contains RGB and Near Infrared (NIR) images, of which the NIR spectrum can provide richer characterization information compared with only the RGB spectrum. Based on NWPU-MOC, the paper presents a multi-spectrum, multi-category object counting framework, which employs a dual-attention module to fuse the features of RGB and NIR and subsequently regress multi-channel density maps corresponding to each object category. In addition, to modeling the dependency between different channels in the density map with each object category, a spatial contrast loss is designed as a penalty for overlapping predictions at the same spatial position. Experimental results demonstrate that the proposed method achieves state-of-the-art performance compared with some mainstream counting algorithms. The dataset, code and models are publicly available at https://github.com/lyongo/NWPU-MOC.
Authors: Bo Zhao, Huan Yang, Jianlong Fu
Face inpainting requires the model to have a precise global understanding of the facial position structure. Benefiting from the powerful capabilities of deep learning backbones, recent works in face inpainting have achieved decent performance in ideal setting (square shape with $512px$). However, existing methods often produce a visually unpleasant result, especially in the position-sensitive details (e.g., eyes and nose), when directly applied to arbitrary-shaped images in real-world scenarios. The visually unpleasant position-sensitive details indicate the shortcomings of existing methods in terms of position information processing capability. In this paper, we propose an \textbf{I}mplicit \textbf{N}eural \textbf{I}npainting \textbf{N}etwork (IN$^2$) to handle arbitrary-shape face images in real-world scenarios by explicit modeling for position information. Specifically, a downsample processing encoder is proposed to reduce information loss while obtaining the global semantic feature. A neighbor hybrid attention block is proposed with a hybrid attention mechanism to improve the facial understanding ability of the model without restricting the shape of the input. Finally, an implicit neural pyramid decoder is introduced to explicitly model position information and bridge the gap between low-resolution features and high-resolution output. Extensive experiments demonstrate the superiority of the proposed method in real-world face inpainting task.
Authors: Divya Jyoti Bajpai, Aastha Jaiswal, Manjesh Kumar Hanawal
The recent advances in Deep Neural Networks (DNNs) stem from their exceptional performance across various domains. However, their inherent large size hinders deploying these networks on resource-constrained devices like edge, mobile, and IoT platforms. Strategies have emerged, from partial cloud computation offloading (split computing) to integrating early exits within DNN layers. Our work presents an innovative unified approach merging early exits and split computing. We determine the 'splitting layer', the optimal depth in the DNN for edge device computations, and whether to infer on edge device or be offloaded to the cloud for inference considering accuracy, computational efficiency, and communication costs. Also, Image classification faces diverse environmental distortions, influenced by factors like time of day, lighting, and weather. To adapt to these distortions, we introduce I-SplitEE, an online unsupervised algorithm ideal for scenarios lacking ground truths and with sequential data. Experimental validation using Caltech-256 and Cifar-10 datasets subjected to varied distortions showcases I-SplitEE's ability to reduce costs by a minimum of 55% with marginal performance degradation of at most 5%.
Authors: Wenlong Liu, Tianyu Yang, Yuhan Wang, Qizhi Yu, Lei Zhang
This work studies the problem of panoptic symbol spotting, which is to spot and parse both countable object instances (windows, doors, tables, etc.) and uncountable stuff (wall, railing, etc.) from computer-aided design (CAD) drawings. Existing methods typically involve either rasterizing the vector graphics into images and using image-based methods for symbol spotting, or directly building graphs and using graph neural networks for symbol recognition. In this paper, we take a different approach, which treats graphic primitives as a set of 2D points that are locally connected and use point cloud segmentation methods to tackle it. Specifically, we utilize a point transformer to extract the primitive features and append a mask2former-like spotting head to predict the final output. To better use the local connection information of primitives and enhance their discriminability, we further propose the attention with connection module (ACM) and contrastive connection learning scheme (CCL). Finally, we propose a KNN interpolation mechanism for the mask attention module of the spotting head to better handle primitive mask downsampling, which is primitive-level in contrast to pixel-level for the image. Our approach, named SymPoint, is simple yet effective, outperforming recent state-of-the-art method GAT-CADNet by an absolute increase of 9.6% PQ and 10.4% RQ on the FloorPlanCAD dataset. The source code and models will be available at https://github.com/nicehuster/SymPoint.
Authors: Yichen Chen, Yiqi Pan, Ruyu Liu, Haoyu Zhang, Guodao Zhang, Bo Sun, Jianhua Zhang
To enhance the performance and effect of AR/VR applications and visual assistance and inspection systems, visual simultaneous localization and mapping (vSLAM) is a fundamental task in computer vision and robotics. However, traditional vSLAM systems are limited by the camera's narrow field-of-view, resulting in challenges such as sparse feature distribution and lack of dense depth information. To overcome these limitations, this paper proposes a 360ORB-SLAM system for panoramic images that combines with a depth completion network. The system extracts feature points from the panoramic image, utilizes a panoramic triangulation module to generate sparse depth information, and employs a depth completion network to obtain a dense panoramic depth map. Experimental results on our novel panoramic dataset constructed based on Carla demonstrate that the proposed method achieves superior scale accuracy compared to existing monocular SLAM methods and effectively addresses the challenges of feature association and scale ambiguity. The integration of the depth completion network enhances system stability and mitigates the impact of dynamic elements on SLAM performance.
Authors: Rui Xu, Yunke Wang, Bo Du
Unsupervised anomaly detection has gained significant attention in the field of medical imaging due to its capability of relieving the costly pixel-level annotation. To achieve this, modern approaches usually utilize generative models to produce healthy references of the diseased images and then identify the abnormalities by comparing the healthy references and the original diseased images. Recently, diffusion models have exhibited promising potential for unsupervised anomaly detection in medical images for their good mode coverage and high sample quality. However, the intrinsic characteristics of the medical images, e.g. the low contrast, and the intricate anatomical structure of the human body make the reconstruction challenging. Besides, the global information of medical images often remain underutilized. To address these two issues, we propose a novel Masked Autoencoder-enhanced Diffusion Model (MAEDiff) for unsupervised anomaly detection in brain images. The MAEDiff involves a hierarchical patch partition. It generates healthy images by overlapping upper-level patches and implements a mechanism based on the masked autoencoders operating on the sub-level patches to enhance the condition on the unnoised regions. Extensive experiments on data of tumors and multiple sclerosis lesions demonstrate the effectiveness of our method.
Authors: Hao Ai, Zidong Cao, Haonan Lu, Chen Chen, Jian Ma, Pengyuan Zhou, Tae-Kyun Kim, Pan Hui, Lin Wang
360 images, with a field-of-view (FoV) of 180x360, provide immersive and realistic environments for emerging virtual reality (VR) applications, such as virtual tourism, where users desire to create diverse panoramic scenes from a narrow FoV photo they take from a viewpoint via portable devices. It thus brings us to a technical challenge: `How to allow the users to freely create diverse and immersive virtual scenes from a narrow FoV image with a specified viewport?' To this end, we propose a transformer-based 360 image outpainting framework called Dream360, which can generate diverse, high-fidelity, and high-resolution panoramas from user-selected viewports, considering the spherical properties of 360 images. Compared with existing methods, e.g., [3], which primarily focus on inputs with rectangular masks and central locations while overlooking the spherical property of 360 images, our Dream360 offers higher outpainting flexibility and fidelity based on the spherical representation. Dream360 comprises two key learning stages: (I) codebook-based panorama outpainting via Spherical-VQGAN (S-VQGAN), and (II) frequency-aware refinement with a novel frequency-aware consistency loss. Specifically, S-VQGAN learns a sphere-specific codebook from spherical harmonic (SH) values, providing a better representation of spherical data distribution for scene modeling. The frequency-aware refinement matches the resolution and further improves the semantic consistency and visual fidelity of the generated results. Our Dream360 achieves significantly lower Frechet Inception Distance (FID) scores and better visual fidelity than existing methods. We also conducted a user study involving 15 participants to interactively evaluate the quality of the generated results in VR, demonstrating the flexibility and superiority of our Dream360 framework.
Authors: Lintai Wu, Junhui Hou, Linqi Song, Yong Xu
3D shapes captured by scanning devices are often incomplete due to occlusion. 3D shape completion methods have been explored to tackle this limitation. However, most of these methods are only trained and tested on a subset of categories, resulting in poor generalization to unseen categories. In this paper, we introduce a novel weakly-supervised framework to reconstruct the complete shapes from unseen categories. We first propose an end-to-end prior-assisted shape learning network that leverages data from the seen categories to infer a coarse shape. Specifically, we construct a prior bank consisting of representative shapes from the seen categories. Then, we design a multi-scale pattern correlation module for learning the complete shape of the input by analyzing the correlation between local patterns within the input and the priors at various scales. In addition, we propose a self-supervised shape refinement model to further refine the coarse shape. Considering the shape variability of 3D objects across categories, we construct a category-specific prior bank to facilitate shape refinement. Then, we devise a voxel-based partial matching loss and leverage the partial scans to drive the refinement process. Extensive experimental results show that our approach is superior to state-of-the-art methods by a large margin.
Authors: Xiangpeng Yang, Linchao Zhu, Xiaohan Wang, Yi Yang
Text-video retrieval is a critical multi-modal task to find the most relevant video for a text query. Although pretrained models like CLIP have demonstrated impressive potential in this area, the rising cost of fully finetuning these models due to increasing model size continues to pose a problem. To address this challenge, prompt tuning has emerged as an alternative. However, existing works still face two problems when adapting pretrained image-text models to downstream video-text tasks: (1) The visual encoder could only encode frame-level features and failed to extract global-level general video information. (2) Equipping the visual and text encoder with separated prompts failed to mitigate the visual-text modality gap. To this end, we propose DGL, a cross-modal Dynamic prompt tuning method with Global-Local video attention. In contrast to previous prompt tuning methods, we employ the shared latent space to generate local-level text and frame prompts that encourage inter-modal interaction. Furthermore, we propose modeling video in a global-local attention mechanism to capture global video information from the perspective of prompt tuning. Extensive experiments reveal that when only 0.67% parameters are tuned, our cross-modal prompt tuning strategy DGL outperforms or is comparable to fully finetuning methods on MSR-VTT, VATEX, LSMDC, and ActivityNet datasets. Code will be available at https://github.com/knightyxp/DGL
Authors: Hongyi Wang, Xiuju Du, Jing Liu, Shuyi Ouyang, Yen-Wei Chen, Lanfen Lin
The advancement of Spatial Transcriptomics (ST) has facilitated the spatially-aware profiling of gene expressions based on histopathology images. Although ST data offers valuable insights into the micro-environment of tumors, its acquisition cost remains expensive. Therefore, directly predicting the ST expressions from digital pathology images is desired. Current methods usually adopt existing regression backbones for this task, which ignore the inherent multi-scale hierarchical data structure of digital pathology images. To address this limit, we propose M2ORT, a many-to-one regression Transformer that can accommodate the hierarchical structure of the pathology images through a decoupled multi-scale feature extractor. Different from traditional models that are trained with one-to-one image-label pairs, M2ORT accepts multiple pathology images of different magnifications at a time to jointly predict the gene expressions at their corresponding common ST spot, aiming at learning a many-to-one relationship through training. We have tested M2ORT on three public ST datasets and the experimental results show that M2ORT can achieve state-of-the-art performance with fewer parameters and floating-point operations (FLOPs). The code is available at: https://github.com/Dootmaan/M2ORT/.
Authors: Jan Heiland, Yongho Kim
With the advancement of neural networks, there has been a notable increase, both in terms of quantity and variety, in research publications concerning the application of autoencoders to reduced-order models. We propose a polytopic autoencoder architecture that includes a lightweight nonlinear encoder, a convex combination decoder, and a smooth clustering network. Supported by several proofs, the model architecture ensures that all reconstructed states lie within a polytope, accompanied by a metric indicating the quality of the constructed polytopes, referred to as polytope error. Additionally, it offers a minimal number of convex coordinates for polytopic linear-parameter varying systems while achieving acceptable reconstruction errors compared to proper orthogonal decomposition (POD). To validate our proposed model, we conduct simulations involving two flow scenarios with the incompressible Navier-Stokes equation. Numerical results demonstrate the guaranteed properties of the model, low reconstruction errors compared to POD, and the improvement in error using a clustering network.
