Authors: Saghir Alfasly, Peyman Nejat, Sobhan Hemati, Jibran Khan, Isaiah Lahr, Areej Alsaafin, Abubakr Shafique, Nneka Comfere, Dennis Murphree, Chady Meroueh, Saba Yasir, Aaron Mangold, Lisa Boardman, Vijay Shah, Joaquin J. Garcia, H.R. Tizhoosh
Recently, several studies have reported on the fine-tuning of foundation models for image-text modeling in the field of medicine, utilizing images from online data sources such as Twitter and PubMed. Foundation models are large, deep artificial neural networks capable of learning the context of a specific domain through training on exceptionally extensive datasets. Through validation, we have observed that the representations generated by such models exhibit inferior performance in retrieval tasks within digital pathology when compared to those generated by significantly smaller, conventional deep networks.
Authors: Qihang Fan, Huaibo Huang, Mingrui Chen, Hongmin Liu, Ran He
Transformer first appears in the field of natural language processing and is later migrated to the computer vision domain, where it demonstrates excellent performance in vision tasks. However, recently, Retentive Network (RetNet) has emerged as an architecture with the potential to replace Transformer, attracting widespread attention in the NLP community. Therefore, we raise the question of whether transferring RetNet's idea to vision can also bring outstanding performance to vision tasks. To address this, we combine RetNet and Transformer to propose RMT. Inspired by RetNet, RMT introduces explicit decay into the vision backbone, bringing prior knowledge related to spatial distances to the vision model. This distance-related spatial prior allows for explicit control of the range of tokens that each token can attend to. Additionally, to reduce the computational cost of global modeling, we decompose this modeling process along the two coordinate axes of the image. Abundant experiments have demonstrated that our RMT exhibits exceptional performance across various computer vision tasks. For example, RMT achieves 84.1% Top1-acc on ImageNet-1k using merely 4.5G FLOPs. To the best of our knowledge, among all models, RMT achieves the highest Top1-acc when models are of similar size and trained with the same strategy. Moreover, RMT significantly outperforms existing vision backbones in downstream tasks such as object detection, instance segmentation, and semantic segmentation. Our work is still in progress.
Authors: Yifeng Xiong, Haoyu Ma, Shanlin Sun, Kun Han, Xiaohui Xie
Single-view novel view synthesis, the task of generating images from new viewpoints based on a single reference image, is an important but challenging task in computer vision. Recently, Denoising Diffusion Probabilistic Model (DDPM) has become popular in this area due to its strong ability to generate high-fidelity images. However, current diffusion-based methods directly rely on camera pose matrices as viewing conditions, globally and implicitly introducing 3D constraints. These methods may suffer from inconsistency among generated images from different perspectives, especially in regions with intricate textures and structures. In this work, we present Light Field Diffusion (LFD), a conditional diffusion-based model for single-view novel view synthesis. Unlike previous methods that employ camera pose matrices, LFD transforms the camera view information into light field encoding and combines it with the reference image. This design introduces local pixel-wise constraints within the diffusion models, thereby encouraging better multi-view consistency. Experiments on several datasets show that our LFD can efficiently generate high-fidelity images and maintain better 3D consistency even in intricate regions. Our method can generate images with higher quality than NeRF-based models, and we obtain sample quality similar to other diffusion-based models but with only one-third of the model size.
Authors: Ofir Gordon, Hai Victor Habi, Arnon Netzer
Quantization of deep neural networks (DNN) has become a key element in the efforts of embedding such networks on end-user devices. However, current quantization methods usually suffer from costly accuracy degradation. In this paper, we propose a new method for Enhanced Post Training Quantization named EPTQ. The method is based on knowledge distillation with an adaptive weighting of layers. In addition, we introduce a new label-free technique for approximating the Hessian trace of the task loss, named Label-Free Hessian. This technique removes the requirement of a labeled dataset for computing the Hessian. The adaptive knowledge distillation uses the Label-Free Hessian technique to give greater attention to the sensitive parts of the model while performing the optimization. Empirically, by employing EPTQ we achieve state-of-the-art results on a wide variety of models, tasks, and datasets, including ImageNet classification, COCO object detection, and Pascal-VOC for semantic segmentation. We demonstrate the performance and compatibility of EPTQ on an extended set of architectures, including CNNs, Transformers, hybrid, and MLP-only models.
Authors: Mohamed Afham, Satya Narayan Shukla, Omid Poursaeed, Pengchuan Zhang, Ashish Shah, Sernam Lim
While most modern video understanding models operate on short-range clips, real-world videos are often several minutes long with semantically consistent segments of variable length. A common approach to process long videos is applying a short-form video model over uniformly sampled clips of fixed temporal length and aggregating the outputs. This approach neglects the underlying nature of long videos since fixed-length clips are often redundant or uninformative. In this paper, we aim to provide a generic and adaptive sampling approach for long-form videos in lieu of the de facto uniform sampling. Viewing videos as semantically consistent segments, we formulate a task-agnostic, unsupervised, and scalable approach based on Kernel Temporal Segmentation (KTS) for sampling and tokenizing long videos. We evaluate our method on long-form video understanding tasks such as video classification and temporal action localization, showing consistent gains over existing approaches and achieving state-of-the-art performance on long-form video modeling.
Authors: Manuel Brack, Patrick Schramowski, Kristian Kersting
Text-conditioned image generation models have recently achieved astonishing image quality and alignment results. Consequently, they are employed in a fast-growing number of applications. Since they are highly data-driven, relying on billion-sized datasets randomly scraped from the web, they also produce unsafe content. As a contribution to the Adversarial Nibbler challenge, we distill a large set of over 1,000 potential adversarial inputs from existing safety benchmarks. Our analysis of the gathered prompts and corresponding images demonstrates the fragility of input filters and provides further insights into systematic safety issues in current generative image models.
Authors: David Li, Brandon Y. Feng, Amitabh Varshney
Recently, several approaches have emerged for generating neural representations with multiple levels of detail (LODs). LODs can improve the rendering by using lower resolutions and smaller model sizes when appropriate. However, existing methods generally focus on a few discrete LODs which suffer from aliasing and flicker artifacts as details are changed and limit their granularity for adapting to resource limitations. In this paper, we propose a method to encode light field networks with continuous LODs, allowing for finely tuned adaptations to rendering conditions. Our training procedure uses summed-area table filtering allowing efficient and continuous filtering at various LODs. Furthermore, we use saliency-based importance sampling which enables our light field networks to distribute their capacity, particularly limited at lower LODs, towards representing the details viewers are most likely to focus on. Incorporating continuous LODs into neural representations enables progressive streaming of neural representations, decreasing the latency and resource utilization for rendering.
Authors: Seyedalireza Khoshsirat, Chandra Kambhamettu
Answer grounding is the task of locating relevant visual evidence for the Visual Question Answering task. While a wide variety of attention methods have been introduced for this task, they suffer from the following three problems: designs that do not allow the usage of pre-trained networks and do not benefit from large data pre-training, custom designs that are not based on well-grounded previous designs, therefore limiting the learning power of the network, or complicated designs that make it challenging to re-implement or improve them. In this paper, we propose a novel architectural block, which we term Sentence Attention Block, to solve these problems. The proposed block re-calibrates channel-wise image feature-maps by explicitly modeling inter-dependencies between the image feature-maps and sentence embedding. We visually demonstrate how this block filters out irrelevant feature-maps channels based on sentence embedding. We start our design with a well-known attention method, and by making minor modifications, we improve the results to achieve state-of-the-art accuracy. The flexibility of our method makes it easy to use different pre-trained backbone networks, and its simplicity makes it easy to understand and be re-implemented. We demonstrate the effectiveness of our method on the TextVQA-X, VQS, VQA-X, and VizWiz-VQA-Grounding datasets. We perform multiple ablation studies to show the effectiveness of our design choices.
Authors: Serkan Savaş, Atilla Ergüzen
Human-Computer Interaction (HCI) has been the subject of research for many years, and recent studies have focused on improving its performance through various techniques. In the past decade, deep learning studies have shown high performance in various research areas, leading researchers to explore their application to HCI. Convolutional neural networks can be used to recognize hand gestures from images using deep architectures. In this study, we evaluated pre-trained high-performance deep architectures on the HG14 dataset, which consists of 14 different hand gesture classes. Among 22 different models, versions of the VGGNet and MobileNet models attained the highest accuracy rates. Specifically, the VGG16 and VGG19 models achieved accuracy rates of 94.64% and 94.36%, respectively, while the MobileNet and MobileNetV2 models achieved accuracy rates of 96.79% and 94.43%, respectively. We performed hand gesture recognition on the dataset using an ensemble learning technique, which combined the four most successful models. By utilizing these models as base learners and applying the Dirichlet ensemble technique, we achieved an accuracy rate of 98.88%. These results demonstrate the effectiveness of the deep ensemble learning technique for HCI and its potential applications in areas such as augmented reality, virtual reality, and game technologies.
Authors: Youssef Abdelkareem, Shady Shehata, Fakhri Karray
Novel view synthesis (NVS) of multi-human scenes imposes challenges due to the complex inter-human occlusions. Layered representations handle the complexities by dividing the scene into multi-layered radiance fields, however, they are mainly constrained to per-scene optimization making them inefficient. Generalizable human view synthesis methods combine the pre-fitted 3D human meshes with image features to reach generalization, yet they are mainly designed to operate on single-human scenes. Another drawback is the reliance on multi-step optimization techniques for parametric pre-fitting of the 3D body models that suffer from misalignment with the images in sparse view settings causing hallucinations in synthesized views. In this work, we propose, GenLayNeRF, a generalizable layered scene representation for free-viewpoint rendering of multiple human subjects which requires no per-scene optimization and very sparse views as input. We divide the scene into multi-human layers anchored by the 3D body meshes. We then ensure pixel-level alignment of the body models with the input views through a novel end-to-end trainable module that carries out iterative parametric correction coupled with multi-view feature fusion to produce aligned 3D models. For NVS, we extract point-wise image-aligned and human-anchored features which are correlated and fused using self-attention and cross-attention modules. We augment low-level RGB values into the features with an attention-based RGB fusion module. To evaluate our approach, we construct two multi-human view synthesis datasets; DeepMultiSyn and ZJU-MultiHuman. The results indicate that our proposed approach outperforms generalizable and non-human per-scene NeRF methods while performing at par with layered per-scene methods without test time optimization.
Authors: Mariano Rivera, Angello Hoyos
We present a novel approach to enhance the capabilities of VQVAE models through the integration of an Attentive Residual Encoder (AREN) and a Residual Pixel Attention layer. The objective of our research is to improve the performance of VQVAE while maintaining practical parameter levels. The AREN encoder is designed to operate effectively at multiple levels, accommodating diverse architectural complexities. The key innovation is the integration of an inter-pixel auto-attention mechanism into the AREN encoder. This approach allows us to efficiently capture and utilize contextual information across latent vectors. Additionally, our models uses additional encoding levels to further enhance the model's representational power. Our attention layer employs a minimal parameter approach, ensuring that latent vectors are modified only when pertinent information from other pixels is available. Experimental results demonstrate that our proposed modifications lead to significant improvements in data representation and generation, making VQVAEs even more suitable for a wide range of applications.
Authors: Duarte Rondao, Lei He, Nabil Aouf
Cameras are rapidly becoming the choice for on-board sensors towards space rendezvous due to their small form factor and inexpensive power, mass, and volume costs. When it comes to docking, however, they typically serve a secondary role, whereas the main work is done by active sensors such as lidar. This paper documents the development of a proposed AI-based (artificial intelligence) navigation algorithm intending to mature the use of on-board visible wavelength cameras as a main sensor for docking and on-orbit servicing (OOS), reducing the dependency on lidar and greatly reducing costs. Specifically, the use of AI enables the expansion of the relative navigation solution towards multiple classes of scenarios, e.g., in terms of targets or illumination conditions, which would otherwise have to be crafted on a case-by-case manner using classical image processing methods. Multiple convolutional neural network (CNN) backbone architectures are benchmarked on synthetically generated data of docking manoeuvres with the International Space Station (ISS), achieving position and attitude estimates close to 1% range-normalised and 1 deg, respectively. The integration of the solution with a physical prototype of the refuelling mechanism is validated in laboratory using a robotic arm to simulate a berthing procedure.