Authors: Cosmin I. Bercea, Benedikt Wiestler, Daniel Rueckert, Julia A. Schnabel
The increasing complexity of medical imaging data underscores the need for advanced anomaly detection methods to automatically identify diverse pathologies. Current methods face challenges in capturing the broad spectrum of anomalies, often limiting their use to specific lesion types in brain scans. To address this challenge, we introduce a novel unsupervised approach, termed \textit{Reversed Auto-Encoders (RA)}, designed to create realistic pseudo-healthy reconstructions that enable the detection of a wider range of pathologies. We evaluate the proposed method across various imaging modalities, including magnetic resonance imaging (MRI) of the brain, pediatric wrist X-ray, and chest X-ray, and demonstrate superior performance in detecting anomalies compared to existing state-of-the-art methods. Our unsupervised anomaly detection approach may enhance diagnostic accuracy in medical imaging by identifying a broader range of unknown pathologies. Our code is publicly available at: \url{https://github.com/ci-ber/RA}.
Authors: Miquel Miró-Nicolau, Antoni Jaume-i-Capó, Gabriel Moyà-Alcover
The use of eXplainable Artificial Intelligence (XAI) systems has introduced a set of challenges that need resolution. Herein, we focus on how to correctly select an XAI method, an open questions within the field. The inherent difficulty of this task is due to the lack of a ground truth. Several authors have proposed metrics to approximate the fidelity of different XAI methods. These metrics lack verification and have concerning disagreements. In this study, we proposed a novel methodology to verify fidelity metrics, using a well-known transparent model, namely a decision tree. This model allowed us to obtain explanations with perfect fidelity. Our proposal constitutes the first objective benchmark for these metrics, facilitating a comparison of existing proposals, and surpassing existing methods. We applied our benchmark to assess the existing fidelity metrics in two different experiments, each using public datasets comprising 52,000 images. The images from these datasets had a size a 128 by 128 pixels and were synthetic data that simplified the training process. All metric values, indicated a lack of fidelity, with the best one showing a 30 \% deviation from the expected values for perfect explanation. Our experimentation led us to conclude that the current fidelity metrics are not reliable enough to be used in real scenarios. From this finding, we deemed it necessary to development new metrics, to avoid the detected problems, and we recommend the usage of our proposal as a benchmark within the scientific community to address these limitations.
Authors: Ali Amiri, Aydin Kaya, Ali Seydi Keceli
Vehicle re-identification (ReID) endeavors to associate vehicle images collected from a distributed network of cameras spanning diverse traffic environments. This task assumes paramount importance within the spectrum of vehicle-centric technologies, playing a pivotal role in deploying Intelligent Transportation Systems (ITS) and advancing smart city initiatives. Rapid advancements in deep learning have significantly propelled the evolution of vehicle ReID technologies in recent years. Consequently, undertaking a comprehensive survey of methodologies centered on deep learning for vehicle re-identification has become imperative and inescapable. This paper extensively explores deep learning techniques applied to vehicle ReID. It outlines the categorization of these methods, encompassing supervised and unsupervised approaches, delves into existing research within these categories, introduces datasets and evaluation criteria, and delineates forthcoming challenges and potential research directions. This comprehensive assessment examines the landscape of deep learning in vehicle ReID and establishes a foundation and starting point for future works. It aims to serve as a complete reference by highlighting challenges and emerging trends, fostering advancements and applications in vehicle ReID utilizing deep learning models.
Authors: Mirza Nihal Baig, Rony Hajong, Mahdi Murshed Patwary, Mohammad Shahidur Rahman, Husne Ara Chowdhury
We propose a comprehensive dataset for object detection in diverse driving environments across 9 districts in Bangladesh. The dataset, collected exclusively from smartphone cameras, provided a realistic representation of real-world scenarios, including day and night conditions. Most existing datasets lack suitable classes for autonomous navigation on Bangladeshi roads, making it challenging for researchers to develop models that can handle the intricacies of road scenarios. To address this issue, the authors proposed a new set of classes based on characteristics rather than local vehicle names. The dataset aims to encourage the development of models that can handle the unique challenges of Bangladeshi road scenarios for the effective deployment of autonomous vehicles. The dataset did not consist of any online images to simulate real-world conditions faced by autonomous vehicles. The classification of vehicles is challenging because of the diverse range of vehicles on Bangladeshi roads, including those not found elsewhere in the world. The proposed classification system is scalable and can accommodate future vehicles, making it a valuable resource for researchers in the autonomous vehicle sector.
Authors: Chen Wu, Zhuoran Zheng, Xiuyi Jia, Wenqi Ren
With the continuous advancement of imaging devices, the prevalence of Ultra-High-Definition (UHD) images is rising. Although many image restoration methods have achieved promising results, they are not directly applicable to UHD images on devices with limited computational resources due to the inherently high computational complexity of UHD images. In this paper, we focus on the task of low-light image enhancement (LLIE) and propose a novel LLIE method called MixNet, which is designed explicitly for UHD images. To capture the long-range dependency of features without introducing excessive computational complexity, we present the Global Feature Modulation Layer (GFML). GFML associates features from different views by permuting the feature maps, enabling efficient modeling of long-range dependency. In addition, we also design the Local Feature Modulation Layer (LFML) and Feed-forward Layer (FFL) to capture local features and transform features into a compact representation. This way, our MixNet achieves effective LLIE with few model parameters and low computational complexity. We conducted extensive experiments on both synthetic and real-world datasets, and the comprehensive results demonstrate that our proposed method surpasses the performance of current state-of-the-art methods. The code will be available at \url{https://github.com/zzr-idam/MixNet}.
Authors: Guido Caccianiga, Julian Nubert, Marco Hutter, Katherine J. Kuchenbecker
New sensing technologies and more advanced processing algorithms are transforming computer-integrated surgery. While researchers are actively investigating depth sensing and 3D reconstruction for vision-based surgical assistance, it remains difficult to achieve real-time, accurate, and robust 3D representations of the abdominal cavity for minimally invasive surgery. Thus, this work uses quantitative testing on fresh ex-vivo porcine tissue to thoroughly characterize the quality with which a 3D laser-based time-of-flight sensor (lidar) can perform anatomical surface reconstruction. Ground-truth surface shapes are captured with a commercial laser scanner, and the resulting signed error fields are analyzed using rigorous statistical tools. When compared to modern learning-based stereo matching from endoscopic images, time-of-flight sensing demonstrates higher precision, lower processing delay, higher frame rate, and superior robustness against sensor distance and poor illumination. Furthermore, we report on the potential negative effect of near-infrared light penetration on the accuracy of lidar measurements across different tissue samples, identifying a significant measured depth offset for muscle in contrast to fat and liver. Our findings highlight the potential of lidar for intraoperative 3D perception and point toward new methods that combine complementary time-of-flight and spectral imaging.
Authors: Haibo Wang, Chenghang Lai, Yixuan Sun, Weifeng Ge
Video Question Answering (VideoQA) aims to answer natural language questions based on the information observed in videos. Despite the recent success of Large Multimodal Models (LMMs) in image-language understanding and reasoning, they deal with VideoQA insufficiently by simply taking uniformly sampled frames as visual inputs, which ignores question-relevant visual clues. Moreover, there are no human annotations for question-critical timestamps in existing VideoQA datasets. In light of this, we propose a novel weakly supervised framework to enforce the LMMs to reason out the answers with question-critical moments as visual inputs. Specifically, we fuse the question and answer pairs as event descriptions to find multiple keyframes as target moments, which will be pseudo-labels. With these pseudo-labels as additionally weak supervision, we devise a lightweight Gaussian-based Contrastive Grounding (GCG) module. GCG learns multiple Gaussian functions to characterize the temporal structure of the video, and sample question-critical frames as positive moments to be the visual inputs of LMMs. Extensive experiments on several VideoQA benchmarks verify the effectiveness of our framework, and we achieve substantial improvements compared to previous state-of-the-art methods.
Authors: Haibi Wang, Weifeng Ge
With the breakthrough of multi-modal large language models, answering complex visual questions that demand advanced reasoning abilities and world knowledge has become a much more important testbed for developing AI models than ever. However, equipping AI models with robust cross-modality reasoning ability remains challenging since the cognition scheme of humans has not been understood systematically. In this paper, we believe that if we can collect visual clues in the given image as much as possible, we will recognize the image more accurately, understand the question better, recall relevant knowledge more easily, and finally reason out the answer. We discover these rich visual clues by mining question-answer pairs in images and sending them into multi-modal large language models as prompts. We call the proposed method Q&A Prompts. Specifically, we first use the image-answer pairs and the corresponding questions in the training set as inputs and outputs to train a visual question generation model. Then, we use an image tagging model to identify various instances and send packaged image-tag pairs into the visual question generation model to generate relevant questions with the extracted image tags as answers. Finally, we encode these generated question-answer pairs as prompts with a visual-aware prompting module and send them into pre-trained multi-modal large language models to reason out the final answers. Experimental results show that, compared with state-of-the-art methods, our Q&A Prompts achieves substantial improvements on the challenging visual question answering datasets requiring reasoning over diverse world knowledge, such as OK-VQA and A-OKVQA.
Authors: Chenyu Wang, Weixin Luo, Qianyu Chen, Haonan Mai, Jindi Guo, Sixun Dong, Xiaohua (Michael) Xuan, Zhengxin Li, Lin Ma, Shenghua Gao
Recently, the astonishing performance of large language models (LLMs) in natural language comprehension and generation tasks triggered lots of exploration of using them as central controllers to build agent systems. Multiple studies focus on bridging the LLMs to external tools to extend the application scenarios. However, the current LLMs' perceiving tool-use ability is limited to a single text query, which may result in ambiguity in understanding the users' real intentions. LLMs are expected to eliminate that by perceiving the visual- or auditory-grounded instructions' information. Therefore, in this paper, we propose Tool-LMM, a system incorporating open-source LLMs and multi-modal encoders so that the learnt LLMs can be conscious of multi-modal input instruction and then select the function-matched tool correctly. To facilitate the evaluation of the model's capability, we collect a dataset featured by consisting of multi-modal input tools from HuggingFace. Another important feature of our dataset is that our dataset also contains multiple potential choices for the same instruction due to the existence of identical functions and synonymous functions, which provides more potential solutions for the same query. The experiments reveal that our LMM is capable of recommending appropriate tools for multi-modal instructions. Codes and data are available at https://github.com/Tool-LMM/Tool-LMM.
Authors: Tianyi Zhao, Maoxun Yuan, Xingxing Wei
Object detection in visible (RGB) and infrared (IR) images has been widely applied in recent years. Leveraging the complementary characteristics of RGB and IR images, the object detector provides reliable and robust object localization from day to night. Existing fusion strategies directly inject RGB and IR images into convolution neural networks, leading to inferior detection performance. Since the RGB and IR features have modality-specific noise, these strategies will worsen the fused features along with the propagation. Inspired by the mechanism of human brain processing multimodal information, this work introduces a new coarse-to-fine perspective to purify and fuse two modality features. Specifically, following this perspective, we design a Redundant Spectrum Removal module to coarsely remove interfering information within each modality and a Dynamic Feature Selection module to finely select the desired features for feature fusion. To verify the effectiveness of the coarse-to-fine fusion strategy, we construct a new object detector called Removal and Selection Detector (RSDet). Extensive experiments on three RGB-IR object detection datasets verify the superior performance of our method.
Authors: Nam Le, Honglei Zhang, Francesco Cricri, Ramin G. Youvalari, Hamed Rezazadegan Tavakoli, Emre Aksu, Miska M. Hannuksela, Esa Rahtu
Image coding for machines (ICM) aims at reducing the bitrate required to represent an image while minimizing the drop in machine vision analysis accuracy. In many use cases, such as surveillance, it is also important that the visual quality is not drastically deteriorated by the compression process. Recent works on using neural network (NN) based ICM codecs have shown significant coding gains against traditional methods; however, the decompressed images, especially at low bitrates, often contain checkerboard artifacts. We propose an effective decoder finetuning scheme based on adversarial training to significantly enhance the visual quality of ICM codecs, while preserving the machine analysis accuracy, without adding extra bitcost or parameters at the inference phase. The results show complete removal of the checkerboard artifacts at the negligible cost of -1.6% relative change in task performance score. In the cases where some amount of artifacts is tolerable, such as when machine consumption is the primary target, this technique can enhance both pixel-fidelity and feature-fidelity scores without losing task performance.