Authors: Wei Jiang, Wei Wang, Yue Chen
We study neural image compression based on the Sparse Visual Representation (SVR), where images are embedded into a discrete latent space spanned by learned visual codebooks. By sharing codebooks with the decoder, the encoder transfers integer codeword indices that are efficient and cross-platform robust, and the decoder retrieves the embedded latent feature using the indices for reconstruction. Previous SVR-based compression lacks effective mechanism for rate-distortion tradeoffs, where one can only pursue either high reconstruction quality or low transmission bitrate. We propose a Masked Adaptive Codebook learning (M-AdaCode) method that applies masks to the latent feature subspace to balance bitrate and reconstruction quality. A set of semantic-class-dependent basis codebooks are learned, which are weighted combined to generate a rich latent feature for high-quality reconstruction. The combining weights are adaptively derived from each input image, providing fidelity information with additional transmission costs. By masking out unimportant weights in the encoder and recovering them in the decoder, we can trade off reconstruction quality for transmission bits, and the masking rate controls the balance between bitrate and distortion. Experiments over the standard JPEG-AI dataset demonstrate the effectiveness of our M-AdaCode approach.
Authors: Krishna Kanth Nakka, Mathieu Salzmann
As 3D human pose estimation can now be achieved with very high accuracy in the supervised learning scenario, tackling the case where 3D pose annotations are not available has received increasing attention. In particular, several methods have proposed to learn image representations in a self-supervised fashion so as to disentangle the appearance information from the pose one. The methods then only need a small amount of supervised data to train a pose regressor using the pose-related latent vector as input, as it should be free of appearance information. In this paper, we carry out in-depth analysis to understand to what degree the state-of-the-art disentangled representation learning methods truly separate the appearance information from the pose one. First, we study disentanglement from the perspective of the self-supervised network, via diverse image synthesis experiments. Second, we investigate disentanglement with respect to the 3D pose regressor following an adversarial attack perspective. Specifically, we design an adversarial strategy focusing on generating natural appearance changes of the subject, and against which we could expect a disentangled network to be robust. Altogether, our analyses show that disentanglement in the three state-of-the-art disentangled representation learning frameworks if far from complete, and that their pose codes contain significant appearance information. We believe that our approach provides a valuable testbed to evaluate the degree of disentanglement of pose from appearance in self-supervised 3D human pose estimation.
Authors: Xinheng Wu, Jie Lu, Zhen Fang, Guangquan Zhang
Out-of-distribution (OOD) detection is crucial to modern deep learning applications by identifying and alerting about the OOD samples that should not be tested or used for making predictions. Current OOD detection methods have made significant progress when in-distribution (ID) and OOD samples are drawn from static distributions. However, this can be unrealistic when applied to real-world systems which often undergo continuous variations and shifts in ID and OOD distributions over time. Therefore, for an effective application in real-world systems, the development of OOD detection methods that can adapt to these dynamic and evolving distributions is essential. In this paper, we propose a novel and more realistic setting called continuously adaptive out-of-distribution (CAOOD) detection which targets on developing an OOD detection model that enables dynamic and quick adaptation to a new arriving distribution, with insufficient ID samples during deployment time. To address CAOOD, we develop meta OOD learning (MOL) by designing a learning-to-adapt diagram such that a good initialized OOD detection model is learned during the training process. In the testing process, MOL ensures OOD detection performance over shifting distributions by quickly adapting to new distributions with a few adaptations. Extensive experiments on several OOD benchmarks endorse the effectiveness of our method in preserving both ID classification accuracy and OOD detection performance on continuously shifting distributions.
Authors: Ping Li, Yu Zhang, Li Yuan, Huaxin Xiao, Binbin Lin, Xianghua Xu
Unsupervised Video Object Segmentation (VOS) aims at identifying the contours of primary foreground objects in videos without any prior knowledge. However, previous methods do not fully use spatial-temporal context and fail to tackle this challenging task in real-time. This motivates us to develop an efficient Long-Short Temporal Attention network (termed LSTA) for unsupervised VOS task from a holistic view. Specifically, LSTA consists of two dominant modules, i.e., Long Temporal Memory and Short Temporal Attention. The former captures the long-term global pixel relations of the past frames and the current frame, which models constantly present objects by encoding appearance pattern. Meanwhile, the latter reveals the short-term local pixel relations of one nearby frame and the current frame, which models moving objects by encoding motion pattern. To speedup the inference, the efficient projection and the locality-based sliding window are adopted to achieve nearly linear time complexity for the two light modules, respectively. Extensive empirical studies on several benchmarks have demonstrated promising performances of the proposed method with high efficiency.
Authors: Elisa Kreiss, Eric Zelikman, Christopher Potts, Nick Haber
Referenceless metrics (e.g., CLIPScore) use pretrained vision--language models to assess image descriptions directly without costly ground-truth reference texts. Such methods can facilitate rapid progress, but only if they truly align with human preference judgments. In this paper, we introduce ContextRef, a benchmark for assessing referenceless metrics for such alignment. ContextRef has two components: human ratings along a variety of established quality dimensions, and ten diverse robustness checks designed to uncover fundamental weaknesses. A crucial aspect of ContextRef is that images and descriptions are presented in context, reflecting prior work showing that context is important for description quality. Using ContextRef, we assess a variety of pretrained models, scoring functions, and techniques for incorporating context. None of the methods is successful with ContextRef, but we show that careful fine-tuning yields substantial improvements. ContextRef remains a challenging benchmark though, in large part due to the challenge of context dependence.
Authors: Fei Pan, Xu Yin, Seokju Lee, Sungeui Yoon, In So Kweon
Unsupervised domain adaptation (UDA) is an effective approach to handle the lack of annotations in the target domain for the semantic segmentation task. In this work, we consider a more practical UDA setting where the target domain contains sequential frames of the unlabeled videos which are easy to collect in practice. A recent study suggests self-supervised learning of the object motion from unlabeled videos with geometric constraints. We design a motion-guided domain adaptive semantic segmentation framework (MoDA), that utilizes self-supervised object motion to learn effective representations in the target domain. MoDA differs from previous methods that use temporal consistency regularization for the target domain frames. Instead, MoDA deals separately with the domain alignment on the foreground and background categories using different strategies. Specifically, MoDA contains foreground object discovery and foreground semantic mining to align the foreground domain gaps by taking the instance-level guidance from the object motion. Additionally, MoDA includes background adversarial training which contains a background category-specific discriminator to handle the background domain gaps. Experimental results on multiple benchmarks highlight the effectiveness of MoDA against existing approaches in the domain adaptive image segmentation and domain adaptive video segmentation. Moreover, MoDA is versatile and can be used in conjunction with existing state-of-the-art approaches to further improve performance.
Authors: Xiao Feng Zhang, Tian Yi Song, Jia Wei Yao
Segment Anything (SAM), an advanced universal image segmentation model trained on an expansive visual dataset, has set a new benchmark in image segmentation and computer vision. However, it faced challenges when it came to distinguishing between shadows and their backgrounds. To address this, we developed Deshadow-Anything, considering the generalization of large-scale datasets, and we performed Fine-tuning on large-scale datasets to achieve image shadow removal. The diffusion model can diffuse along the edges and textures of an image, helping to remove shadows while preserving the details of the image. Furthermore, we design Multi-Self-Attention Guidance (MSAG) and adaptive input perturbation (DDPM-AIP) to accelerate the iterative training speed of diffusion. Experiments on shadow removal tasks demonstrate that these methods can effectively improve image restoration performance.
Authors: Shunli Wang, Qing Yu, Shuaibing Wang, Dingkang Yang, Liuzhen Su, Xiao Zhao, Haopeng Kuang, Peixuan Zhang, Peng Zhai, Lihua Zhang
The fine-grained medical action analysis task has received considerable attention from pattern recognition communities recently, but it faces the problems of data and algorithm shortage. Cardiopulmonary Resuscitation (CPR) is an essential skill in emergency treatment. Currently, the assessment of CPR skills mainly depends on dummies and trainers, leading to high training costs and low efficiency. For the first time, this paper constructs a vision-based system to complete error action recognition and skill assessment in CPR. Specifically, we define 13 types of single-error actions and 74 types of composite error actions during external cardiac compression and then develop a video dataset named CPR-Coach. By taking the CPR-Coach as a benchmark, this paper thoroughly investigates and compares the performance of existing action recognition models based on different data modalities. To solve the unavoidable Single-class Training & Multi-class Testing problem, we propose a humancognition-inspired framework named ImagineNet to improve the model's multierror recognition performance under restricted supervision. Extensive experiments verify the effectiveness of the framework. We hope this work could advance research toward fine-grained medical action analysis and skill assessment. The CPR-Coach dataset and the code of ImagineNet are publicly available on Github.
Authors: Kaizhao Liang, Xu Cao, Kuei-Da Liao, Tianren Gao, Zhengyu Chen, Tejas Nama
Disease progression simulation is a crucial area of research that has significant implications for clinical diagnosis, prognosis, and treatment. One major challenge in this field is the lack of continuous medical imaging monitoring of individual patients over time. To address this issue, we develop a novel framework termed Progressive Image Editing (PIE) that enables controlled manipulation of disease-related image features, facilitating precise and realistic disease progression simulation. Specifically, we leverage recent advancements in text-to-image generative models to simulate disease progression accurately and personalize it for each patient. We theoretically analyze the iterative refining process in our framework as a gradient descent with an exponentially decayed learning rate. To validate our framework, we conduct experiments in three medical imaging domains. Our results demonstrate the superiority of PIE over existing methods such as Stable Diffusion Walk and Style-Based Manifold Extrapolation based on CLIP score (Realism) and Disease Classification Confidence (Alignment). Our user study collected feedback from 35 veteran physicians to assess the generated progressions. Remarkably, 76.2% of the feedback agrees with the fidelity of the generated progressions. To our best knowledge, PIE is the first of its kind to generate disease progression images meeting real-world standards. It is a promising tool for medical research and clinical practice, potentially allowing healthcare providers to model disease trajectories over time, predict future treatment responses, and improve patient outcomes.
Authors: Yinpeng Dong, Huanran Chen, Jiawei Chen, Zhengwei Fang, Xiao Yang, Yichi Zhang, Yu Tian, Hang Su, Jun Zhu
Multimodal Large Language Models (MLLMs) that integrate text and other modalities (especially vision) have achieved unprecedented performance in various multimodal tasks. However, due to the unsolved adversarial robustness problem of vision models, MLLMs can have more severe safety and security risks by introducing the vision inputs. In this work, we study the adversarial robustness of Google's Bard, a competitive chatbot to ChatGPT that released its multimodal capability recently, to better understand the vulnerabilities of commercial MLLMs. By attacking white-box surrogate vision encoders or MLLMs, the generated adversarial examples can mislead Bard to output wrong image descriptions with a 22% success rate based solely on the transferability. We show that the adversarial examples can also attack other MLLMs, e.g., a 26% attack success rate against Bing Chat and a 86% attack success rate against ERNIE bot. Moreover, we identify two defense mechanisms of Bard, including face detection and toxicity detection of images. We design corresponding attacks to evade these defenses, demonstrating that the current defenses of Bard are also vulnerable. We hope this work can deepen our understanding on the robustness of MLLMs and facilitate future research on defenses. Our code is available at https://github.com/thu-ml/Attack-Bard.