Authors: Théophile Rageau, Laurence Likforman-Sulem, Attilio Fiandrotti, Victoria Eyharabide, Béatrice Caseau, Jean-Claude Cheynet
Seals are small coin-shaped artifacts, mostly made of lead, held with strings to seal letters. This work presents the first attempt towards automatic reading of text on Byzantine seal images.Byzantine seals are generally decorated with iconography on the obverse side and Greek text on the reverse side. Text may include the sender's name, position in the Byzantine aristocracy, and elements of prayers. Both text and iconography are precious literary sources that wait to be exploited electronically, so the development of computerized systems for interpreting seals images is of paramount importance. This work's contribution is hence a deep, two-stages, character reading pipeline for transcribing Byzantine seal images. A first deep convolutional neural network (CNN) detects characters in the seal (character localization). A second convolutional network reads the localized characters (character classification). Finally, a diplomatic transcription of the seal is provided by post-processing the two network outputs. We provide an experimental evaluation of each CNN in isolation and both CNNs in combination. All performances are evaluated by cross-validation. Character localization achieves a mean average precision (mAP@0.5) greater than 0.9. Classification of characters cropped from ground truth bounding boxes achieves Top-1 accuracy greater than 0.92. End-to-end evaluation shows the efficiency of the proposed approach when compared to the SoTA for similar tasks.
Authors: Chao Pang, Xingxing Weng, Jiang Wu, Qiang Wang, Gui-Song Xia
Advanced change detection techniques primarily target image pairs of equal and high quality. However, variations in imaging conditions and platforms frequently lead to image pairs with distinct qualities: one image being high-quality, while the other being low-quality. These disparities in image quality present significant challenges for understanding image pairs semantically and extracting change features, ultimately resulting in a notable decline in performance. To tackle this challenge, we introduce an innovative training strategy grounded in knowledge distillation. The core idea revolves around leveraging task knowledge acquired from high-quality image pairs to guide the model's learning process when dealing with image pairs that exhibit differences in quality. Additionally, we develop a hierarchical correlation distillation approach (involving self-correlation, cross-correlation, and global correlation). This approach compels the student model to replicate the correlations inherent in the teacher model, rather than focusing solely on individual features. This ensures effective knowledge transfer while maintaining the student model's training flexibility.
Authors: Jukka I. Ahonen, Nam Le, Honglei Zhang, Antti Hallapuro, Francesco Cricri, Hamed Rezazadegan Tavakoli, Miska M. Hannuksela, Esa Rahtu
The recent progress in artificial intelligence has led to an ever-increasing usage of images and videos by machine analysis algorithms, mainly neural networks. Nonetheless, compression, storage and transmission of media have traditionally been designed considering human beings as the viewers of the content. Recent research on image and video coding for machine analysis has progressed mainly in two almost orthogonal directions. The first is represented by end-to-end (E2E) learned codecs which, while offering high performance on image coding, are not yet on par with state-of-the-art conventional video codecs and lack interoperability. The second direction considers using the Versatile Video Coding (VVC) standard or any other conventional video codec (CVC) together with pre- and post-processing operations targeting machine analysis. While the CVC-based methods benefit from interoperability and broad hardware and software support, the machine task performance is often lower than the desired level, particularly in low bitrates. This paper proposes a hybrid codec for machines called NN-VVC, which combines the advantages of an E2E-learned image codec and a CVC to achieve high performance in both image and video coding for machines. Our experiments show that the proposed system achieved up to -43.20% and -26.8% Bj{\o}ntegaard Delta rate reduction over VVC for image and video data, respectively, when evaluated on multiple different datasets and machine vision tasks. To the best of our knowledge, this is the first research paper showing a hybrid video codec that outperforms VVC on multiple datasets and multiple machine vision tasks.
Authors: Anton Sergeev, Victor Minchenkov, Aleksei Soldatov
Systems of intelligent control of manual operations in industrial production are being implemented in many industries nowadays. Such systems use high-resolution cameras and computer vision algorithms to automatically track the operator's manipulations and prevent technological errors in the assembly process. At the same time compliance with safety regulations in the workspace is monitored. As a result, the defect rate of manufactured products and the number of accidents during the manual assembly of any device are decreased. Before implementing an intelligent control system into a real production it is necessary to calculate its efficiency. In order to do it experiments on the stand for manual operations control systems were carried out. This paper proposes the methodology for calculating the efficiency indicators. This mathematical approach is based on the IoU calculation of real- and predicted-time intervals between assembly stages. The results show high precision in tracking the validity of manual assembly and do not depend on the duration of the assembly process.
Authors: Zuoyue Li, Zhenqiang Li, Zhaopeng Cui, Marc Pollefeys, Martin R. Oswald
Directly generating scenes from satellite imagery offers exciting possibilities for integration into applications like games and map services. However, challenges arise from significant view changes and scene scale. Previous efforts mainly focused on image or video generation, lacking exploration into the adaptability of scene generation for arbitrary views. Existing 3D generation works either operate at the object level or are difficult to utilize the geometry obtained from satellite imagery. To overcome these limitations, we propose a novel architecture for direct 3D scene generation by introducing diffusion models into 3D sparse representations and combining them with neural rendering techniques. Specifically, our approach generates texture colors at the point level for a given geometry using a 3D diffusion model first, which is then transformed into a scene representation in a feed-forward manner. The representation can be utilized to render arbitrary views which would excel in both single-frame quality and inter-frame consistency. Experiments in two city-scale datasets show that our model demonstrates proficiency in generating photo-realistic street-view image sequences and cross-view urban scenes from satellite imagery.
Authors: Lynn Vonder Haar, Timothy Elvira, Luke Newcomb, Omar Ochoa
Although accuracy and other common metrics can provide a useful window into the performance of an object detection model, they lack a deeper view of the model's decision process. Regardless of the quality of the training data and process, the features that an object detection model learns cannot be guaranteed. A model may learn a relationship between certain background context, i.e., scene level objects, and the presence of the labeled classes. Furthermore, standard performance verification and metrics would not identify this phenomenon. This paper presents a new black box explainability method for additional verification of object detection models by finding the impact of scene level objects on the identification of the objects within the image. By comparing the accuracies of a model on test data with and without certain scene level objects, the contributions of these objects to the model's performance becomes clearer. The experiment presented here will assess the impact of buildings and people in image context on the detection of emergency road vehicles by a fine-tuned YOLOv8 model. A large increase in accuracy in the presence of a scene level object will indicate the model's reliance on that object to make its detections. The results of this research lead to providing a quantitative explanation of the object detection model's decision process, enabling a deeper understanding of the model's performance.
Authors: Eric Peh, Paritosh Parmar, Basura Fernando
In this work, we introduce the novel concept of visually Connecting Actions and Their Effects (CATE) in video understanding. CATE can have applications in areas like task planning and learning from demonstration. We propose different CATE-based task formulations, such as action selection and action specification, where video understanding models connect actions and effects at semantic and fine-grained levels. We observe that different formulations produce representations capturing intuitive action properties. We also design various baseline models for action selection and action specification. Despite the intuitive nature of the task, we observe that models struggle, and humans outperform them by a large margin. The study aims to establish a foundation for future efforts, showcasing the flexibility and versatility of connecting actions and effects in video understanding, with the hope of inspiring advanced formulations and models.
Authors: Fernando Pérez-García, Harshita Sharma, Sam Bond-Taylor, Kenza Bouzid, Valentina Salvatelli, Maximilian Ilse, Shruthi Bannur, Daniel C. Castro, Anton Schwaighofer, Matthew P. Lungren, Maria Wetscherek, Noel Codella, Stephanie L. Hyland, Javier Alvarez-Valle, Ozan Oktay
Language-supervised pre-training has proven to be a valuable method for extracting semantically meaningful features from images, serving as a foundational element in multimodal systems within the computer vision and medical imaging domains. However, resulting features are limited by the information contained within the text. This is particularly problematic in medical imaging, where radiologists' written findings focus on specific observations; a challenge compounded by the scarcity of paired imaging-text data due to concerns over leakage of personal health information. In this work, we fundamentally challenge the prevailing reliance on language supervision for learning general purpose biomedical imaging encoders. We introduce RAD-DINO, a biomedical image encoder pre-trained solely on unimodal biomedical imaging data that obtains similar or greater performance than state-of-the-art biomedical language supervised models on a diverse range of benchmarks. Specifically, the quality of learned representations is evaluated on standard imaging tasks (classification and semantic segmentation), and a vision-language alignment task (text report generation from images). To further demonstrate the drawback of language supervision, we show that features from RAD-DINO correlate with other medical records (e.g., sex or age) better than language-supervised models, which are generally not mentioned in radiology reports. Finally, we conduct a series of ablations determining the factors in RAD-DINO's performance; notably, we observe that RAD-DINO's downstream performance scales well with the quantity and diversity of training data, demonstrating that image-only supervision is a scalable approach for training a foundational biomedical image encoder.
Authors: Boxiao Pan, Zhan Xu, Chun-Hao Paul Huang, Krishna Kumar Singh, Yang Zhou, Leonidas J. Guibas, Jimei Yang
Generating video background that tailors to foreground subject motion is an important problem for the movie industry and visual effects community. This task involves synthesizing background that aligns with the motion and appearance of the foreground subject, while also complies with the artist's creative intention. We introduce ActAnywhere, a generative model that automates this process which traditionally requires tedious manual efforts. Our model leverages the power of large-scale video diffusion models, and is specifically tailored for this task. ActAnywhere takes a sequence of foreground subject segmentation as input and an image that describes the desired scene as condition, to produce a coherent video with realistic foreground-background interactions while adhering to the condition frame. We train our model on a large-scale dataset of human-scene interaction videos. Extensive evaluations demonstrate the superior performance of our model, significantly outperforming baselines. Moreover, we show that ActAnywhere generalizes to diverse out-of-distribution samples, including non-human subjects. Please visit our project webpage at https://actanywhere.github.io.
Authors: Matthew Kowal, Achal Dave, Rares Ambrus, Adrien Gaidon, Konstantinos G. Derpanis, Pavel Tokmakov
This paper studies the problem of concept-based interpretability of transformer representations for videos. Concretely, we seek to explain the decision-making process of video transformers based on high-level, spatiotemporal concepts that are automatically discovered. Prior research on concept-based interpretability has concentrated solely on image-level tasks. Comparatively, video models deal with the added temporal dimension, increasing complexity and posing challenges in identifying dynamic concepts over time. In this work, we systematically address these challenges by introducing the first Video Transformer Concept Discovery (VTCD) algorithm. To this end, we propose an efficient approach for unsupervised identification of units of video transformer representations - concepts, and ranking their importance to the output of a model. The resulting concepts are highly interpretable, revealing spatio-temporal reasoning mechanisms and object-centric representations in unstructured video models. Performing this analysis jointly over a diverse set of supervised and self-supervised representations, we discover that some of these mechanism are universal in video transformers. Finally, we demonstrate that VTCDcan be used to improve model performance for fine-grained tasks.