Authors: Jiaxin Zhang, Shiyuan Chen, Haoran Yin, Ruohong Mei, Xuan Liu, Cong Yang, Qian Zhang, Wei Sui
The recent development of online static map element (a.k.a. HD Map) construction algorithms has raised a vast demand for data with ground truth annotations. However, available public datasets currently cannot provide high-quality training data regarding consistency and accuracy. To this end, we present CAMA: a vision-centric approach for Consistent and Accurate Map Annotation. Without LiDAR inputs, our proposed framework can still generate high-quality 3D annotations of static map elements. Specifically, the annotation can achieve high reprojection accuracy across all surrounding cameras and is spatial-temporal consistent across the whole sequence. We apply our proposed framework to the popular nuScenes dataset to provide efficient and highly accurate annotations. Compared with the original nuScenes static map element, models trained with annotations from CAMA achieve lower reprojection errors (e.g., 4.73 vs. 8.03 pixels).
Authors: Guan-Cheng Lee
Recently, multi-modality models have been introduced because of the complementary information from different sensors such as LiDAR and cameras. It requires paired data along with precise calibrations for all modalities, the complicated calibration among modalities hugely increases the cost of collecting such high-quality datasets, and hinder it from being applied to practical scenarios. Inherit from the previous works, we not only fuse the information from multi-modality without above issues, and also exhaust the information in the RGB modality. We introduced the 2D Detection Annotations Transmittable Aggregation(\textbf{2DDATA}), designing a data-specific branch, called \textbf{Local Object Branch}, which aims to deal with points in a certain bounding box, because of its easiness of acquiring 2D bounding box annotations. We demonstrate that our simple design can transmit bounding box prior information to the 3D encoder model, proving the feasibility of large multi-modality models fused with modality-specific data.
Authors: Chengliang Wang, Xinrun Chen, Haojian Ning, Shiying Li
In the analysis of optical coherence tomography angiography (OCTA) images, the operation of segmenting specific targets is necessary. Existing methods typically train on supervised datasets with limited samples (approximately a few hundred), which can lead to overfitting. To address this, the low-rank adaptation technique is adopted for foundation model fine-tuning and proposed corresponding prompt point generation strategies to process various segmentation tasks on OCTA datasets. This method is named SAM-OCTA and has been experimented on the publicly available OCTA-500 dataset. While achieving state-of-the-art performance metrics, this method accomplishes local vessel segmentation as well as effective artery-vein segmentation, which was not well-solved in previous works. The code is available at: https://github.com/ShellRedia/SAM-OCTA.
Authors: Rajesh Kumar, Can Isik, Chilukuri K. Mohan
We present a novel adversarial model for authentication systems that use gait patterns recorded by the inertial measurement unit (IMU) built into smartphones. The attack idea is inspired by and named after the concept of a dictionary attack on knowledge (PIN or password) based authentication systems. In particular, this work investigates whether it is possible to build a dictionary of IMUGait patterns and use it to launch an attack or find an imitator who can actively reproduce IMUGait patterns that match the target's IMUGait pattern. Nine physically and demographically diverse individuals walked at various levels of four predefined controllable and adaptable gait factors (speed, step length, step width, and thigh-lift), producing 178 unique IMUGait patterns. Each pattern attacked a wide variety of user authentication models. The deeper analysis of error rates (before and after the attack) challenges the belief that authentication systems based on IMUGait patterns are the most difficult to spoof; further research is needed on adversarial models and associated countermeasures.
Authors: Tongtong Zhang, Yuanxiang Li
Existing NeRF models for satellite images suffer from slow speeds, mandatory solar information as input, and limitations in handling large satellite images. In response, we present SatensoRF, which significantly accelerates the entire process while employing fewer parameters for satellite imagery of large size. Besides, we observed that the prevalent assumption of Lambertian surfaces in neural radiance fields falls short for vegetative and aquatic elements. In contrast to the traditional hierarchical MLP-based scene representation, we have chosen a multiscale tensor decomposition approach for color, volume density, and auxiliary variables to model the lightfield with specular color. Additionally, to rectify inconsistencies in multi-date imagery, we incorporate total variation loss to restore the density tensor field and treat the problem as a denosing task.To validate our approach, we conducted assessments of SatensoRF using subsets from the spacenet multi-view dataset, which includes both multi-date and single-date multi-view RGB images. Our results clearly demonstrate that SatensoRF surpasses the state-of-the-art Sat-NeRF series in terms of novel view synthesis performance. Significantly, SatensoRF requires fewer parameters for training, resulting in faster training and inference speeds and reduced computational demands.
Authors: Qingtian Wu, Liming Zhang
Extreme head postures pose a common challenge across a spectrum of facial analysis tasks, including face detection, facial landmark detection (FLD), and head pose estimation (HPE). These tasks are interdependent, where accurate FLD relies on robust face detection, and HPE is intricately associated with these key points. This paper focuses on the integration of these tasks, particularly when addressing the complexities posed by large-angle face poses. The primary contribution of this study is the proposal of a real-time multi-task detection system capable of simultaneously performing joint detection of faces, facial landmarks, and head poses. This system builds upon the widely adopted YOLOv8 detection framework. It extends the original object detection head by incorporating additional landmark regression head, enabling efficient localization of crucial facial landmarks. Furthermore, we conduct optimizations and enhancements on various modules within the original YOLOv8 framework. To validate the effectiveness and real-time performance of our proposed model, we conduct extensive experiments on 300W-LP and AFLW2000-3D datasets. The results obtained verify the capability of our model to tackle large-angle face pose challenges while delivering real-time performance across these interconnected tasks.
Authors: Thanh Nguyen, Trung Pham, Chaoning Zhang, Tung Luu, Thang Vu, Chang D. Yoo
Self-supervised learning (SSL) has gained remarkable success, for which contrastive learning (CL) plays a key role. However, the recent development of new non-CL frameworks has achieved comparable or better performance with high improvement potential, prompting researchers to enhance these frameworks further. Assimilating CL into non-CL frameworks has been thought to be beneficial, but empirical evidence indicates no visible improvements. In view of that, this paper proposes a strategy of performing CL along the dimensional direction instead of along the batch direction as done in conventional contrastive learning, named Dimensional Contrastive Learning (DimCL). DimCL aims to enhance the feature diversity, and it can serve as a regularizer to prior SSL frameworks. DimCL has been found to be effective, and the hardness-aware property is identified as a critical reason for its success. Extensive experimental results reveal that assimilating DimCL into SSL frameworks leads to performance improvement by a non-trivial margin on various datasets and backbone architectures.
Authors: Zixuan Yin, Han Sun, Ningzhong Liu, Huiyu Zhou, Jiaquan Shen
Lidars and cameras are critical sensors that provide complementary information for 3D detection in autonomous driving. While most prevalent methods progressively downscale the 3D point clouds and camera images and then fuse the high-level features, the downscaled features inevitably lose low-level detailed information. In this paper, we propose Fine-Grained Lidar-Camera Fusion (FGFusion) that make full use of multi-scale features of image and point cloud and fuse them in a fine-grained way. First, we design a dual pathway hierarchy structure to extract both high-level semantic and low-level detailed features of the image. Second, an auxiliary network is introduced to guide point cloud features to better learn the fine-grained spatial information. Finally, we propose multi-scale fusion (MSF) to fuse the last N feature maps of image and point cloud. Extensive experiments on two popular autonomous driving benchmarks, i.e. KITTI and Waymo, demonstrate the effectiveness of our method.
Authors: Abhijit Ramesh, Anantha Nandanan, Anantha Nandanan, Priya Nair MD, Gilad Gressel
Pancreatic cancer is a lethal form of cancer that significantly contributes to cancer-related deaths worldwide. Early detection is essential to improve patient prognosis and survival rates. Despite advances in medical imaging techniques, pancreatic cancer remains a challenging disease to detect. Endoscopic ultrasound (EUS) is the most effective diagnostic tool for detecting pancreatic cancer. However, it requires expert interpretation of complex ultrasound images to complete a reliable patient scan. To obtain complete imaging of the pancreas, practitioners must learn to guide the endoscope into multiple "EUS stations" (anatomical locations), which provide different views of the pancreas. This is a difficult skill to learn, involving over 225 proctored procedures with the support of an experienced doctor. We build an AI-assisted tool that utilizes deep learning techniques to identify these stations of the stomach in real time during EUS procedures. This computer-assisted diagnostic (CAD) will help train doctors more efficiently. Historically, the challenge faced in developing such a tool has been the amount of retrospective labeling required by trained clinicians. To solve this, we developed an open-source user-friendly labeling web app that streamlines the process of annotating stations during the EUS procedure with minimal effort from the clinicians. Our research shows that employing only 43 procedures with no hyperparameter fine-tuning obtained a balanced accuracy of 90%, comparable to the current state of the art. In addition, we employ Grad-CAM, a visualization technology that provides clinicians with interpretable and explainable visualizations.
Authors: Haozhi Cao, Yuecong Xu, Jianfei Yang, Pengyu Yin, Shenghai Yuan, Lihua Xie
Multi-modal unsupervised domain adaptation (MM-UDA) for 3D semantic segmentation is a practical solution to embed semantic understanding in autonomous systems without expensive point-wise annotations. While previous MM-UDA methods can achieve overall improvement, they suffer from significant class-imbalanced performance, restricting their adoption in real applications. This imbalanced performance is mainly caused by: 1) self-training with imbalanced data and 2) the lack of pixel-wise 2D supervision signals. In this work, we propose Multi-modal Prior Aided (MoPA) domain adaptation to improve the performance of rare objects. Specifically, we develop Valid Ground-based Insertion (VGI) to rectify the imbalance supervision signals by inserting prior rare objects collected from the wild while avoiding introducing artificial artifacts that lead to trivial solutions. Meanwhile, our SAM consistency loss leverages the 2D prior semantic masks from SAM as pixel-wise supervision signals to encourage consistent predictions for each object in the semantic mask. The knowledge learned from modal-specific prior is then shared across modalities to achieve better rare object segmentation. Extensive experiments show that our method achieves state-of-the-art performance on the challenging MM-UDA benchmark. Code will be available at https://github.com/AronCao49/MoPA.
Authors: Ting Jiang, Chuan Wang, Xinpeng Li, Ru Li, Haoqiang Fan, Shuaicheng Liu
In this paper, we introduce a new approach for high-quality multi-exposure image fusion (MEF). We show that the fusion weights of an exposure can be encoded into a 1D lookup table (LUT), which takes pixel intensity value as input and produces fusion weight as output. We learn one 1D LUT for each exposure, then all the pixels from different exposures can query 1D LUT of that exposure independently for high-quality and efficient fusion. Specifically, to learn these 1D LUTs, we involve attention mechanism in various dimensions including frame, channel and spatial ones into the MEF task so as to bring us significant quality improvement over the state-of-the-art (SOTA). In addition, we collect a new MEF dataset consisting of 960 samples, 155 of which are manually tuned by professionals as ground-truth for evaluation. Our network is trained by this dataset in an unsupervised manner. Extensive experiments are conducted to demonstrate the effectiveness of all the newly proposed components, and results show that our approach outperforms the SOTA in our and another representative dataset SICE, both qualitatively and quantitatively. Moreover, our 1D LUT approach takes less than 4ms to run a 4K image on a PC GPU. Given its high quality, efficiency and robustness, our method has been shipped into millions of Android mobiles across multiple brands world-wide. Code is available at: https://github.com/Hedlen/MEFLUT.
Authors: Haodong Ouyang
Recently, end-to-end object detectors have gained significant attention from the research community due to their outstanding performance. However, DETR typically relies on supervised pretraining of the backbone on ImageNet, which limits the practical application of DETR and the design of the backbone, affecting the model's potential generalization ability. In this paper, we propose a new training method called step-by-step training. Specifically, in the first stage, the one-to-many pre-trained YOLO detector is used to initialize the end-to-end detector. In the second stage, the backbone and encoder are consistent with the DETR-like model, but only the detector needs to be trained from scratch. Due to this training method, the object detector does not need the additional dataset (ImageNet) to train the backbone, which makes the design of the backbone more flexible and dramatically reduces the training cost of the detector, which is helpful for the practical application of the object detector. At the same time, compared with the DETR-like model, the step-by-step training method can achieve higher accuracy than the traditional training method of the DETR-like model. With the aid of this novel training method, we propose a brand-new end-to-end real-time object detection model called DEYOv3. DEYOv3-N achieves 41.1% on COCO val2017 and 270 FPS on T4 GPU, while DEYOv3-L achieves 51.3% AP and 102 FPS. Without the use of additional training data, DEYOv3 surpasses all existing real-time object detectors in terms of both speed and accuracy. It is worth noting that for models of N, S, and M scales, the training on the COCO dataset can be completed using a single 24GB RTX3090 GPU.
Authors: Junlong Li, Bingyao Yu, Yongming Rao, Jie Zhou, Jiwen Lu
In recent years, significant progress has been made in video instance segmentation (VIS), with many offline and online methods achieving state-of-the-art performance. While offline methods have the advantage of producing temporally consistent predictions, they are not suitable for real-time scenarios. Conversely, online methods are more practical, but maintaining temporal consistency remains a challenging task. In this paper, we propose a novel online method for video instance segmentation, called TCOVIS, which fully exploits the temporal information in a video clip. The core of our method consists of a global instance assignment strategy and a spatio-temporal enhancement module, which improve the temporal consistency of the features from two aspects. Specifically, we perform global optimal matching between the predictions and ground truth across the whole video clip, and supervise the model with the global optimal objective. We also capture the spatial feature and aggregate it with the semantic feature between frames, thus realizing the spatio-temporal enhancement. We evaluate our method on four widely adopted VIS benchmarks, namely YouTube-VIS 2019/2021/2022 and OVIS, and achieve state-of-the-art performance on all benchmarks without bells-and-whistles. For instance, on YouTube-VIS 2021, TCOVIS achieves 49.5 AP and 61.3 AP with ResNet-50 and Swin-L backbones, respectively. Code is available at https://github.com/jun-long-li/TCOVIS.
Authors: Zhisheng Wang, Zihan Deng, Fenglin Liu, Yixing Huang, Haijun Yu, Junning Cui
Recently, linear computed tomography (LCT) systems have actively attracted attention. To weaken projection truncation and image the region of interest (ROI) for LCT, the backprojection filtration (BPF) algorithm is an effective solution. However, in BPF for LCT, it is difficult to achieve stable interior reconstruction, and for differentiated backprojection (DBP) images of LCT, multiple rotation-finite inversion of Hilbert transform (Hilbert filtering)-inverse rotation operations will blur the image. To satisfy multiple reconstruction scenarios for LCT, including interior ROI, complete object, and exterior region beyond field-of-view (FOV), and avoid the rotation operations of Hilbert filtering, we propose two types of reconstruction architectures. The first overlays multiple DBP images to obtain a complete DBP image, then uses a network to learn the overlying Hilbert filtering function, referred to as the Overlay-Single Network (OSNet). The second uses multiple networks to train different directional Hilbert filtering models for DBP images of multiple linear scannings, respectively, and then overlays the reconstructed results, i.e., Multiple Networks Overlaying (MNetO). In two architectures, we introduce a Swin Transformer (ST) block to the generator of pix2pixGAN to extract both local and global features from DBP images at the same time. We investigate two architectures from different networks, FOV sizes, pixel sizes, number of projections, geometric magnification, and processing time. Experimental results show that two architectures can both recover images. OSNet outperforms BPF in various scenarios. For the different networks, ST-pix2pixGAN is superior to pix2pixGAN and CycleGAN. MNetO exhibits a few artifacts due to the differences among the multiple models, but any one of its models is suitable for imaging the exterior edge in a certain direction.
Authors: Shuang Zeng, Lei Zhu, Xinliang Zhang, Zifeng Tian, Qian Chen, Lujia Jin, Jiayi Wang, Yanye Lu
Contrastive learning, which is a powerful technique for learning image-level representations from unlabeled data, leads a promising direction to dealing with the dilemma between large-scale pre-training and limited labeled data. However, most existing contrastive learning strategies are designed mainly for downstream tasks of natural images, therefore they are sub-optimal and even worse than learning from scratch when directly applied to medical images whose downstream tasks are usually segmentation. In this work, we propose a novel asymmetric contrastive learning framework named JCL for medical image segmentation with self-supervised pre-training. Specifically, (1) A novel asymmetric contrastive learning strategy is proposed to pre-train both encoder and decoder simultaneously in one-stage to provide better initialization for segmentation models. (2) A multi-level contrastive loss is designed to take the correspondence among feature-level, image-level and pixel-level projections, respectively into account to make sure multi-level representations can be learned by the encoder and decoder during pre-training. (3) Experiments on multiple medical image datasets indicate our JCL framework outperforms existing SOTA contrastive learning strategies.
Authors: Vaibhav Mudgal, Qingyang Wang, Lorin Sweeney, Alan F. Smeaton
Video memorability is a measure of how likely a particular video is to be remembered by a viewer when that viewer has no emotional connection with the video content. It is an important characteristic as videos that are more memorable are more likely to be shared, viewed, and discussed. This paper presents results of a series of experiments where we improved the memorability of a video by selectively cropping frames based on image saliency. We present results of a basic fixed cropping as well as the results from dynamic cropping where both the size of the crop and the position of the crop within the frame, move as the video is played and saliency is tracked. Our results indicate that especially for videos of low initial memorability, the memorability score can be improved.
Authors: Zongqian Zhan, Rui Xia, Yifei Yu, Yibo Xu, Xin Wang
Over the last decades, ample achievements have been made on Structure from motion (SfM). However, the vast majority of them basically work in an offline manner, i.e., images are firstly captured and then fed together into a SfM pipeline for obtaining poses and sparse point cloud. In this work, on the contrary, we present an on-the-fly SfM: running online SfM while image capturing, the newly taken On-the-Fly image is online estimated with the corresponding pose and points, i.e., what you capture is what you get. Specifically, our approach firstly employs a vocabulary tree that is unsupervised trained using learning-based global features for fast image retrieval of newly fly-in image. Then, a robust feature matching mechanism with least squares (LSM) is presented to improve image registration performance. Finally, via investigating the influence of newly fly-in image's connected neighboring images, an efficient hierarchical weighted local bundle adjustment (BA) is used for optimization. Extensive experimental results demonstrate that on-the-fly SfM can meet the goal of robustly registering the images while capturing in an online way.
Authors: Aniket Jagtap, RamaKrishna Venkatesh Saripalli, Joe Lemley, Waseem Shariff, Alan F. Smeaton
Event cameras, also known as neuromorphic cameras, are an emerging technology that offer advantages over traditional shutter and frame-based cameras, including high temporal resolution, low power consumption, and selective data acquisition. In this study, we propose to harnesses the capabilities of event-based cameras to capture subtle changes in the surface of the skin caused by the pulsatile flow of blood in the wrist region. We investigate whether an event camera could be used for continuous noninvasive monitoring of heart rate (HR). Event camera video data from 25 participants, comprising varying age groups and skin colours, was collected and analysed. Ground-truth HR measurements obtained using conventional methods were used to evaluate of the accuracy of automatic detection of HR from event camera data. Our experimental results and comparison to the performance of other non-contact HR measurement methods demonstrate the feasibility of using event cameras for pulse detection. We also acknowledge the challenges and limitations of our method, such as light-induced flickering and the sub-conscious but naturally-occurring tremors of an individual during data capture.
Authors: Sylwia Majchrowska, Anders Hildeman, Philip Teare, Tom Diethe
Supervised training of deep learning models for medical imaging applications requires a significant amount of labeled data. This is posing a challenge as the images are required to be annotated by medical professionals. To address this limitation, we introduce the Adaptive Locked Agnostic Network (ALAN), a concept involving self-supervised visual feature extraction using a large backbone model to produce anatomically robust semantic self-segmentation. In the ALAN methodology, this self-supervised training occurs only once on a large and diverse dataset. Due to the intuitive interpretability of the segmentation, downstream models tailored for specific tasks can be easily designed using white-box models with few parameters. This, in turn, opens up the possibility of communicating the inner workings of a model with domain experts and introducing prior knowledge into it. It also means that the downstream models become less data-hungry compared to fully supervised approaches. These characteristics make ALAN particularly well-suited for resource-scarce scenarios, such as costly clinical trials and rare diseases. In this paper, we apply the ALAN approach to three publicly available echocardiography datasets: EchoNet-Dynamic, CAMUS, and TMED-2. Our findings demonstrate that the self-supervised backbone model robustly identifies anatomical subregions of the heart in an apical four-chamber view. Building upon this, we design two downstream models, one for segmenting a target anatomical region, and a second for echocardiogram view classification.
Authors: Yanbo Gao, Wenjia Huang, Shuai Li, Hui Yuan, Mao Ye, Siwei Ma
Learned video compression (LVC) has witnessed remarkable advancements in recent years. Similar as the traditional video coding, LVC inherits motion estimation/compensation, residual coding and other modules, all of which are implemented with neural networks (NNs). However, within the framework of NNs and its training mechanism using gradient backpropagation, most existing works often struggle to consistently generate stable motion information, which is in the form of geometric features, from the input color features. Moreover, the modules such as the inter-prediction and residual coding are independent from each other, making it inefficient to fully reduce the spatial-temporal redundancy. To address the above problems, in this paper, we propose a novel Spatial-Temporal Transformer based Video Compression (STT-VC) framework. It contains a Relaxed Deformable Transformer (RDT) with Uformer based offsets estimation for motion estimation and compensation, a Multi-Granularity Prediction (MGP) module based on multi-reference frames for prediction refinement, and a Spatial Feature Distribution prior based Transformer (SFD-T) for efficient temporal-spatial joint residual compression. Specifically, RDT is developed to stably estimate the motion information between frames by thoroughly investigating the relationship between the similarity based geometric motion feature extraction and self-attention. MGP is designed to fuse the multi-reference frame information by effectively exploring the coarse-grained prediction feature generated with the coded motion information. SFD-T is to compress the residual information by jointly exploring the spatial feature distributions in both residual and temporal prediction to further reduce the spatial-temporal redundancy. Experimental results demonstrate that our method achieves the best result with 13.5% BD-Rate saving over VTM.
Authors: Xiaozhou You, Jian Zhang
Text-guided image generation aimed to generate desired images conditioned on given texts, while text-guided image manipulation refers to semantically edit parts of a given image based on specified texts. For these two similar tasks, the key point is to ensure image fidelity as well as semantic consistency. Many previous approaches require complex multi-stage generation and adversarial training, while struggling to provide a unified framework for both tasks. In this work, we propose TextCLIP, a unified framework for text-guided image generation and manipulation without adversarial training. The proposed method accepts input from images or random noise corresponding to these two different tasks, and under the condition of the specific texts, a carefully designed mapping network that exploits the powerful generative capabilities of StyleGAN and the text image representation capabilities of Contrastive Language-Image Pre-training (CLIP) generates images of up to $1024\times1024$ resolution that can currently be generated. Extensive experiments on the Multi-modal CelebA-HQ dataset have demonstrated that our proposed method outperforms existing state-of-the-art methods, both on text-guided generation tasks and manipulation tasks.