Authors: Prakhar Kaushik, Aayush Mishra, Adam Kortylewski, Alan Yuille
We consider the problem of source-free unsupervised category-level pose estimation from only RGB images to a target domain without any access to source domain data or 3D annotations during adaptation. Collecting and annotating real-world 3D data and corresponding images is laborious, expensive, yet unavoidable process, since even 3D pose domain adaptation methods require 3D data in the target domain. We introduce 3DUDA, a method capable of adapting to a nuisance-ridden target domain without 3D or depth data. Our key insight stems from the observation that specific object subparts remain stable across out-of-domain (OOD) scenarios, enabling strategic utilization of these invariant subcomponents for effective model updates. We represent object categories as simple cuboid meshes, and harness a generative model of neural feature activations modeled at each mesh vertex learnt using differential rendering. We focus on individual locally robust mesh vertex features and iteratively update them based on their proximity to corresponding features in the target domain even when the global pose is not correct. Our model is then trained in an EM fashion, alternating between updating the vertex features and the feature extractor. We show that our method simulates fine-tuning on a global pseudo-labeled dataset under mild assumptions, which converges to the target domain asymptotically. Through extensive empirical validation, including a complex extreme UDA setup which combines real nuisances, synthetic noise, and occlusion, we demonstrate the potency of our simple approach in addressing the domain shift challenge and significantly improving pose estimation accuracy.
Authors: André O. Françani, Marcos R. O. A. Maximo
Deep learning algorithms have driven expressive progress in many complex tasks. The loss function is a core component of deep learning techniques, guiding the learning process of neural networks. This paper contributes by introducing a consistency loss for visual odometry with deep learning-based approaches. The motion consistency loss explores repeated motions that appear in consecutive overlapped video clips. Experimental results show that our approach increased the performance of a model on the KITTI odometry benchmark.
Authors: Emma Harvey, Hauke Sandhaus, Abigail Z. Jacobs, Emanuel Moss, Mona Sloane
Motion capture systems, used across various domains, make body representations concrete through technical processes. We argue that the measurement of bodies and the validation of measurements for motion capture systems can be understood as social practices. By analyzing the findings of a systematic literature review (N=278) through the lens of social practice theory, we show how these practices, and their varying attention to errors, become ingrained in motion capture design and innovation over time. Moreover, we show how contemporary motion capture systems perpetuate assumptions about human bodies and their movements. We suggest that social practices of measurement and validation are ubiquitous in the development of data- and sensor-driven systems more broadly, and provide this work as a basis for investigating hidden design assumptions and their potential negative consequences in human-computer interaction.
Authors: Dominik A. Kloepfer, João F. Henriques, Dylan Campbell
Extracting point correspondences from two or more views of a scene is a fundamental computer vision problem with particular importance for relative camera pose estimation and structure-from-motion. Existing local feature matching approaches, trained with correspondence supervision on large-scale datasets, obtain highly-accurate matches on the test sets. However, they do not generalise well to new datasets with different characteristics to those they were trained on, unlike classic feature extractors. Instead, they require finetuning, which assumes that ground-truth correspondences or ground-truth camera poses and 3D structure are available. We relax this assumption by removing the requirement of 3D structure, e.g., depth maps or point clouds, and only require camera pose information, which can be obtained from odometry. We do so by replacing correspondence losses with epipolar losses, which encourage putative matches to lie on the associated epipolar line. While weaker than correspondence supervision, we observe that this cue is sufficient for finetuning existing models on new data. We then further relax the assumption of known camera poses by using pose estimates in a novel bootstrapping approach. We evaluate on highly challenging datasets, including an indoor drone dataset and an outdoor smartphone camera dataset, and obtain state-of-the-art results without strong supervision.
Authors: Boyi Li, Jathushan Rajasegaran, Yossi Gandelsman, Alexei A. Efros, Jitendra Malik
In this paper, we present a diffusion model-based framework for animating people from a single image for a given target 3D motion sequence. Our approach has two core components: a) learning priors about invisible parts of the human body and clothing, and b) rendering novel body poses with proper clothing and texture. For the first part, we learn an in-filling diffusion model to hallucinate unseen parts of a person given a single image. We train this model on texture map space, which makes it more sample-efficient since it is invariant to pose and viewpoint. Second, we develop a diffusion-based rendering pipeline, which is controlled by 3D human poses. This produces realistic renderings of novel poses of the person, including clothing, hair, and plausible in-filling of unseen regions. This disentangled approach allows our method to generate a sequence of images that are faithful to the target motion in the 3D pose and, to the input image in terms of visual similarity. In addition to that, the 3D control allows various synthetic camera trajectories to render a person. Our experiments show that our method is resilient in generating prolonged motions and varied challenging and complex poses compared to prior methods. Please check our website for more details: https://boyiliee.github.io/3DHM.github.io/.
Authors: Ekin Ugurel, Steffen Coenen, Minda Zhou Chen, Cynthia Chen
Rapid identification and response to breaking events, particularly those that pose a threat to human life such as natural disasters or conflicts, is of paramount importance. The prevalence of mobile devices and the ubiquity of network connectivity has generated a massive amount of temporally- and spatially-stamped data. Numerous studies have used mobile data to derive individual human mobility patterns for various applications. Similarly, the increasing number of orbital satellites has made it easier to gather high-resolution images capturing a snapshot of a geographical area in sub-daily temporal frequency. We propose a novel data fusion methodology integrating satellite imagery with privacy-enhanced mobile data to augment the event inference task, whether in real-time or historical. In the absence of boots on the ground, mobile data is able to give an approximation of human mobility, proximity to one another, and the built environment. On the other hand, satellite imagery can provide visual information on physical changes to the built and natural environment. The expected use cases for our methodology include small-scale disaster detection (i.e., tornadoes, wildfires, and floods) in rural regions, search and rescue operation augmentation for lost hikers in remote wilderness areas, and identification of active conflict areas and population displacement in war-torn states. Our implementation is open-source on GitHub: https://github.com/ekinugurel/SatMobFusion.
Authors: Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao
This work presents Depth Anything, a highly practical solution for robust monocular depth estimation. Without pursuing novel technical modules, we aim to build a simple yet powerful foundation model dealing with any images under any circumstances. To this end, we scale up the dataset by designing a data engine to collect and automatically annotate large-scale unlabeled data (~62M), which significantly enlarges the data coverage and thus is able to reduce the generalization error. We investigate two simple yet effective strategies that make data scaling-up promising. First, a more challenging optimization target is created by leveraging data augmentation tools. It compels the model to actively seek extra visual knowledge and acquire robust representations. Second, an auxiliary supervision is developed to enforce the model to inherit rich semantic priors from pre-trained encoders. We evaluate its zero-shot capabilities extensively, including six public datasets and randomly captured photos. It demonstrates impressive generalization ability. Further, through fine-tuning it with metric depth information from NYUv2 and KITTI, new SOTAs are set. Our better depth model also results in a better depth-conditioned ControlNet. Our models are released at https://github.com/LiheYoung/Depth-Anything.
Authors: Zhuoran Shen, Mingyuan Zhang, Haiyu Zhao, Shuai Yi, Hongsheng Li
Dot-product attention has wide applications in computer vision and natural language processing. However, its memory and computational costs grow quadratically with the input size. Such growth prohibits its application on high-resolution inputs. To remedy this drawback, this paper proposes a novel efficient attention mechanism equivalent to dot-product attention but with substantially less memory and computational costs. Its resource efficiency allows more widespread and flexible integration of attention modules into a network, which leads to better accuracies. Empirical evaluations demonstrated the effectiveness of its advantages. Efficient attention modules brought significant performance boosts to object detectors and instance segmenters on MS-COCO 2017. Further, the resource efficiency democratizes attention to complex models, where high costs prohibit the use of dot-product attention. As an exemplar, a model with efficient attention achieved state-of-the-art accuracies for stereo depth estimation on the Scene Flow dataset. Code is available at https://github.com/cmsflash/efficient-attention.
Authors: Chen Liang, Yu Wu, Yawei Luo, Yi Yang
Text-based video segmentation is a challenging task that segments out the natural language referred objects in videos. It essentially requires semantic comprehension and fine-grained video understanding. Existing methods introduce language representation into segmentation models in a bottom-up manner, which merely conducts vision-language interaction within local receptive fields of ConvNets. We argue that such interaction is not fulfilled since the model can barely construct region-level relationships given partial observations, which is contrary to the description logic of natural language/referring expressions. In fact, people usually describe a target object using relations with other objects, which may not be easily understood without seeing the whole video. To address the issue, we introduce a novel top-down approach by imitating how we human segment an object with the language guidance. We first figure out all candidate objects in videos and then choose the refereed one by parsing relations among those high-level objects. Three kinds of object-level relations are investigated for precise relationship understanding, i.e., positional relation, text-guided semantic relation, and temporal relation. Extensive experiments on A2D Sentences and J-HMDB Sentences show our method outperforms state-of-the-art methods by a large margin. Qualitative results also show our results are more explainable.
Authors: Chen Liang, Yu Wu, Tianfei Zhou, Wenguan Wang, Zongxin Yang, Yunchao Wei, Yi Yang
Referring video object segmentation (RVOS) aims to segment video objects with the guidance of natural language reference. Previous methods typically tackle RVOS through directly grounding linguistic reference over the image lattice. Such bottom-up strategy fails to explore object-level cues, easily leading to inferior results. In this work, we instead put forward a two-stage, top-down RVOS solution. First, an exhaustive set of object tracklets is constructed by propagating object masks detected from several sampled frames to the entire video. Second, a Transformer-based tracklet-language grounding module is proposed, which models instance-level visual relations and cross-modal interactions simultaneously and efficiently. Our model ranks first place on CVPR2021 Referring Youtube-VOS challenge.
Authors: Chen Liang, Wenguan Wang, Tianfei Zhou, Jiaxu Miao, Yawei Luo, Yi Yang
We explore the task of language-guided video segmentation (LVS). Previous algorithms mostly adopt 3D CNNs to learn video representation, struggling to capture long-term context and easily suffering from visual-linguistic misalignment. In light of this, we present Locater (local-global context aware Transformer), which augments the Transformer architecture with a finite memory so as to query the entire video with the language expression in an efficient manner. The memory is designed to involve two components -- one for persistently preserving global video content, and one for dynamically gathering local temporal context and segmentation history. Based on the memorized local-global context and the particular content of each frame, Locater holistically and flexibly comprehends the expression as an adaptive query vector for each frame. The vector is used to query the corresponding frame for mask generation. The memory also allows Locater to process videos with linear time complexity and constant size memory, while Transformer-style self-attention computation scales quadratically with sequence length. To thoroughly examine the visual grounding capability of LVS models, we contribute a new LVS dataset, A2D-S+, which is built upon A2D-S dataset but poses increased challenges in disambiguating among similar objects. Experiments on three LVS datasets and our A2D-S+ show that Locater outperforms previous state-of-the-arts. Further, we won the 1st place in the Referring Video Object Segmentation Track of the 3rd Large-scale Video Object Segmentation Challenge, where Locater served as the foundation for the winning solution. Our code and dataset are available at: https://github.com/leonnnop/Locater
Authors: Mahan Agha Zahedi, Niloofar Gholamrezaei, Alex Doboli
Computational modeling of artwork meaning is complex and difficult. This is because art interpretation is multidimensional and highly subjective. This paper experimentally investigated the degree to which a state-of-the-art Deep Convolutional Neural Network (DCNN), a popular Machine Learning approach, can correctly distinguish modern conceptual art work into the galleries devised by art curators. Two hypotheses were proposed to state that the DCNN model uses Exhibited Properties for classification, like shape and color, but not Non-Exhibited Properties, such as historical context and artist intention. The two hypotheses were experimentally validated using a methodology designed for this purpose. VGG-11 DCNN pre-trained on ImageNet dataset and discriminatively fine-tuned was trained on handcrafted datasets designed from real-world conceptual photography galleries. Experimental results supported the two hypotheses showing that the DCNN model ignores Non-Exhibited Properties and uses only Exhibited Properties for artwork classification. This work points to current DCNN limitations, which should be addressed by future DNN models.