Authors: Lukáš Korel, Petr Pulc, Jiří Tumpach, Martin Holeňa
This paper provides an insight into the possibility of scene recognition from a video sequence with a small set of repeated shooting locations (such as in television series) using artificial neural networks. The basic idea of the presented approach is to select a set of frames from each scene, transform them by a pre-trained singleimage pre-processing convolutional network, and classify the scene location with subsequent layers of the neural network. The considered networks have been tested and compared on a dataset obtained from The Big Bang Theory television series. We have investigated different neural network layers to combine individual frames, particularly AveragePooling, MaxPooling, Product, Flatten, LSTM, and Bidirectional LSTM layers. We have observed that only some of the approaches are suitable for the task at hand.
Authors: Bo Ye, Kai Gan, Tong Wei, Min-Ling Zhang
In open-world semi-supervised learning, a machine learning model is tasked with uncovering novel categories from unlabeled data while maintaining performance on seen categories from labeled data. The central challenge is the substantial learning gap between seen and novel categories, as the model learns the former faster due to accurate supervisory information. To address this, we introduce 1) an adaptive margin loss based on estimated class distribution, which encourages a large negative margin for samples in seen classes, to synchronize learning paces, and 2) pseudo-label contrastive clustering, which pulls together samples which are likely from the same class in the output space, to enhance novel class discovery. Our extensive evaluations on multiple datasets demonstrate that existing models still hinder novel class learning, whereas our approach strikingly balances both seen and novel classes, achieving a remarkable 3% average accuracy increase on the ImageNet dataset compared to the prior state-of-the-art. Additionally, we find that fine-tuning the self-supervised pre-trained backbone significantly boosts performance over the default in prior literature. After our paper is accepted, we will release the code.
Authors: Ping Li, Yu Zhang, Li Yuan, Xianghua Xu
Referring Video Object Segmentation (RVOS) requires segmenting the object in video referred by a natural language query. Existing methods mainly rely on sophisticated pipelines to tackle such cross-modal task, and do not explicitly model the object-level spatial context which plays an important role in locating the referred object. Therefore, we propose an end-to-end RVOS framework completely built upon transformers, termed \textit{Fully Transformer-Equipped Architecture} (FTEA), which treats the RVOS task as a mask sequence learning problem and regards all the objects in video as candidate objects. Given a video clip with a text query, the visual-textual features are yielded by encoder, while the corresponding pixel-level and word-level features are aligned in terms of semantic similarity. To capture the object-level spatial context, we have developed the Stacked Transformer, which individually characterizes the visual appearance of each candidate object, whose feature map is decoded to the binary mask sequence in order directly. Finally, the model finds the best matching between mask sequence and text query. In addition, to diversify the generated masks for candidate objects, we impose a diversity loss on the model for capturing more accurate mask of the referred object. Empirical studies have shown the superiority of the proposed method on three benchmarks, e.g., FETA achieves 45.1% and 38.7% in terms of mAP on A2D Sentences (3782 videos) and J-HMDB Sentences (928 videos), respectively; it achieves 56.6% in terms of $\mathcal{J\&F}$ on Ref-YouTube-VOS (3975 videos and 7451 objects). Particularly, compared to the best candidate method, it has a gain of 2.1% and 3.2% in terms of P$@$0.5 on the former two, respectively, while it has a gain of 2.9% in terms of $\mathcal{J}$ on the latter one.
Authors: Jonas Brenig, Radu Timofte
Self-supervised representation learning has seen remarkable progress in the last few years, with some of the recent methods being able to learn useful image representations without labels. These methods are trained using backpropagation, the de facto standard. Recently, Geoffrey Hinton proposed the forward-forward algorithm as an alternative training method. It utilizes two forward passes and a separate loss function for each layer to train the network without backpropagation.
In this study, for the first time, we study the performance of forward-forward vs. backpropagation for self-supervised representation learning and provide insights into the learned representation spaces. Our benchmark employs four standard datasets, namely MNIST, F-MNIST, SVHN and CIFAR-10, and three commonly used self-supervised representation learning techniques, namely rotation, flip and jigsaw.
Our main finding is that while the forward-forward algorithm performs comparably to backpropagation during (self-)supervised training, the transfer performance is significantly lagging behind in all the studied settings. This may be caused by a combination of factors, including having a loss function for each layer and the way the supervised training is realized in the forward-forward paradigm. In comparison to backpropagation, the forward-forward algorithm focuses more on the boundaries and drops part of the information unnecessary for making decisions which harms the representation learning goal. Further investigation and research are necessary to stabilize the forward-forward strategy for self-supervised learning, to work beyond the datasets and configurations demonstrated by Geoffrey Hinton.
Authors: Taeho Kang, Kyungjin Lee, Jinrui Zhang, Youngki Lee
We present Ego3DPose, a highly accurate binocular egocentric 3D pose reconstruction system. The binocular egocentric setup offers practicality and usefulness in various applications, however, it remains largely under-explored. It has been suffering from low pose estimation accuracy due to viewing distortion, severe self-occlusion, and limited field-of-view of the joints in egocentric 2D images. Here, we notice that two important 3D cues, stereo correspondences, and perspective, contained in the egocentric binocular input are neglected. Current methods heavily rely on 2D image features, implicitly learning 3D information, which introduces biases towards commonly observed motions and leads to low overall accuracy. We observe that they not only fail in challenging occlusion cases but also in estimating visible joint positions. To address these challenges, we propose two novel approaches. First, we design a two-path network architecture with a path that estimates pose per limb independently with its binocular heatmaps. Without full-body information provided, it alleviates bias toward trained full-body distribution. Second, we leverage the egocentric view of body limbs, which exhibits strong perspective variance (e.g., a significantly large-size hand when it is close to the camera). We propose a new perspective-aware representation using trigonometry, enabling the network to estimate the 3D orientation of limbs. Finally, we develop an end-to-end pose reconstruction network that synergizes both techniques. Our comprehensive evaluations demonstrate that Ego3DPose outperforms state-of-the-art models by a pose estimation error (i.e., MPJPE) reduction of 23.1% in the UnrealEgo dataset. Our qualitative results highlight the superiority of our approach across a range of scenarios and challenges.
Authors: Floris Erich, Naoya Chiba, Yusuke Yoshiyasu, Noriaki Ando, Ryo Hanai, Yukiyasu Domae
We present NeuralLabeling, a labeling approach and toolset for annotating a scene using either bounding boxes or meshes and generating segmentation masks, affordance maps, 2D bounding boxes, 3D bounding boxes, 6DOF object poses, depth maps and object meshes. NeuralLabeling uses Neural Radiance Fields (NeRF) as renderer, allowing labeling to be performed using 3D spatial tools while incorporating geometric clues such as occlusions, relying only on images captured from multiple viewpoints as input. To demonstrate the applicability of NeuralLabeling to a practical problem in robotics, we added ground truth depth maps to 30000 frames of transparent object RGB and noisy depth maps of glasses placed in a dishwasher captured using an RGBD sensor, yielding the Dishwasher30k dataset. We show that training a simple deep neural network with supervision using the annotated depth maps yields a higher reconstruction performance than training with the previously applied weakly supervised approach.
Authors: Philipp Ausserlechner, David Haberger, Stefan Thalhammer, Jean-Baptiste Weibel, Markus Vincze
As robotic systems increasingly encounter complex and unconstrained real-world scenarios, there is a demand to recognize diverse objects. The state-of-the-art 6D object pose estimation methods rely on object-specific training and therefore do not generalize to unseen objects. Recent novel object pose estimation methods are solving this issue using task-specific fine-tuned CNNs for deep template matching. This adaptation for pose estimation still requires expensive data rendering and training procedures. MegaPose for example is trained on a dataset consisting of two million images showing 20,000 different objects to reach such generalization capabilities. To overcome this shortcoming we introduce ZS6D, for zero-shot novel object 6D pose estimation. Visual descriptors, extracted using pre-trained Vision Transformers (ViT), are used for matching rendered templates against query images of objects and for establishing local correspondences. These local correspondences enable deriving geometric correspondences and are used for estimating the object's 6D pose with RANSAC-based PnP. This approach showcases that the image descriptors extracted by pre-trained ViTs are well-suited to achieve a notable improvement over two state-of-the-art novel object 6D pose estimation methods, without the need for task-specific fine-tuning. Experiments are performed on LMO, YCBV, and TLESS. In comparison to one of the two methods we improve the Average Recall on all three datasets and compared to the second method we improve on two datasets.
Authors: Rajitha de Silva, Grzegorz Cielniak, Junfeng Gao
Vision-based mobile robot navigation systems in arable fields are mostly limited to in-row navigation. The process of switching from one crop row to the next in such systems is often aided by GNSS sensors or multiple camera setups. This paper presents a novel vision-based crop row-switching algorithm that enables a mobile robot to navigate an entire field of arable crops using a single front-mounted camera. The proposed row-switching manoeuvre uses deep learning-based RGB image segmentation and depth data to detect the end of the crop row, and re-entry point to the next crop row which would be used in a multi-state row switching pipeline. Each state of this pipeline use visual feedback or wheel odometry of the robot to successfully navigate towards the next crop row. The proposed crop row navigation pipeline was tested in a real sugar beet field containing crop rows with discontinuities, varying light levels, shadows and irregular headland surfaces. The robot could successfully exit from one crop row and re-enter the next crop row using the proposed pipeline with absolute median errors averaging at 19.25 cm and 6.77{\deg} for linear and rotational steps of the proposed manoeuvre.
Authors: Silvia Sellán, Alec Jacobson
Reconstructing a surface from a point cloud is an underdetermined problem. We use a neural network to study and quantify this reconstruction uncertainty under a Poisson smoothness prior. Our algorithm addresses the main limitations of existing work and can be fully integrated into the 3D scanning pipeline, from obtaining an initial reconstruction to deciding on the next best sensor position and updating the reconstruction upon capturing more data.
Authors: Eduardo Augusto Roeder
Pneumonia is the leading infectious cause of infant death in the world. When identified early, it is possible to alter the prognosis of the patient, one could use imaging exams to help in the diagnostic confirmation. Performing and interpreting the exams as soon as possible is vital for a good treatment, with the most common exam for this pathology being chest X-ray. The objective of this study was to develop a software that identify the presence or absence of pneumonia in chest radiographs. The software was developed as a computational model based on machine learning using transfer learning technique. For the training process, images were collected from a database available online with children's chest X-rays images taken at a hospital in China. After training, the model was then exposed to new images, achieving relevant results on identifying such pathology, reaching 98% sensitivity and 97.3% specificity for the sample used for testing. It can be concluded that it is possible to develop a software that identifies pneumonia in chest X-ray images.
Authors: Yiping Wei, Kunyu Peng, Alina Roitberg, Jiaming Zhang, Junwei Zheng, Ruiping Liu, Yufan Chen, Kailun Yang, Rainer Stiefelhagen
Self-supervised representation learning for human action recognition has developed rapidly in recent years. Most of the existing works are based on skeleton data while using a multi-modality setup. These works overlooked the differences in performance among modalities, which led to the propagation of erroneous knowledge between modalities while only three fundamental modalities, i.e., joints, bones, and motions are used, hence no additional modalities are explored.
In this work, we first propose an Implicit Knowledge Exchange Module (IKEM) which alleviates the propagation of erroneous knowledge between low-performance modalities. Then, we further propose three new modalities to enrich the complementary information between modalities. Finally, to maintain efficiency when introducing new modalities, we propose a novel teacher-student framework to distill the knowledge from the secondary modalities into the mandatory modalities considering the relationship constrained by anchors, positives, and negatives, named relational cross-modality knowledge distillation. The experimental results demonstrate the effectiveness of our approach, unlocking the efficient use of skeleton-based multi-modality data. Source code will be made publicly available at https://github.com/desehuileng0o0/IKEM.