Authors: Jia Cheng Hu, Roberto Cavicchioli, Alessandro Capotondi
We introduce a method called the Expansion mechanism that processes the input unconstrained by the number of elements in the sequence. By doing so, the model can learn more effectively compared to traditional attention-based approaches. To support this claim, we design a novel architecture ExpansionNet v2 that achieved strong results on the MS COCO 2014 Image Captioning challenge and the State of the Art in its respective category, with a score of 143.7 CIDErD in the offline test split, 140.8 CIDErD in the online evaluation server and 72.9 AllCIDEr on the nocaps validation set. Additionally, we introduce an End to End training algorithm up to 2.8 times faster than established alternatives. Source code available at: https://github.com/jchenghu/ExpansionNet_v2
Authors: Changzhen Li, Jie Zhang, Shuzhe Wu, Xin Jin, Shiguang Shan
Recently action recognition has received more and more attention for its comprehensive and practical applications in intelligent surveillance and human-computer interaction. However, few-shot action recognition has not been well explored and remains challenging because of data scarcity. In this paper, we propose a novel hierarchical compositional representations (HCR) learning approach for few-shot action recognition. Specifically, we divide a complicated action into several sub-actions by carefully designed hierarchical clustering and further decompose the sub-actions into more fine-grained spatially attentional sub-actions (SAS-actions). Although there exist large differences between base classes and novel classes, they can share similar patterns in sub-actions or SAS-actions. Furthermore, we adopt the Earth Mover's Distance in the transportation problem to measure the similarity between video samples in terms of sub-action representations. It computes the optimal matching flows between sub-actions as distance metric, which is favorable for comparing fine-grained patterns. Extensive experiments show our method achieves the state-of-the-art results on HMDB51, UCF101 and Kinetics datasets.
Authors: Ginger Delmas, Philippe Weinzaepfel, Thomas Lucas, Francesc Moreno-Noguer, Grégory Rogez
Natural language plays a critical role in many computer vision applications, such as image captioning, visual question answering, and cross-modal retrieval, to provide fine-grained semantic information. Unfortunately, while human pose is key to human understanding, current 3D human pose datasets lack detailed language descriptions. To address this issue, we have introduced the PoseScript dataset. This dataset pairs more than six thousand 3D human poses from AMASS with rich human-annotated descriptions of the body parts and their spatial relationships. Additionally, to increase the size of the dataset to a scale that is compatible with data-hungry learning algorithms, we have proposed an elaborate captioning process that generates automatic synthetic descriptions in natural language from given 3D keypoints. This process extracts low-level pose information, known as "posecodes", using a set of simple but generic rules on the 3D keypoints. These posecodes are then combined into higher level textual descriptions using syntactic rules. With automatic annotations, the amount of available data significantly scales up (100k), making it possible to effectively pretrain deep models for finetuning on human captions. To showcase the potential of annotated poses, we present three multi-modal learning tasks that utilize the PoseScript dataset. Firstly, we develop a pipeline that maps 3D poses and textual descriptions into a joint embedding space, allowing for cross-modal retrieval of relevant poses from large-scale datasets. Secondly, we establish a baseline for a text-conditioned model generating 3D poses. Thirdly, we present a learned process for generating pose descriptions. These applications demonstrate the versatility and usefulness of annotated poses in various tasks and pave the way for future research in the field.
Authors: Jielin Qiu, Yi Zhu, Xingjian Shi, Florian Wenzel, Zhiqiang Tang, Ding Zhao, Bo Li, Mu Li
Multimodal image-text models have shown remarkable performance in the past few years. However, evaluating robustness against distribution shifts is crucial before adopting them in real-world applications. In this work, we investigate the robustness of 12 popular open-sourced image-text models under common perturbations on five tasks (image-text retrieval, visual reasoning, visual entailment, image captioning, and text-to-image generation). In particular, we propose several new multimodal robustness benchmarks by applying 17 image perturbation and 16 text perturbation techniques on top of existing datasets. We observe that multimodal models are not robust to image and text perturbations, especially to image perturbations. Among the tested perturbation methods, character-level perturbations constitute the most severe distribution shift for text, and zoom blur is the most severe shift for image data. We also introduce two new robustness metrics (\textbf{MMI} for MultiModal Impact score and \textbf{MOR} for Missing Object Rate) for proper evaluations of multimodal models. We hope our extensive study sheds light on new directions for the development of robust multimodal models. More details can be found on the project webpage: \url{https://MMRobustness.github.io}.
Authors: Shaoyuan Xie, Zichao Li, Zeyu Wang, Cihang Xie
In recent years, camera-based 3D object detection has gained widespread attention for its ability to achieve high performance with low computational cost. However, the robustness of these methods to adversarial attacks has not been thoroughly examined, especially when considering their deployment in safety-critical domains like autonomous driving. In this study, we conduct the first comprehensive investigation of the robustness of leading camera-based 3D object detection approaches under various adversarial conditions. We systematically analyze the resilience of these models under two attack settings: white-box and black-box; focusing on two primary objectives: classification and localization. Additionally, we delve into two types of adversarial attack techniques: pixel-based and patch-based. Our experiments yield four interesting findings: (a) bird's-eye-view-based representations exhibit stronger robustness against localization attacks; (b) depth-estimation-free approaches have the potential to show stronger robustness; (c) accurate depth estimation effectively improves robustness for depth-estimation-based methods; (d) incorporating multi-frame benign inputs can effectively mitigate adversarial attacks. We hope our findings can steer the development of future camera-based object detection models with enhanced adversarial robustness.
Authors: Guoyang Xie, Jinbao Wang, Jiaqi Liu, Jiayi Lyu, Yong Liu, Chengjie Wang, Feng Zheng, Yaochu Jin
Image anomaly detection (IAD) is an emerging and vital computer vision task in industrial manufacturing (IM). Recently, many advanced algorithms have been reported, but their performance deviates considerably with various IM settings. We realize that the lack of a uniform IM benchmark is hindering the development and usage of IAD methods in real-world applications. In addition, it is difficult for researchers to analyze IAD algorithms without a uniform benchmark. To solve this problem, we propose a uniform IM benchmark, for the first time, to assess how well these algorithms perform, which includes various levels of supervision (unsupervised versus fully supervised), learning paradigms (few-shot, continual and noisy label), and efficiency (memory usage and inference speed). Then, we construct a comprehensive image anomaly detection benchmark (IM-IAD), which includes 19 algorithms on seven major datasets with a uniform setting. Extensive experiments (17,017 total) on IM-IAD provide in-depth insights into IAD algorithm redesign or selection. Moreover, the proposed IM-IAD benchmark challenges existing algorithms and suggests future research directions. To foster reproducibility and accessibility, the source code of IM-IAD is uploaded on the website, https://github.com/M-3LAB/IM-IAD.
Authors: Shikun Liu, Linxi Fan, Edward Johns, Zhiding Yu, Chaowei Xiao, Anima Anandkumar
Recent vision-language models have shown impressive multi-modal generation capabilities. However, typically they require training huge models on massive datasets. As a more scalable alternative, we introduce Prismer, a data- and parameter-efficient vision-language model that leverages an ensemble of task-specific experts. Prismer only requires training of a small number of components, with the majority of network weights inherited from multiple readily-available, pre-trained experts, and kept frozen during training. By leveraging experts from a wide range of domains, we show Prismer can efficiently pool this expert knowledge and adapt it to various vision-language reasoning tasks. In our experiments, we show that Prismer achieves fine-tuned and few-shot learning performance which is competitive with current state-of-the-arts, whilst requiring up to two orders of magnitude less training data. Code is available at https://github.com/NVlabs/prismer.
Authors: Jieren Deng, Xin Zhou, Hao Tian, Zhihong Pan, Derek Aguiar
Distilling the structured information captured in feature maps has contributed to improved results for object detection tasks, but requires careful selection of baseline architectures and substantial pre-training. Self-distillation addresses these limitations and has recently achieved state-of-the-art performance for object detection despite making several simplifying architectural assumptions. Building on this work, we propose Smooth and Stepwise Self-Distillation (SSSD) for object detection. Our SSSD architecture forms an implicit teacher from object labels and a feature pyramid network backbone to distill label-annotated feature maps using Jensen-Shannon distance, which is smoother than distillation losses used in prior work. We additionally add a distillation coefficient that is adaptively configured based on the learning rate. We extensively benchmark SSSD against a baseline and two state-of-the-art object detector architectures on the COCO dataset by varying the coefficients and backbone and detector networks. We demonstrate that SSSD achieves higher average precision in most experimental settings, is robust to a wide range of coefficients, and benefits from our stepwise distillation procedure.
Authors: Marin Scalbert, Maria Vakalopoulou, Florent Couzinié-Devy
In Self-Supervised Learning (SSL), models are typically pretrained, fine-tuned, and evaluated on the same domains. However, they tend to perform poorly when evaluated on unseen domains, a challenge that Unsupervised Domain Generalization (UDG) seeks to address. Current UDG methods rely on domain labels, which are often challenging to collect, and domain-specific architectures that lack scalability when confronted with numerous domains, making the current methodology impractical and rigid. Inspired by contrastive-based UDG methods that mitigate spurious correlations by restricting comparisons to examples from the same domain, we hypothesize that eliminating style variability within a batch could provide a more convenient and flexible way to reduce spurious correlations without requiring domain labels. To verify this hypothesis, we introduce Batch Styles Standardization (BSS), a relatively simple yet powerful Fourier-based method to standardize the style of images in a batch specifically designed for integration with SSL methods to tackle UDG. Combining BSS with existing SSL methods offers serious advantages over prior UDG methods: (1) It eliminates the need for domain labels or domain-specific network components to enhance domain-invariance in SSL representations, and (2) offers flexibility as BSS can be seamlessly integrated with diverse contrastive-based but also non-contrastive-based SSL methods. Experiments on several UDG datasets demonstrate that it significantly improves downstream task performances on unseen domains, often outperforming or rivaling with UDG methods. Finally, this work clarifies the underlying mechanisms contributing to BSS's effectiveness in improving domain-invariance in SSL representations and performances on unseen domain.
Authors: Bastian Wittmann, Johannes C. Paetzold, Chinmay Prabhakar, Daniel Rueckert, Bjoern Menze
Link prediction algorithms aim to infer the existence of connections (or links) between nodes in network-structured data and are typically applied to refine the connectivity among nodes. In this work, we focus on link prediction for flow-driven spatial networks, which are embedded in a Euclidean space and relate to physical exchange and transportation processes (e.g., blood flow in vessels or traffic flow in road networks). To this end, we propose the Graph Attentive Vectors (GAV) link prediction framework. GAV models simplified dynamics of physical flow in spatial networks via an attentive, neighborhood-aware message-passing paradigm, updating vector embeddings in a constrained manner. We evaluate GAV on eight flow-driven spatial networks given by whole-brain vessel graphs and road networks. GAV demonstrates superior performances across all datasets and metrics and outperformed the state-of-the-art on the ogbl-vessel benchmark at the time of submission by 12% (98.38 vs. 87.98 AUC). All code is publicly available on GitHub.
Authors: Wenbo Hu, Hongjian Zhan, Cong Liu, Bing Yin, Yue Lu
In the field of historical manuscript research, scholars frequently encounter novel symbols in ancient texts, investing considerable effort in their identification and documentation. Although some object detection methods have achieved impressive performance, they primarily excel at detecting categories included in training datasets, often failing to recognize novel symbols without retraining. To overcome this limitation, we propose a novel One-shot learning-based Text Spotting (OTS) approach that accurately and reliably spots novel characters with just one annotated support sample. Drawing inspiration from cognitive research, we introduce a spatial alignment module that finds, focuses on, and learns the most discriminative spatial regions in the query image based on one support image. Especially, since the low-resource spotting task often faces the problem of example imbalance, we propose a novel loss function called torus loss which can make the embedding space of distance metric more discriminative. Our approach is highly efficient and requires only a few training samples while exhibiting the remarkable ability to handle novel characters and symbols. To enhance dataset diversity, a new manuscript dataset that contains the ancient Dongba hieroglyphics (DBH) is created, a script associated with China and developed by the ancestors of the Naxi minority. We conduct experiments on publicly available DBH, EGY, VML-HD, TKH, and NC datasets. The experimental results demonstrate that OTS outperforms the state-of-the-art methods in one-shot text spotting. Overall, our proposed method offers promising applications in text spotting in historical manuscripts.
Authors: Yang Liu, Muzhi Zhu, Hengtao Li, Hao Chen, Xinlong Wang, Chunhua Shen
Powered by large-scale pre-training, vision foundation models exhibit significant potential in open-world image understanding. However, unlike large language models that excel at directly tackling various language tasks, vision foundation models require a task-specific model structure followed by fine-tuning on specific tasks. In this work, we present Matcher, a novel perception paradigm that utilizes off-the-shelf vision foundation models to address various perception tasks. Matcher can segment anything by using an in-context example without training. Additionally, we design three effective components within the Matcher framework to collaborate with these foundation models and unleash their full potential in diverse perception tasks. Matcher demonstrates impressive generalization performance across various segmentation tasks, all without training. For example, it achieves 52.7% mIoU on COCO-20$^i$ with one example, surpassing the state-of-the-art specialist model by 1.6%. In addition, Matcher achieves 33.0% mIoU on the proposed LVIS-92$^i$ for one-shot semantic segmentation, outperforming the state-of-the-art generalist model by 14.4%. Our visualization results further showcase the open-world generality and flexibility of Matcher when applied to images in the wild. Our code can be found at https://github.com/aim-uofa/Matcher.
Authors: Jieren Deng, Xin Zhou, Hao Tian, Zhihong Pan, Derek Aguiar
The bokeh effect is an artistic technique that blurs out-of-focus areas in a photograph and has gained interest due to recent developments in text-to-image synthesis and the ubiquity of smart-phone cameras and photo-sharing apps. Prior work on rendering bokeh effects have focused on post hoc image manipulation to produce similar blurring effects in existing photographs using classical computer graphics or neural rendering techniques, but have either depth discontinuity artifacts or are restricted to reproducing bokeh effects that are present in the training data. More recent diffusion based models can synthesize images with an artistic style, but either require the generation of high-dimensional masks, expensive fine-tuning, or affect global image characteristics. In this paper, we present GBSD, the first generative text-to-image model that synthesizes photorealistic images with a bokeh style. Motivated by how image synthesis occurs progressively in diffusion models, our approach combines latent diffusion models with a 2-stage conditioning algorithm to render bokeh effects on semantically defined objects. Since we can focus the effect on objects, this semantic bokeh effect is more versatile than classical rendering techniques. We evaluate GBSD both quantitatively and qualitatively and demonstrate its ability to be applied in both text-to-image and image-to-image settings.
Authors: Fanda Fan, Chaoxu Guo, Litong Gong, Biao Wang, Tiezheng Ge, Yuning Jiang, Chunjie Luo, Jianfeng Zhan
Video outpainting aims to adequately complete missing areas at the edges of video frames. Compared to image outpainting, it presents an additional challenge as the model should maintain the temporal consistency of the filled area. In this paper, we introduce a masked 3D diffusion model for video outpainting. We use the technique of mask modeling to train the 3D diffusion model. This allows us to use multiple guide frames to connect the results of multiple video clip inferences, thus ensuring temporal consistency and reducing jitter between adjacent frames. Meanwhile, we extract the global frames of the video as prompts and guide the model to obtain information other than the current video clip using cross-attention. We also introduce a hybrid coarse-to-fine inference pipeline to alleviate the artifact accumulation problem. The existing coarse-to-fine pipeline only uses the infilling strategy, which brings degradation because the time interval of the sparse frames is too large. Our pipeline benefits from bidirectional learning of the mask modeling and thus can employ a hybrid strategy of infilling and interpolation when generating sparse frames. Experiments show that our method achieves state-of-the-art results in video outpainting tasks. More results and codes are provided at our https://fanfanda.github.io/M3DDM/.
Authors: Jinglong Wang, Xiawei Li, Jing Zhang, Qingyuan Xu, Qin Zhou, Qian Yu, Lu Sheng, Dong Xu
The pre-trained text-image discriminative models, such as CLIP, has been explored for open-vocabulary semantic segmentation with unsatisfactory results due to the loss of crucial localization information and awareness of object shapes. Recently, there has been a growing interest in expanding the application of generative models from generation tasks to semantic segmentation. These approaches utilize generative models either for generating annotated data or extracting features to facilitate semantic segmentation. This typically involves generating a considerable amount of synthetic data or requiring additional mask annotations. To this end, we uncover the potential of generative text-to-image diffusion models (e.g., Stable Diffusion) as highly efficient open-vocabulary semantic segmenters, and introduce a novel training-free approach named DiffSegmenter. The insight is that to generate realistic objects that are semantically faithful to the input text, both the complete object shapes and the corresponding semantics are implicitly learned by diffusion models. We discover that the object shapes are characterized by the self-attention maps while the semantics are indicated through the cross-attention maps produced by the denoising U-Net, forming the basis of our segmentation results.Additionally, we carefully design effective textual prompts and a category filtering mechanism to further enhance the segmentation results. Extensive experiments on three benchmark datasets show that the proposed DiffSegmenter achieves impressive results for open-vocabulary semantic segmentation.
Authors: Gang Wu, Junjun Jiang, Kui Jiang, Xianming Liu
Contrastive learning has emerged as a prevailing paradigm for high-level vision tasks, which, by introducing properly negative samples, has also been exploited for low-level vision tasks to achieve a compact optimization space to account for their ill-posed nature. However, existing methods rely on manually predefined and task-oriented negatives, which often exhibit pronounced task-specific biases. To address this challenge, our paper introduces an innovative method termed 'learning from history', which dynamically generates negative samples from the target model itself. Our approach, named Model Contrastive paradigm for Image Restoration (MCIR), rejuvenates latency models as negative models, making it compatible with diverse image restoration tasks. We propose the Self-Prior guided Negative loss (SPN) to enable it. This approach significantly enhances existing models when retrained with the proposed model contrastive paradigm. The results show significant improvements in image restoration across various tasks and architectures. For example, models retrained with SPN outperform the original FFANet and DehazeFormer by 3.41 dB and 0.57 dB on the RESIDE indoor dataset for image dehazing. Similarly, they achieve notable improvements of 0.47 dB on SPA-Data over IDT for image deraining and 0.12 dB on Manga109 for a 4x scale super-resolution over lightweight SwinIR, respectively. Code and retrained models are available at https://github.com/Aitical/MCIR.
Authors: YuTeng Ye, Jiale Cai, Hang Zhou, Guanwen Li, Youjia Zhang, Zikai Song, Chenxing Gao, Junqing Yu, Wei Yang
In spite of the rapidly evolving landscape of text-to-image generation, the synthesis and manipulation of multiple entities while adhering to specific relational constraints pose enduring challenges. This paper introduces an innovative progressive synthesis and editing operation that systematically incorporates entities into the target image, ensuring their adherence to spatial and relational constraints at each sequential step. Our key insight stems from the observation that while a pre-trained text-to-image diffusion model adeptly handles one or two entities, it often falters when dealing with a greater number. To address this limitation, we propose harnessing the capabilities of a Large Language Model (LLM) to decompose intricate and protracted text descriptions into coherent directives adhering to stringent formats. To facilitate the execution of directives involving distinct semantic operations-namely insertion, editing, and erasing-we formulate the Stimulus, Response, and Fusion (SRF) framework. Within this framework, latent regions are gently stimulated in alignment with each operation, followed by the fusion of the responsive latent components to achieve cohesive entity manipulation. Our proposed framework yields notable advancements in object synthesis, particularly when confronted with intricate and lengthy textual inputs. Consequently, it establishes a new benchmark for text-to-image generation tasks, further elevating the field's performance standards.
Authors: Xinyi Yu, Guanbin Li, Wei Lou, Siqi Liu, Xiang Wan, Yan Chen, Haofeng Li
Nuclei segmentation is a fundamental but challenging task in the quantitative analysis of histopathology images. Although fully-supervised deep learning-based methods have made significant progress, a large number of labeled images are required to achieve great segmentation performance. Considering that manually labeling all nuclei instances for a dataset is inefficient, obtaining a large-scale human-annotated dataset is time-consuming and labor-intensive. Therefore, augmenting a dataset with only a few labeled images to improve the segmentation performance is of significant research and application value. In this paper, we introduce the first diffusion-based augmentation method for nuclei segmentation. The idea is to synthesize a large number of labeled images to facilitate training the segmentation model. To achieve this, we propose a two-step strategy. In the first step, we train an unconditional diffusion model to synthesize the Nuclei Structure that is defined as the representation of pixel-level semantic and distance transform. Each synthetic nuclei structure will serve as a constraint on histopathology image synthesis and is further post-processed to be an instance map. In the second step, we train a conditioned diffusion model to synthesize histopathology images based on nuclei structures. The synthetic histopathology images paired with synthetic instance maps will be added to the real dataset for training the segmentation model. The experimental results show that by augmenting 10% labeled real dataset with synthetic samples, one can achieve comparable segmentation results with the fully-supervised baseline. The code is released in: https://github.com/lhaof/Nudiff
Authors: Kaichen Zhou, Jia-Xing Zhong, Sangyun Shin, Kai Lu, Yiyuan Yang, Andrew Markham, Niki Trigoni
The introduction of neural radiance fields has greatly improved the effectiveness of view synthesis for monocular videos. However, existing algorithms face difficulties when dealing with uncontrolled or lengthy scenarios, and require extensive training time specific to each new scenario. To tackle these limitations, we propose DynPoint, an algorithm designed to facilitate the rapid synthesis of novel views for unconstrained monocular videos. Rather than encoding the entirety of the scenario information into a latent representation, DynPoint concentrates on predicting the explicit 3D correspondence between neighboring frames to realize information aggregation. Specifically, this correspondence prediction is achieved through the estimation of consistent depth and scene flow information across frames. Subsequently, the acquired correspondence is utilized to aggregate information from multiple reference frames to a target frame, by constructing hierarchical neural point clouds. The resulting framework enables swift and accurate view synthesis for desired views of target frames. The experimental results obtained demonstrate the considerable acceleration of training time achieved - typically an order of magnitude - by our proposed method while yielding comparable outcomes compared to prior approaches. Furthermore, our method exhibits strong robustness in handling long-duration videos without learning a canonical representation of video content.
Authors: Mikaela Angelina Uy, Kiyohiro Nakayama, Guandao Yang, Rahul Krishna Thomas, Leonidas Guibas, Ke Li
Neural radiance fields (NeRF) rely on volume rendering to synthesize novel views. Volume rendering requires evaluating an integral along each ray, which is numerically approximated with a finite sum that corresponds to the exact integral along the ray under piecewise constant volume density. As a consequence, the rendered result is unstable w.r.t. the choice of samples along the ray, a phenomenon that we dub quadrature instability. We propose a mathematically principled solution by reformulating the sample-based rendering equation so that it corresponds to the exact integral under piecewise linear volume density. This simultaneously resolves multiple issues: conflicts between samples along different rays, imprecise hierarchical sampling, and non-differentiability of quantiles of ray termination distances w.r.t. model parameters. We demonstrate several benefits over the classical sample-based rendering equation, such as sharper textures, better geometric reconstruction, and stronger depth supervision. Our proposed formulation can be also be used as a drop-in replacement to the volume rendering equation of existing NeRF-based methods. Our project page can be found at pl-nerf.github.io.
Authors: Jieyu Yao, Hao Yu, Paul Judge, Jiabin Jia, Sasa Djokic, Verner Püvi, Matti Lehtonen, Jan Meyer
Even though the use of power electronics PE loads offers enhanced electrical energy conversion efficiency and control, they remain the primary sources of harmonics in grids. When diverse loads are connected in the distribution system, their interactions complicate establishing analytical models for the relationship between harmonic voltages and currents. To solve this, our paper presents a data-driven model using MCReSANet to construct the highly nonlinear between harmonic voltage and current. Two datasets from PCCs in Finland and Germany are utilized, which demonstrates that MCReSANet is capable of establishing accurate nonlinear mappings, even in the presence of various network characteristics for selected Finland and Germany datasets. The model built by MCReSANet can improve the MAE by 10% and 14% compared to the CNN, and by 8% and 17% compared to the MLP for both Finnish and German datasets, also showing much lower model uncertainty than others. This is a crucial prerequisite for more precise SHAP value-based feature importance analysis, which is a method for the model interpretability analysis in this paper. The results by feature importance analysis show the detailed relationships between each order of harmonic voltage and current in the distribution system. There is an interactive impact on each order of harmonic current, but some orders of harmonic voltages have a dominant influence on harmonic current emissions: positive sequence and zero sequence harmonics have the dominant importance in the Finnish and German networks, respectively, which conforms to the pattern of connected load types in two selected Finnish and German datasets. This paper enhances the potential for understanding and predicting harmonic current emissions by diverse PE loads in distribution systems, which is beneficial to more effective management for optimizing power quality in diverse grid environments.