Authors: Haopeng Zhang, Zijing Lin, Feng Gao, Junyu Dong, Qian Du, Heng-Chao Li
Synthetic aperture radar (SAR) image change detection is a critical task and has received increasing attentions in the remote sensing community. However, existing SAR change detection methods are mainly based on convolutional neural networks (CNNs), with limited consideration of global attention mechanism. In this letter, we explore Transformer-like architecture for SAR change detection to incorporate global attention. To this end, we propose a convolution and attention mixer (CAMixer). First, to compensate the inductive bias for Transformer, we combine self-attention with shift convolution in a parallel way. The parallel design effectively captures the global semantic information via the self-attention and performs local feature extraction through shift convolution simultaneously. Second, we adopt a gating mechanism in the feed-forward network to enhance the non-linear feature transformation. The gating mechanism is formulated as the element-wise multiplication of two parallel linear layers. Important features can be highlighted, leading to high-quality representations against speckle noise. Extensive experiments conducted on three SAR datasets verify the superior performance of the proposed CAMixer. The source codes will be publicly available at https://github.com/summitgao/CAMixer .
Authors: Utsav Kumar Nareti, Chandranath Adak, Soumi Chattopadhyay
In the film industry, movie posters have been an essential part of advertising and marketing for many decades, and continue to play a vital role even today in the form of digital posters through online, social media and OTT platforms. Typically, movie posters can effectively promote and communicate the essence of a film, such as its genre, visual style/ tone, vibe and storyline cue/ theme, which are essential to attract potential viewers. Identifying the genres of a movie often has significant practical applications in recommending the film to target audiences. Previous studies on movie genre identification are limited to subtitles, plot synopses, and movie scenes that are mostly accessible after the movie release. Posters usually contain pre-release implicit information to generate mass interest. In this paper, we work for automated multi-label genre identification only from movie poster images, without any aid of additional textual/meta-data information about movies, which is one of the earliest attempts of its kind. Here, we present a deep transformer network with a probabilistic module to identify the movie genres exclusively from the poster. For experimental analysis, we procured 13882 number of posters of 13 genres from the Internet Movie Database (IMDb), where our model performances were encouraging and even outperformed some major contemporary architectures.
Authors: Muhammad Sulaiman, Mina Farmanbar, Ahmed Nabil Belbachir, Chunming Rong
Building segmentation is essential in infrastructure development, population management, and geological observations. This article targets shallow models due to their interpretable nature to assess the presence of LiDAR data for supervised segmentation. The benchmark data used in this article are published in NORA MapAI competition for deep learning model. Shallow models are compared with deep learning models based on Intersection over Union (IoU) and Boundary Intersection over Union (BIoU). In the proposed work, boundary masks from the original mask are generated to improve the BIoU score, which relates to building shapes' borderline. The influence of LiDAR data is tested by training the model with only aerial images in task 1 and a combination of aerial and LiDAR data in task 2 and then compared. shallow models outperform deep learning models in IoU by 8% using aerial images (task 1) only and 2% in combined aerial images and LiDAR data (task 2). In contrast, deep learning models show better performance on BIoU scores. Boundary masks improve BIoU scores by 4% in both tasks. Light Gradient-Boosting Machine (LightGBM) performs better than RF and Extreme Gradient Boosting (XGBoost).
Authors: Yifei Chen, Kunyu Peng, Alina Roitberg, David Schneider, Jiaming Zhang, Junwei Zheng, Ruiping Liu, Yufan Chen, Kailun Yang, Rainer Stiefelhagen
To integrate action recognition methods into autonomous robotic systems, it is crucial to consider adverse situations involving target occlusions. Such a scenario, despite its practical relevance, is rarely addressed in existing self-supervised skeleton-based action recognition methods. To empower robots with the capacity to address occlusion, we propose a simple and effective method. We first pre-train using occluded skeleton sequences, then use k-means clustering (KMeans) on sequence embeddings to group semantically similar samples. Next, we employ K-nearest-neighbor (KNN) to fill in missing skeleton data based on the closest sample neighbors. Imputing incomplete skeleton sequences to create relatively complete sequences as input provides significant benefits to existing skeleton-based self-supervised models. Meanwhile, building on the state-of-the-art Partial Spatio-Temporal Learning (PSTL), we introduce an Occluded Partial Spatio-Temporal Learning (OPSTL) framework. This enhancement utilizes Adaptive Spatial Masking (ASM) for better use of high-quality, intact skeletons. The effectiveness of our imputation methods is verified on the challenging occluded versions of the NTURGB+D 60 and NTURGB+D 120. The source code will be made publicly available at https://github.com/cyfml/OPSTL.
Authors: Adrian Suwała, Bartosz Wójcik, Magdalena Proszewska, Jacek Tabor, Przemysław Spurek, Marek Śmieja
Conditional GANs are frequently used for manipulating the attributes of face images, such as expression, hairstyle, pose, or age. Even though the state-of-the-art models successfully modify the requested attributes, they simultaneously modify other important characteristics of the image, such as a person's identity. In this paper, we focus on solving this problem by introducing PluGeN4Faces, a plugin to StyleGAN, which explicitly disentangles face attributes from a person's identity. Our key idea is to perform training on images retrieved from movie frames, where a given person appears in various poses and with different attributes. By applying a type of contrastive loss, we encourage the model to group images of the same person in similar regions of latent space. Our experiments demonstrate that the modifications of face attributes performed by PluGeN4Faces are significantly less invasive on the remaining characteristics of the image than in the existing state-of-the-art models.
Authors: Martin Vonheim Larsen, Sigmund Rolfsjord, Daniel Gusland, Jörgen Ahlberg, Kim Mathiassen
The field of visual object tracking is dominated by methods that combine simple tracking algorithms and ad hoc schemes. Probabilistic tracking algorithms, which are leading in other fields, are surprisingly absent from the leaderboards. We found that accounting for distance in target kinematics, exploiting detector confidence and modelling non-uniform clutter characteristics is critical for a probabilistic tracker to work in visual tracking. Previous probabilistic methods fail to address most or all these aspects, which we believe is why they fall so far behind current state-of-the-art (SOTA) methods (there are no probabilistic trackers in the MOT17 top 100). To rekindle progress among probabilistic approaches, we propose a set of pragmatic models addressing these challenges, and demonstrate how they can be incorporated into a probabilistic framework. We present BASE (Bayesian Approximation Single-hypothesis Estimator), a simple, performant and easily extendible visual tracker, achieving state-of-the-art (SOTA) on MOT17 and MOT20, without using Re-Id. Code will be made available at https://github.com/ffi-no
Authors: Xiaoyu Liu, Ming Liu, Junyi Li, Shuai Liu, Xiaotao Wang, Lei Lei, Wangmeng Zuo
For improving image composition and aesthetic quality, most existing methods modulate the captured images by striking out redundant content near the image borders. However, such image cropping methods are limited in the range of image views. Some methods have been suggested to extrapolate the images and predict cropping boxes from the extrapolated image. Nonetheless, the synthesized extrapolated regions may be included in the cropped image, making the image composition result not real and potentially with degraded image quality. In this paper, we circumvent this issue by presenting a joint framework for both unbounded recommendation of camera view and image composition (i.e., UNIC). In this way, the cropped image is a sub-image of the image acquired by the predicted camera view, and thus can be guaranteed to be real and consistent in image quality. Specifically, our framework takes the current camera preview frame as input and provides a recommendation for view adjustment, which contains operations unlimited by the image borders, such as zooming in or out and camera movement. To improve the prediction accuracy of view adjustment prediction, we further extend the field of view by feature extrapolation. After one or several times of view adjustments, our method converges and results in both a camera view and a bounding box showing the image composition recommendation. Extensive experiments are conducted on the datasets constructed upon existing image cropping datasets, showing the effectiveness of our UNIC in unbounded recommendation of camera view and image composition. The source code, dataset, and pretrained models is available at https://github.com/liuxiaoyu1104/UNIC.
Authors: Kiseok Choi, Inchul Kim, Dongyoung Choi, Julio Marco, Diego Gutierrez, Min H. Kim
Existing time-resolved non-line-of-sight (NLOS) imaging methods reconstruct hidden scenes by inverting the optical paths of indirect illumination measured at visible relay surfaces. These methods are prone to reconstruction artifacts due to inversion ambiguities and capture noise, which are typically mitigated through the manual selection of filtering functions and parameters. We introduce a fully-differentiable end-to-end NLOS inverse rendering pipeline that self-calibrates the imaging parameters during the reconstruction of hidden scenes, using as input only the measured illumination while working both in the time and frequency domains. Our pipeline extracts a geometric representation of the hidden scene from NLOS volumetric intensities and estimates the time-resolved illumination at the relay wall produced by such geometric information using differentiable transient rendering. We then use gradient descent to optimize imaging parameters by minimizing the error between our simulated time-resolved illumination and the measured illumination. Our end-to-end differentiable pipeline couples diffraction-based volumetric NLOS reconstruction with path-space light transport and a simple ray marching technique to extract detailed, dense sets of surface points and normals of hidden scenes. We demonstrate the robustness of our method to consistently reconstruct geometry and albedo, even under significant noise levels.
Authors: Karolina Seweryn, Anna Wróblewska, Szymon Łukasik
Action scene understanding in soccer is a challenging task due to the complex and dynamic nature of the game, as well as the interactions between players. This article provides a comprehensive overview of this task divided into action recognition, spotting, and spatio-temporal action localization, with a particular emphasis on the modalities used and multimodal methods. We explore the publicly available data sources and metrics used to evaluate models' performance. The article reviews recent state-of-the-art methods that leverage deep learning techniques and traditional methods. We focus on multimodal methods, which integrate information from multiple sources, such as video and audio data, and also those that represent one source in various ways. The advantages and limitations of methods are discussed, along with their potential for improving the accuracy and robustness of models. Finally, the article highlights some of the open research questions and future directions in the field of soccer action recognition, including the potential for multimodal methods to advance this field. Overall, this survey provides a valuable resource for researchers interested in the field of action scene understanding in soccer.
Authors: Fuping Wu, Le Zhang, Yang Sun, Yuanhan Mo, Thomas Nichols, Bartlomiej W. Papiez
Multi-task learning (MTL) has shown great potential in medical image analysis, improving the generalizability of the learned features and the performance in individual tasks. However, most of the work on MTL focuses on either architecture design or gradient manipulation, while in both scenarios, features are learned in a competitive manner. In this work, we propose to formulate MTL as a multi/bi-level optimization problem, and therefore force features to learn from each task in a cooperative approach. Specifically, we update the sub-model for each task alternatively taking advantage of the learned sub-models of the other tasks. To alleviate the negative transfer problem during the optimization, we search for flat minima for the current objective function with regard to features from other tasks. To demonstrate the effectiveness of the proposed approach, we validate our method on three publicly available datasets. The proposed method shows the advantage of cooperative learning, and yields promising results when compared with the state-of-the-art MTL approaches. The code will be available online.
Authors: Dimitrije Marković, Karl J. Friston, Stefan J. Kiebel
Deep learning's immense capabilities are often constrained by the complexity of its models, leading to an increasing demand for effective sparsification techniques. Bayesian sparsification for deep learning emerges as a crucial approach, facilitating the design of models that are both computationally efficient and competitive in terms of performance across various deep learning applications. The state-of-the-art -- in Bayesian sparsification of deep neural networks -- combines structural shrinkage priors on model weights with an approximate inference scheme based on black-box stochastic variational inference. However, model inversion of the full generative model is exceptionally computationally demanding, especially when compared to standard deep learning of point estimates. In this context, we advocate for the use of Bayesian model reduction (BMR) as a more efficient alternative for pruning of model weights. As a generalization of the Savage-Dickey ratio, BMR allows a post-hoc elimination of redundant model weights based on the posterior estimates under a straightforward (non-hierarchical) generative model. Our comparative study highlights the computational efficiency and the pruning rate of the BMR method relative to the established stochastic variational inference (SVI) scheme, when applied to the full hierarchical generative model. We illustrate the potential of BMR to prune model parameters across various deep learning architectures, from classical networks like LeNet to modern frameworks such as Vision Transformers and MLP-Mixers.
Authors: Mehmet Bahadir Erden, Selahattin Cansiz, Onur Caki, Haya Khattak, Durmus Etiz, Melek Cosar Yakar, Kerem Duruer, Berke Barut, Cigdem Gunduz-Demir
Encoder-decoder networks become a popular choice for various medical image segmentation tasks. When they are trained with a standard loss function, these networks are not explicitly enforced to preserve the shape integrity of an object in an image. However, this ability of the network is important to obtain more accurate results, especially when there is a low-contrast difference between the object and its surroundings. In response to this issue, this work introduces a new shape-aware loss function, which we name FourierLoss. This loss function relies on quantifying the shape dissimilarity between the ground truth and the predicted segmentation maps through the Fourier descriptors calculated on their objects, and penalizing this dissimilarity in network training. Different than the previous studies, FourierLoss offers an adaptive loss function with trainable hyperparameters that control the importance of the level of the shape details that the network is enforced to learn in the training process. This control is achieved by the proposed adaptive loss update mechanism, which end-to-end learns the hyperparameters simultaneously with the network weights by backpropagation. As a result of using this mechanism, the network can dynamically change its attention from learning the general outline of an object to learning the details of its contour points, or vice versa, in different training epochs. Working on 2879 computed tomography images of 93 subjects, our experiments revealed that the proposed adaptive shape-aware loss function led to statistically significantly better results for liver segmentation, compared to its counterparts.
Authors: Alberto Baldrati, Marco Bertini, Tiberio Uricchio, Alberto Del Bimbo
Given the recent advances in multimodal image pretraining where visual models trained with semantically dense textual supervision tend to have better generalization capabilities than those trained using categorical attributes or through unsupervised techniques, in this work we investigate how recent CLIP model can be applied in several tasks in artwork domain. We perform exhaustive experiments on the NoisyArt dataset which is a dataset of artwork images crawled from public resources on the web. On such dataset CLIP achieves impressive results on (zero-shot) classification and promising results in both artwork-to-artwork and description-to-artwork domain.
Authors: Matthias Hadlich, Zdravko Marinov, Rainer Stiefelhagen
Tumor segmentation in medical imaging is crucial and relies on precise delineation. Fluorodeoxyglucose Positron-Emission Tomography (FDG-PET) is widely used in clinical practice to detect metabolically active tumors. However, FDG-PET scans may misinterpret irregular glucose consumption in healthy or benign tissues as cancer. Combining PET with Computed Tomography (CT) can enhance tumor segmentation by integrating metabolic and anatomic information. FDG-PET/CT scans are pivotal for cancer staging and reassessment, utilizing radiolabeled fluorodeoxyglucose to highlight metabolically active regions. Accurately distinguishing tumor-specific uptake from physiological uptake in normal tissues is a challenging aspect of precise tumor segmentation. The AutoPET challenge addresses this by providing a dataset of 1014 FDG-PET/CT studies, encouraging advancements in accurate tumor segmentation and analysis within the FDG-PET/CT domain. Code: https://github.com/matt3o/AutoPET2-Submission/
Authors: Sanjeet Vardam, Luuk Spreeuwers
In recent years face recognition systems have been brought to the mainstream due to development in hardware and software. Consistent efforts are being made to make them better and more secure. This has also brought developments in 3D face recognition systems at a rapid pace. These 3DFR systems are expected to overcome certain vulnerabilities of 2DFR systems. One such problem that the domain of 2DFR systems face is face image morphing. A substantial amount of research is being done for generation of high quality face morphs along with detection of attacks from these morphs. Comparatively the understanding of vulnerability of 3DFR systems against 3D face morphs is less. But at the same time an expectation is set from 3DFR systems to be more robust against such attacks. This paper attempts to research and gain more information on this matter. The paper describes a couple of methods that can be used to generate 3D face morphs. The face morphs that are generated using this method are then compared to the contributing faces to obtain similarity scores. The highest MMPMR is obtained around 40% with RMMR of 41.76% when 3DFRS are attacked with look-a-like morphs.
Authors: Travis Zhang, Katie Luo, Cheng Perng Phoo, Yurong You, Wei-Lun Chao, Bharath Hariharan, Mark Campbell, Kilian Q. Weinberger
The rapid development of 3D object detection systems for self-driving cars has significantly improved accuracy. However, these systems struggle to generalize across diverse driving environments, which can lead to safety-critical failures in detecting traffic participants. To address this, we propose a method that utilizes unlabeled repeated traversals of multiple locations to adapt object detectors to new driving environments. By incorporating statistics computed from repeated LiDAR scans, we guide the adaptation process effectively. Our approach enhances LiDAR-based detection models using spatial quantized historical features and introduces a lightweight regression head to leverage the statistics for feature regularization. Additionally, we leverage the statistics for a novel self-training process to stabilize the training. The framework is detector model-agnostic and experiments on real-world datasets demonstrate significant improvements, achieving up to a 20-point performance gain, especially in detecting pedestrians and distant objects. Code is available at https://github.com/zhangtravis/Hist-DA.
Authors: Mauro Barni, Patrizio Campisi, Edward J. Delp, Gwenael Doërr, Jessica Fridrich, Nasir Memon, Fernando Pérez-González, Anderson Rocha, Luisa Verdoliva, Min Wu
Information Forensics and Security (IFS) is an active R&D area whose goal is to ensure that people use devices, data, and intellectual properties for authorized purposes and to facilitate the gathering of solid evidence to hold perpetrators accountable. For over a quarter century since the 1990s, the IFS research area has grown tremendously to address the societal needs of the digital information era. The IEEE Signal Processing Society (SPS) has emerged as an important hub and leader in this area, and the article below celebrates some landmark technical contributions. In particular, we highlight the major technological advances on some selected focus areas in the field developed in the last 25 years from the research community and present future trends.
Authors: Sagar M. Waghmare, Kimberly Wilber, Dave Hawkey, Xuan Yang, Matthew Wilson, Stephanie Debats, Cattalyya Nuengsigkapian, Astuti Sharma, Lars Pandikow, Huisheng Wang, Hartwig Adam, Mikhail Sirotenko
We introduce SANPO, a large-scale egocentric video dataset focused on dense prediction in outdoor environments. It contains stereo video sessions collected across diverse outdoor environments, as well as rendered synthetic video sessions. (Synthetic data was provided by Parallel Domain.) All sessions have (dense) depth and odometry labels. All synthetic sessions and a subset of real sessions have temporally consistent dense panoptic segmentation labels. To our knowledge, this is the first human egocentric video dataset with both large scale dense panoptic segmentation and depth annotations. In addition to the dataset we also provide zero-shot baselines and SANPO benchmarks for future research. We hope that the challenging nature of SANPO will help advance the state-of-the-art in video segmentation, depth estimation, multi-task visual modeling, and synthetic-to-real domain adaptation, while enabling human navigation systems.
SANPO is available here: https://google-research-datasets.github.io/sanpo_dataset/
Authors: Eui Jun Hwang, Huije Lee, Jong C. Park
Gloss-free Sign Language Production (SLP) offers a direct translation of spoken language sentences into sign language, bypassing the need for gloss intermediaries. This paper presents the Sign language Vector Quantization Network, a novel approach to SLP that leverages Vector Quantization to derive discrete representations from sign pose sequences. Our method, rooted in both manual and non-manual elements of signing, supports advanced decoding methods and integrates latent-level alignment for enhanced linguistic coherence. Through comprehensive evaluations, we demonstrate superior performance of our method over prior SLP methods and highlight the reliability of Back-Translation and Fr\'echet Gesture Distance as evaluation metrics.
Authors: Yu Cheng, Bo Wang, Robby T. Tan
In 3D human shape and pose estimation from a monocular video, models trained with limited labeled data cannot generalize well to videos with occlusion, which is common in the wild videos. The recent human neural rendering approaches focusing on novel view synthesis initialized by the off-the-shelf human shape and pose methods have the potential to correct the initial human shape. However, the existing methods have some drawbacks such as, erroneous in handling occlusion, sensitive to inaccurate human segmentation, and ineffective loss computation due to the non-regularized opacity field. To address these problems, we introduce ORTexME, an occlusion-robust temporal method that utilizes temporal information from the input video to better regularize the occluded body parts. While our ORTexME is based on NeRF, to determine the reliable regions for the NeRF ray sampling, we utilize our novel average texture learning approach to learn the average appearance of a person, and to infer a mask based on the average texture. In addition, to guide the opacity-field updates in NeRF to suppress blur and noise, we propose the use of human body mesh. The quantitative evaluation demonstrates that our method achieves significant improvement on the challenging multi-person 3DPW dataset, where our method achieves 1.8 P-MPJPE error reduction. The SOTA rendering-based methods fail and enlarge the error up to 5.6 on the same dataset.
Authors: Guangyao Zhai, Xiaoni Cai, Dianye Huang, Yan Di, Fabian Manhardt, Federico Tombari, Nassir Navab, Benjamin Busam
Object rearrangement is pivotal in robotic-environment interactions, representing a significant capability in embodied AI. In this paper, we present SG-Bot, a novel rearrangement framework that utilizes a coarse-to-fine scheme with a scene graph as the scene representation. Unlike previous methods that rely on either known goal priors or zero-shot large models, SG-Bot exemplifies lightweight, real-time, and user-controllable characteristics, seamlessly blending the consideration of commonsense knowledge with automatic generation capabilities. SG-Bot employs a three-fold procedure--observation, imagination, and execution--to adeptly address the task. Initially, objects are discerned and extracted from a cluttered scene during the observation. These objects are first coarsely organized and depicted within a scene graph, guided by either commonsense or user-defined criteria. Then, this scene graph subsequently informs a generative model, which forms a fine-grained goal scene considering the shape information from the initial scene and object semantics. Finally, for execution, the initial and envisioned goal scenes are matched to formulate robotic action policies. Experimental results demonstrate that SG-Bot outperforms competitors by a large margin.
Authors: Razia Sultana Misu
Brain tumors are collections of abnormal cells that can develop into masses or clusters. Because they have the potential to infiltrate other tissues, they pose a risk to the patient. The main imaging technique used, MRI, may be able to identify a brain tumor with accuracy. The fast development of Deep Learning methods for use in computer vision applications has been facilitated by a vast amount of training data and improvements in model construction that offer better approximations in a supervised setting. The need for these approaches has been the main driver of this expansion. Deep learning methods have shown promise in improving the precision of brain tumor detection and classification using magnetic resonance imaging (MRI). The study on the use of deep learning techniques, especially ResNet50, for brain tumor identification is presented in this abstract. As a result, this study investigates the possibility of automating the detection procedure using deep learning techniques. In this study, I utilized five transfer learning models which are VGG16, VGG19, DenseNet121, ResNet50 and YOLO V4 where ResNet50 provide the best or highest accuracy 99.54%. The goal of the study is to guide researchers and medical professionals toward powerful brain tumor detecting systems by employing deep learning approaches by way of this evaluation and analysis.
Authors: Gabriel Kasmi, Laurent Dubus, Yves-Marie Saint-Drenan, Philippe Blanc
Photovoltaic (PV) energy is crucial for the decarbonization of energy systems. Due to the lack of centralized data, remote sensing of rooftop PV installations is the best option to monitor the evolution of the rooftop PV installed fleet at a regional scale. However, current techniques lack reliability and are notably sensitive to shifts in the acquisition conditions. To overcome this, we leverage the wavelet scale attribution method (WCAM), which decomposes a model's prediction in the space-scale domain. The WCAM enables us to assess on which scales the representation of a PV model rests and provides insights to derive methods that improve the robustness to acquisition conditions, thus increasing trust in deep learning systems to encourage their use for the safe integration of clean energy in electric systems.