Authors: Gabriel De Araujo, Shanlin Sun, Xiaohui Xie
Image registration has traditionally been done using two distinct approaches: learning based methods, relying on robust deep neural networks, and optimization-based methods, applying complex mathematical transformations to warp images accordingly. Of course, both paradigms offer advantages and disadvantages, and, in this work, we seek to combine their respective strengths into a single streamlined framework, using the outputs of the learning based method as initial parameters for optimization while prioritizing computational power for the image pairs that offer the greatest loss. Our investigations showed improvements of up to 1.6% in test data, while maintaining the same inference time, and a substantial 1.0% points performance gain in deformation field smoothness.
Authors: Lukas Hirsch, Yu Huang, Hernan A. Makse, Danny F. Martinez, Mary Hughes, Sarah Eskreis-Winkler, Katja Pinker, Elizabeth Morris, Lucas C. Parra, Elizabeth J. Sutton
Women with an increased life-time risk of breast cancer undergo supplemental annual screening MRI. We propose to predict the risk of developing breast cancer within one year based on the current MRI, with the objective of reducing screening burden and facilitating early detection. An AI algorithm was developed on 53,858 breasts from 12,694 patients who underwent screening or diagnostic MRI and accrued over 12 years, with 2,331 confirmed cancers. A first U-Net was trained to segment lesions and identify regions of concern. A second convolutional network was trained to detect malignant cancer using features extracted by the U-Net. This network was then fine-tuned to estimate the risk of developing cancer within a year in cases that radiologists considered normal or likely benign. Risk predictions from this AI were evaluated with a retrospective analysis of 9,183 breasts from a high-risk screening cohort, which were not used for training. Statistical analysis focused on the tradeoff between number of omitted exams versus negative predictive value, and number of potential early detections versus positive predictive value. The AI algorithm identified regions of concern that coincided with future tumors in 52% of screen-detected cancers. Upon directed review, a radiologist found that 71.3% of cancers had a visible correlate on the MRI prior to diagnosis, 65% of these correlates were identified by the AI model. Reevaluating these regions in 10% of all cases with higher AI-predicted risk could have resulted in up to 33% early detections by a radiologist. Additionally, screening burden could have been reduced in 16% of lower-risk cases by recommending a later follow-up without compromising current interval cancer rate. With increasing datasets and improving image quality we expect this new AI-aided, adaptive screening to meaningfully reduce screening burden and improve early detection.
Authors: Jingchun Zhou, Tianyu Liang, Dehuan Zhang, Zongxin He
Neural Radiance Field (NeRF) technology demonstrates immense potential in novel viewpoint synthesis tasks, due to its physics-based volumetric rendering process, which is particularly promising in underwater scenes. Addressing the limitations of existing underwater NeRF methods in handling light attenuation caused by the water medium and the lack of real Ground Truth (GT) supervision, this study proposes WaterHE-NeRF. We develop a new water-ray tracing field by Retinex theory that precisely encodes color, density, and illuminance attenuation in three-dimensional space. WaterHE-NeRF, through its illuminance attenuation mechanism, generates both degraded and clear multi-view images and optimizes image restoration by combining reconstruction loss with Wasserstein distance. Additionally, the use of histogram equalization (HE) as pseudo-GT enhances the network's accuracy in preserving original details and color distribution. Extensive experiments on real underwater datasets and synthetic datasets validate the effectiveness of WaterHE-NeRF. Our code will be made publicly available.
Authors: Jingchun Zhou, Qilin Gai, Kin-man Lam, Xianping Fu
In underwater environments, variations in suspended particle concentration and turbidity cause severe image degradation, posing significant challenges to image enhancement (IE) and object detection (OD) tasks. Currently, in-air image enhancement and detection methods have made notable progress, but their application in underwater conditions is limited due to the complexity and variability of these environments. Fine-tuning in-air models saves high overhead and has more optional reference work than building an underwater model from scratch. To address these issues, we design a transfer plugin with multiple priors for converting in-air models to underwater applications, named IA2U. IA2U enables efficient application in underwater scenarios, thereby improving performance in Underwater IE and OD. IA2U integrates three types of underwater priors: the water type prior that characterizes the degree of image degradation, such as color and visibility; the degradation prior, focusing on differences in details and textures; and the sample prior, considering the environmental conditions at the time of capture and the characteristics of the photographed object. Utilizing a Transformer-like structure, IA2U employs these priors as query conditions and a joint task loss function to achieve hierarchical enhancement of task-level underwater image features, therefore considering the requirements of two different tasks, IE and OD. Experimental results show that IA2U combined with an in-air model can achieve superior performance in underwater image enhancement and object detection tasks. The code will be made publicly available.
Authors: Jingchun Zhou, Zongxin He, Dehuan Zhang, Kin-man Lam, Xianping Fu, Yi Wang
Underwater image enhancement (UIE) is a challenging task due to the complex degradation caused by underwater environments. To solve this issue, previous methods often idealize the degradation process, and neglect the impact of medium noise and object motion on the distribution of image features, limiting the generalization and adaptability of the model. Previous methods use the reference gradient that is constructed from original images and synthetic ground-truth images. This may cause the network performance to be influenced by some low-quality training data. Our approach utilizes predicted images to dynamically update pseudo-labels, adding a dynamic gradient to optimize the network's gradient space. This process improves image quality and avoids local optima. Moreover, we propose a Feature Restoration and Reconstruction module (FRR) based on a Channel Combination Inference (CCI) strategy and a Frequency Domain Smoothing module (FRS). These modules decouple other degradation features while reducing the impact of various types of noise on network performance. Experiments on multiple public datasets demonstrate the superiority of our method over existing state-of-the-art approaches, especially in achieving performance milestones: PSNR of 25.6dB and SSIM of 0.93 on the UIEB dataset. Its efficiency in terms of parameter size and inference time further attests to its broad practicality. The code will be made publicly available.
Authors: Qi Tang, Yao Zhao, Meiqin Liu, Jian Jin, Chao Yao
As a critical clue of video super-resolution (VSR), inter-frame alignment significantly impacts overall performance. However, accurate pixel-level alignment is a challenging task due to the intricate motion interweaving in the video. In response to this issue, we introduce a novel paradigm for VSR named Semantic Lens, predicated on semantic priors drawn from degraded videos. Specifically, video is modeled as instances, events, and scenes via a Semantic Extractor. Those semantics assist the Pixel Enhancer in understanding the recovered contents and generating more realistic visual results. The distilled global semantics embody the scene information of each frame, while the instance-specific semantics assemble the spatial-temporal contexts related to each instance. Furthermore, we devise a Semantics-Powered Attention Cross-Embedding (SPACE) block to bridge the pixel-level features with semantic knowledge, composed of a Global Perspective Shifter (GPS) and an Instance-Specific Semantic Embedding Encoder (ISEE). Concretely, the GPS module generates pairs of affine transformation parameters for pixel-level feature modulation conditioned on global semantics. After that, the ISEE module harnesses the attention mechanism to align the adjacent frames in the instance-centric semantic space. In addition, we incorporate a simple yet effective pre-alignment module to alleviate the difficulty of model training. Extensive experiments demonstrate the superiority of our model over existing state-of-the-art VSR methods.
Authors: Shahzad Ahmad, Sukalpa Chanda, Yogesh S Rawat
Recent advancements in large-scale pre-training of visual-language models on paired image-text data have demonstrated impressive generalization capabilities for zero-shot tasks. Building on this success, efforts have been made to adapt these image-based visual-language models, such as CLIP, for videos extending their zero-shot capabilities to the video domain. While these adaptations have shown promising results, they come at a significant computational cost and struggle with effectively modeling the crucial temporal aspects inherent to the video domain. In this study, we present EZ-CLIP, a simple and efficient adaptation of CLIP that addresses these challenges. EZ-CLIP leverages temporal visual prompting for seamless temporal adaptation, requiring no fundamental alterations to the core CLIP architecture while preserving its remarkable generalization abilities. Moreover, we introduce a novel learning objective that guides the temporal visual prompts to focus on capturing motion, thereby enhancing its learning capabilities from video data. We conducted extensive experiments on five different benchmark datasets, thoroughly evaluating EZ-CLIP for zero-shot learning and base-to-novel video action recognition, and also demonstrating its potential for few-shot generalization.Impressively, with a mere 5.2 million learnable parameters (as opposed to the 71.1 million in the prior best model), EZ-CLIP can be efficiently trained on a single GPU, outperforming existing approaches in several evaluations.
Authors: Fahim Ahmed Zaman, Wahidul Alam, Tarun Kanti Roy, Amanda Chang, Kan Liu, Xiaodong Wu
Researchers have shown significant correlations among segmented objects in various medical imaging modalities and disease related pathologies. Several studies showed that using hand crafted features for disease prediction neglects the immense possibility to use latent features from deep learning (DL) models which may reduce the overall accuracy of differential diagnosis. However, directly using classification or segmentation models on medical to learn latent features opt out robust feature selection and may lead to overfitting. To fill this gap, we propose a novel feature selection technique using the latent space of a segmentation model that can aid diagnosis. We evaluated our method in differentiating a rare cardiac disease: Takotsubo Syndrome (TTS) from the ST elevation myocardial infarction (STEMI) using echocardiogram videos (echo). TTS can mimic clinical features of STEMI in echo and extremely hard to distinguish. Our approach shows promising results in differential diagnosis of TTS with 82% diagnosis accuracy beating the previous state-of-the-art (SOTA) approach. Moreover, the robust feature selection technique using LASSO algorithm shows great potential in reducing the redundant features and creates a robust pipeline for short- and long-term disease prognoses in the downstream analysis.
Authors: Ingyun Lee, Wooju Lee, Hyun Myung
Deep neural networks have shown remarkable performance in image classification. However, their performance significantly deteriorates with corrupted input data. Domain generalization methods have been proposed to train robust models against out-of-distribution data. Data augmentation in the frequency domain is one of such approaches that enable a model to learn phase features to establish domain-invariant representations. This approach changes the amplitudes of the input data while preserving the phases. However, using fixed phases leads to susceptibility to phase fluctuations because amplitudes and phase fluctuations commonly occur in out-of-distribution. In this study, to address this problem, we introduce an approach using finite variation of the phases of input data rather than maintaining fixed phases. Based on the assumption that the degree of domain-invariant features varies for each phase, we propose a method to distinguish phases based on this degree. In addition, we propose a method called vital phase augmentation (VIPAug) that applies the variation to the phases differently according to the degree of domain-invariant features of given phases. The model depends more on the vital phases that contain more domain-invariant features for attaining robustness to amplitude and phase fluctuations. We present experimental evaluations of our proposed approach, which exhibited improved performance for both clean and corrupted data. VIPAug achieved SOTA performance on the benchmark CIFAR-10 and CIFAR-100 datasets, as well as near-SOTA performance on the ImageNet-100 and ImageNet datasets. Our code is available at https://github.com/excitedkid/vipaug.
Authors: Maghsood Salimi, Mohammad Loni, Sara Afshar, Antonio Cicchetti, Marjan Sirjani
The increasing demand for autonomous machines in construction environments necessitates the development of robust object detection algorithms that can perform effectively across various weather and environmental conditions. This paper introduces a new semantic segmentation dataset specifically tailored for construction sites, taking into account the diverse challenges posed by adverse weather and environmental conditions. The dataset is designed to enhance the training and evaluation of object detection models, fostering their adaptability and reliability in real-world construction applications. Our dataset comprises annotated images captured under a wide range of different weather conditions, including but not limited to sunny days, rainy periods, foggy atmospheres, and low-light situations. Additionally, environmental factors such as the existence of dirt/mud on the camera lens are integrated into the dataset through actual captures and synthetic generation to simulate the complex conditions prevalent in construction sites. We also generate synthetic images of the annotations including precise semantic segmentation masks for various objects commonly found in construction environments, such as wheel loader machines, personnel, cars, and structural elements. To demonstrate the dataset's utility, we evaluate state-of-the-art object detection algorithms on our proposed benchmark. The results highlight the dataset's success in adversarial training models across diverse conditions, showcasing its efficacy compared to existing datasets that lack such environmental variability.
Authors: Shanchuan Lin, Xiao Yang
Diffusion models trained with mean squared error loss tend to generate unrealistic samples. Current state-of-the-art models rely on classifier-free guidance to improve sample quality, yet its surprising effectiveness is not fully understood. In this paper, we show that the effectiveness of classifier-free guidance partly originates from it being a form of implicit perceptual guidance. As a result, we can directly incorporate perceptual loss in diffusion training to improve sample quality. Since the score matching objective used in diffusion training strongly resembles the denoising autoencoder objective used in unsupervised training of perceptual networks, the diffusion model itself is a perceptual network and can be used to generate meaningful perceptual loss. We propose a novel self-perceptual objective that results in diffusion models capable of generating more realistic samples. For conditional generation, our method only improves sample quality without entanglement with the conditional input and therefore does not sacrifice sample diversity. Our method can also improve sample quality for unconditional generation, which was not possible with classifier-free guidance before.
Authors: Joao P. C. Bertoldo, Dick Ameln, Ashwin Vaidya, Samet Akçay
Recent advances in visual anomaly detection research have seen AUROC and AUPRO scores on public benchmark datasets such as MVTec and VisA converge towards perfect recall, giving the impression that these benchmarks are near-solved. However, high AUROC and AUPRO scores do not always reflect qualitative performance, which limits the validity of these metrics in real-world applications. We argue that the artificial ceiling imposed by the lack of an adequate evaluation metric restrains progression of the field, and it is crucial that we revisit the evaluation metrics used to rate our algorithms. In response, we introduce Per-IMage Overlap (PIMO), a novel metric that addresses the shortcomings of AUROC and AUPRO. PIMO retains the recall-based nature of the existing metrics but introduces two distinctions: the assignment of curves (and respective area under the curve) is per-image, and its X-axis relies solely on normal images. Measuring recall per image simplifies instance score indexing and is more robust to noisy annotations. As we show, it also accelerates computation and enables the usage of statistical tests to compare models. By imposing low tolerance for false positives on normal images, PIMO provides an enhanced model validation procedure and highlights performance variations across datasets. Our experiments demonstrate that PIMO offers practical advantages and nuanced performance insights that redefine anomaly detection benchmarks -- notably challenging the perception that MVTec AD and VisA datasets have been solved by contemporary models. Available on GitHub: https://github.com/jpcbertoldo/aupimo.
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: Zhaobo Qi, Yibo Yuan, Xiaowen Ruan, Shuhui Wang, Weigang Zhang, Qingming Huang
Temporal Sentence Grounding in Video (TSGV) is troubled by dataset bias issue, which is caused by the uneven temporal distribution of the target moments for samples with similar semantic components in input videos or query texts. Existing methods resort to utilizing prior knowledge about bias to artificially break this uneven distribution, which only removes a limited amount of significant language biases. In this work, we propose the bias-conflict sample synthesis and adversarial removal debias strategy (BSSARD), which dynamically generates bias-conflict samples by explicitly leveraging potentially spurious correlations between single-modality features and the temporal position of the target moments. Through adversarial training, its bias generators continuously introduce biases and generate bias-conflict samples to deceive its grounding model. Meanwhile, the grounding model continuously eliminates the introduced biases, which requires it to model multi-modality alignment information. BSSARD will cover most kinds of coupling relationships and disrupt language and visual biases simultaneously. Extensive experiments on Charades-CD and ActivityNet-CD demonstrate the promising debiasing capability of BSSARD. Source codes are available at https://github.com/qzhb/BSSARD.
Authors: Win Kent Ong, Kam Woh Ng, Chee Seng Chan, Yi Zhe Song, Tao Xiang
Neural Radiance Field (NeRF) models have gained significant attention in the computer vision community in the recent past with state-of-the-art visual quality and produced impressive demonstrations. Since then, technopreneurs have sought to leverage NeRF models into a profitable business. Therefore, NeRF models make it worth the risk of plagiarizers illegally copying, re-distributing, or misusing those models. This paper proposes a comprehensive intellectual property (IP) protection framework for the NeRF model in both black-box and white-box settings, namely IPR-NeRF. In the black-box setting, a diffusion-based solution is introduced to embed and extract the watermark via a two-stage optimization process. In the white-box setting, a designated digital signature is embedded into the weights of the NeRF model by adopting the sign loss objective. Our extensive experiments demonstrate that not only does our approach maintain the fidelity (\ie, the rendering quality) of IPR-NeRF models, but it is also robust against both ambiguity and removal attacks compared to prior arts.
Authors: Ryosuke Watanabe, Keisuke Nonaka, Eduardo Pavez, Tatsuya Kobayashi, Antonio Ortega
Point clouds are utilized in various 3D applications such as cross-reality (XR) and realistic 3D displays. In some applications, e.g., for live streaming using a 3D point cloud, real-time point cloud denoising methods are required to enhance the visual quality. However, conventional high-precision denoising methods cannot be executed in real time for large-scale point clouds owing to the complexity of graph constructions with K nearest neighbors and noise level estimation. This paper proposes a fast graph-based denoising (FGBD) for a large-scale point cloud. First, high-speed graph construction is achieved by scanning a point cloud in various directions and searching adjacent neighborhoods on the scanning lines. Second, we propose a fast noise level estimation method using eigenvalues of the covariance matrix on a graph. Finally, we also propose a new low-cost filter selection method to enhance denoising accuracy to compensate for the degradation caused by the acceleration algorithms. In our experiments, we succeeded in reducing the processing time dramatically while maintaining accuracy relative to conventional denoising methods. Denoising was performed at 30fps, with frames containing approximately 1 million points.
Authors: Jun Wang, Chengfeng Zhou, Zhaoyan Ming, Lina Wei, Xudong Jiang, Dahong Qian
One of the fundamental challenges in microscopy (MS) image analysis is instance segmentation (IS), particularly when segmenting cluster regions where multiple objects of varying sizes and shapes may be connected or even overlapped in arbitrary orientations. Existing IS methods usually fail in handling such scenarios, as they rely on coarse instance representations such as keypoints and horizontal bounding boxes (h-bboxes). In this paper, we propose a novel one-stage framework named A2B-IS to address this challenge and enhance the accuracy of IS in MS images. Our approach represents each instance with a pixel-level mask map and a rotated bounding box (r-bbox). Unlike two-stage methods that use box proposals for segmentations, our method decouples mask and box predictions, enabling simultaneous processing to streamline the model pipeline. Additionally, we introduce a Gaussian skeleton map to aid the IS task in two key ways: (1) It guides anchor placement, reducing computational costs while improving the model's capacity to learn RoI-aware features by filtering out noise from background regions. (2) It ensures accurate isolation of densely packed instances by rectifying erroneous box predictions near instance boundaries. To further enhance the performance, we integrate two modules into the framework: (1) An Atrous Attention Block (A2B) designed to extract high-resolution feature maps with fine-grained multiscale information, and (2) A Semi-Supervised Learning (SSL) strategy that leverages both labeled and unlabeled images for model training. Our method has been thoroughly validated on two large-scale MS datasets, demonstrating its superiority over most state-of-the-art approaches.
Authors: Yunpeng Gong, Zhun Zhong, Zhiming Luo, Yansong Qu, Rongrong Ji, Min Jiang
In recent years, there has been significant research focusing on addressing security concerns in single-modal person re-identification (ReID) systems that are based on RGB images. However, the safety of cross-modality scenarios, which are more commonly encountered in practical applications involving images captured by infrared cameras, has not received adequate attention. The main challenge in cross-modality ReID lies in effectively dealing with visual differences between different modalities. For instance, infrared images are typically grayscale, unlike visible images that contain color information. Existing attack methods have primarily focused on the characteristics of the visible image modality, overlooking the features of other modalities and the variations in data distribution among different modalities. This oversight can potentially undermine the effectiveness of these methods in image retrieval across diverse modalities. This study represents the first exploration into the security of cross-modality ReID models and proposes a universal perturbation attack specifically designed for cross-modality ReID. This attack optimizes perturbations by leveraging gradients from diverse modality data, thereby disrupting the discriminator and reinforcing the differences between modalities. We conducted experiments on two widely used cross-modality datasets, namely RegDB and SYSU, which not only demonstrated the effectiveness of our method but also provided insights for future enhancements in the robustness of cross-modality ReID systems.
Authors: Xianfu Cheng, Weixiao Zhou, Xiang Li, Xiaoming Chen, Jian Yang, Tongliang Li, Zhoujun Li
Scene Text Recognition (STR) is a challenging task that involves recognizing text within images of natural scenes. Although current state-of-the-art models for STR exhibit high performance, they typically suffer from low inference efficiency due to their reliance on hybrid architectures comprised of visual encoders and sequence decoders. In this work, we propose the VIsion Permutable extractor for fast and efficient scene Text Recognition (VIPTR), which achieves an impressive balance between high performance and rapid inference speeds in the domain of STR. Specifically, VIPTR leverages a visual-semantic extractor with a pyramid structure, characterized by multiple self-attention layers, while eschewing the traditional sequence decoder. This design choice results in a lightweight and efficient model capable of handling inputs of varying sizes. Extensive experimental results on various standard datasets for both Chinese and English scene text recognition validate the superiority of VIPTR. Notably, the VIPTR-T (Tiny) variant delivers highly competitive accuracy on par with other lightweight models and achieves SOTA inference speeds. Meanwhile, the VIPTR-L (Large) variant attains greater recognition accuracy, while maintaining a low parameter count and favorable inference speed. Our proposed method provides a compelling solution for the STR challenge, which blends high accuracy with efficiency and greatly benefits real-world applications requiring fast and reliable text recognition. The code is publicly available at https://github.com/cxfyxl/VIPTR.
Authors: Changgu Chen, Junwei Shu, Lianggangxu Chen, Gaoqi He, Changbo Wang, Yang Li
Recent large-scale pre-trained diffusion models have demonstrated a powerful generative ability to produce high-quality videos from detailed text descriptions. However, exerting control over the motion of objects in videos generated by any video diffusion model is a challenging problem. In this paper, we propose a novel zero-shot moving object trajectory control framework, Motion-Zero, to enable a bounding-box-trajectories-controlled text-to-video diffusion model.To this end, an initial noise prior module is designed to provide a position-based prior to improve the stability of the appearance of the moving object and the accuracy of position. In addition, based on the attention map of the U-net, spatial constraints are directly applied to the denoising process of diffusion models, which further ensures the positional and spatial consistency of moving objects during the inference. Furthermore, temporal consistency is guaranteed with a proposed shift temporal attention mechanism. Our method can be flexibly applied to various state-of-the-art video diffusion models without any training process. Extensive experiments demonstrate our proposed method can control the motion trajectories of objects and generate high-quality videos.
Authors: Grzegorz Rypeść, Sebastian Cygert, Valeriya Khan, Tomasz Trzciński, Bartosz Zieliński, Bartłomiej Twardowski
Class-incremental learning is becoming more popular as it helps models widen their applicability while not forgetting what they already know. A trend in this area is to use a mixture-of-expert technique, where different models work together to solve the task. However, the experts are usually trained all at once using whole task data, which makes them all prone to forgetting and increasing computational burden. To address this limitation, we introduce a novel approach named SEED. SEED selects only one, the most optimal expert for a considered task, and uses data from this task to fine-tune only this expert. For this purpose, each expert represents each class with a Gaussian distribution, and the optimal expert is selected based on the similarity of those distributions. Consequently, SEED increases diversity and heterogeneity within the experts while maintaining the high stability of this ensemble method. The extensive experiments demonstrate that SEED achieves state-of-the-art performance in exemplar-free settings across various scenarios, showing the potential of expert diversification through data in continual learning.