Authors: Muhammad Muneeb Saad, Mubashir Husain Rehmani, Ruairi O'Reilly
Biomedical image datasets can be imbalanced due to the rarity of targeted diseases. Generative Adversarial Networks play a key role in addressing this imbalance by enabling the generation of synthetic images to augment datasets. It is important to generate synthetic images that incorporate a diverse range of features to accurately represent the distribution of features present in the training imagery. Furthermore, the absence of diverse features in synthetic images can degrade the performance of machine learning classifiers. The mode collapse problem impacts Generative Adversarial Networks' capacity to generate diversified images. Mode collapse comes in two varieties: intra-class and inter-class. In this paper, both varieties of the mode collapse problem are investigated, and their subsequent impact on the diversity of synthetic X-ray images is evaluated. This work contributes an empirical demonstration of the benefits of integrating the adaptive input-image normalization with the Deep Convolutional GAN and Auxiliary Classifier GAN to alleviate the mode collapse problems. Synthetically generated images are utilized for data augmentation and training a Vision Transformer model. The classification performance of the model is evaluated using accuracy, recall, and precision scores. Results demonstrate that the DCGAN and the ACGAN with adaptive input-image normalization outperform the DCGAN and ACGAN with un-normalized X-ray images as evidenced by the superior diversity scores and classification scores.
Authors: Ruizhao Zhu, Peng Huang, Eshed Ohn-Bar, Venkatesh Saligrama
Human drivers can seamlessly adapt their driving decisions across geographical locations with diverse conditions and rules of the road, e.g., left vs. right-hand traffic. In contrast, existing models for autonomous driving have been thus far only deployed within restricted operational domains, i.e., without accounting for varying driving behaviors across locations or model scalability. In this work, we propose AnyD, a single geographically-aware conditional imitation learning (CIL) model that can efficiently learn from heterogeneous and globally distributed data with dynamic environmental, traffic, and social characteristics. Our key insight is to introduce a high-capacity geo-location-based channel attention mechanism that effectively adapts to local nuances while also flexibly modeling similarities among regions in a data-driven manner. By optimizing a contrastive imitation objective, our proposed approach can efficiently scale across inherently imbalanced data distributions and location-dependent events. We demonstrate the benefits of our AnyD agent across multiple datasets, cities, and scalable deployment paradigms, i.e., centralized, semi-supervised, and distributed agent training. Specifically, AnyD outperforms CIL baselines by over 14% in open-loop evaluation and 30% in closed-loop testing on CARLA.
Authors: Irmak Guzey, Yinlong Dai, Ben Evans, Soumith Chintala, Lerrel Pinto
Equipping multi-fingered robots with tactile sensing is crucial for achieving the precise, contact-rich, and dexterous manipulation that humans excel at. However, relying solely on tactile sensing fails to provide adequate cues for reasoning about objects' spatial configurations, limiting the ability to correct errors and adapt to changing situations. In this paper, we present Tactile Adaptation from Visual Incentives (TAVI), a new framework that enhances tactile-based dexterity by optimizing dexterous policies using vision-based rewards. First, we use a contrastive-based objective to learn visual representations. Next, we construct a reward function using these visual representations through optimal-transport based matching on one human demonstration. Finally, we use online reinforcement learning on our robot to optimize tactile-based policies that maximize the visual reward. On six challenging tasks, such as peg pick-and-place, unstacking bowls, and flipping slender objects, TAVI achieves a success rate of 73% using our four-fingered Allegro robot hand. The increase in performance is 108% higher than policies using tactile and vision-based rewards and 135% higher than policies without tactile observational input. Robot videos are best viewed on our project website: https://see-to-touch.github.io/.
Authors: Stefan Smeu, Elena Burceanu, Emanuela Haller, Andrei Liviu Nicolicioiu
We tackle the problem of robust novelty detection, where we aim to detect novelties in terms of semantic content while being invariant to changes in other, irrelevant factors. Specifically, we operate in a setup with multiple environments, where we determine the set of features that are associated more with the environments, rather than to the content relevant for the task. Thus, we propose a method that starts with a pretrained embedding and a multi-env setup and manages to rank the features based on their environment-focus. First, we compute a per-feature score based on the feature distribution variance between envs. Next, we show that by dropping the highly scored ones, we manage to remove spurious correlations and improve the overall performance by up to 6%, both in covariance and sub-population shift cases, both for a real and a synthetic benchmark, that we introduce for this task.
Authors: Peiying Zhang, Nanxuan Zhao, Jing Liao
Vector graphics are widely used in digital art and valued by designers for their scalability and layer-wise topological properties. However, the creation and editing of vector graphics necessitate creativity and design expertise, leading to a time-consuming process. In this paper, we propose a novel pipeline that generates high-quality customized vector graphics based on textual prompts while preserving the properties and layer-wise information of a given exemplar SVG. Our method harnesses the capabilities of large pre-trained text-to-image models. By fine-tuning the cross-attention layers of the model, we generate customized raster images guided by textual prompts. To initialize the SVG, we introduce a semantic-based path alignment method that preserves and transforms crucial paths from the exemplar SVG. Additionally, we optimize path parameters using both image-level and vector-level losses, ensuring smooth shape deformation while aligning with the customized raster image. We extensively evaluate our method using multiple metrics from vector-level, image-level, and text-level perspectives. The evaluation results demonstrate the effectiveness of our pipeline in generating diverse customizations of vector graphics with exceptional quality. The project page is https://intchous.github.io/SVGCustomization.
Authors: Shilin Yan, Xiaohao Xu, Lingyi Hong, Wenchao Chen, Wenqiang Zhang, Wei Zhang
Panoramic videos contain richer spatial information and have attracted tremendous amounts of attention due to their exceptional experience in some fields such as autonomous driving and virtual reality. However, existing datasets for video segmentation only focus on conventional planar images. To address the challenge, in this paper, we present a panoramic video dataset, PanoVOS. The dataset provides 150 videos with high video resolutions and diverse motions. To quantify the domain gap between 2D planar videos and panoramic videos, we evaluate 15 off-the-shelf video object segmentation (VOS) models on PanoVOS. Through error analysis, we found that all of them fail to tackle pixel-level content discontinues of panoramic videos. Thus, we present a Panoramic Space Consistency Transformer (PSCFormer), which can effectively utilize the semantic boundary information of the previous frame for pixel-level matching with the current frame. Extensive experiments demonstrate that compared with the previous SOTA models, our PSCFormer network exhibits a great advantage in terms of segmentation results under the panoramic setting. Our dataset poses new challenges in panoramic VOS and we hope that our PanoVOS can advance the development of panoramic segmentation/tracking.
Authors: Junseok Ahn, Youngjoon Jang, Joon Son Chung
The objective of this work is the effective extraction of spatial and dynamic features for Continuous Sign Language Recognition (CSLR). To accomplish this, we utilise a two-pathway SlowFast network, where each pathway operates at distinct temporal resolutions to separately capture spatial (hand shapes, facial expressions) and dynamic (movements) information. In addition, we introduce two distinct feature fusion methods, carefully designed for the characteristics of CSLR: (1) Bi-directional Feature Fusion (BFF), which facilitates the transfer of dynamic semantics into spatial semantics and vice versa; and (2) Pathway Feature Enhancement (PFE), which enriches dynamic and spatial representations through auxiliary subnetworks, while avoiding the need for extra inference time. As a result, our model further strengthens spatial and dynamic representations in parallel. We demonstrate that the proposed framework outperforms the current state-of-the-art performance on popular CSLR datasets, including PHOENIX14, PHOENIX14-T, and CSL-Daily.
Authors: Chaeyoung Jung, Suyeon Lee, Kihyun Nam, Kyeongha Rho, You Jin Kim, Youngjoon Jang, Joon Son Chung
The goal of this work is Active Speaker Detection (ASD), a task to determine whether a person is speaking or not in a series of video frames. Previous works have dealt with the task by exploring network architectures while learning effective representations has been less explored. In this work, we propose TalkNCE, a novel talk-aware contrastive loss. The loss is only applied to part of the full segments where a person on the screen is actually speaking. This encourages the model to learn effective representations through the natural correspondence of speech and facial movements. Our loss can be jointly optimized with the existing objectives for training ASD models without the need for additional supervision or training data. The experiments demonstrate that our loss can be easily integrated into the existing ASD frameworks, improving their performance. Our method achieves state-of-the-art performances on AVA-ActiveSpeaker and ASW datasets.
Authors: Jianing Yang, Xuweiyi Chen, Shengyi Qian, Nikhil Madaan, Madhavan Iyengar, David F. Fouhey, Joyce Chai
3D visual grounding is a critical skill for household robots, enabling them to navigate, manipulate objects, and answer questions based on their environment. While existing approaches often rely on extensive labeled data or exhibit limitations in handling complex language queries, we propose LLM-Grounder, a novel zero-shot, open-vocabulary, Large Language Model (LLM)-based 3D visual grounding pipeline. LLM-Grounder utilizes an LLM to decompose complex natural language queries into semantic constituents and employs a visual grounding tool, such as OpenScene or LERF, to identify objects in a 3D scene. The LLM then evaluates the spatial and commonsense relations among the proposed objects to make a final grounding decision. Our method does not require any labeled training data and can generalize to novel 3D scenes and arbitrary text queries. We evaluate LLM-Grounder on the ScanRefer benchmark and demonstrate state-of-the-art zero-shot grounding accuracy. Our findings indicate that LLMs significantly improve the grounding capability, especially for complex language queries, making LLM-Grounder an effective approach for 3D vision-language tasks in robotics. Videos and interactive demos can be found on the project website https://chat-with-nerf.github.io/ .
Authors: Jeremy A. Collins, Cody Houff, You Liang Tan, Charles C. Kemp
We present ForceSight, a system for text-guided mobile manipulation that predicts visual-force goals using a deep neural network. Given a single RGBD image combined with a text prompt, ForceSight determines a target end-effector pose in the camera frame (kinematic goal) and the associated forces (force goal). Together, these two components form a visual-force goal. Prior work has demonstrated that deep models outputting human-interpretable kinematic goals can enable dexterous manipulation by real robots. Forces are critical to manipulation, yet have typically been relegated to lower-level execution in these systems. When deployed on a mobile manipulator equipped with an eye-in-hand RGBD camera, ForceSight performed tasks such as precision grasps, drawer opening, and object handovers with an 81% success rate in unseen environments with object instances that differed significantly from the training data. In a separate experiment, relying exclusively on visual servoing and ignoring force goals dropped the success rate from 90% to 45%, demonstrating that force goals can significantly enhance performance. The appendix, videos, code, and trained models are available at https://force-sight.github.io/.
Authors: Kan Wu, Houwen Peng, Zhenghong Zhou, Bin Xiao, Mengchen Liu, Lu Yuan, Hong Xuan, Michael Valenzuela, Xi (Stephen) Chen, Xinggang Wang, Hongyang Chao, Han Hu
In this paper, we propose a novel cross-modal distillation method, called TinyCLIP, for large-scale language-image pre-trained models. The method introduces two core techniques: affinity mimicking and weight inheritance. Affinity mimicking explores the interaction between modalities during distillation, enabling student models to mimic teachers' behavior of learning cross-modal feature alignment in a visual-linguistic affinity space. Weight inheritance transmits the pre-trained weights from the teacher models to their student counterparts to improve distillation efficiency. Moreover, we extend the method into a multi-stage progressive distillation to mitigate the loss of informative weights during extreme compression. Comprehensive experiments demonstrate the efficacy of TinyCLIP, showing that it can reduce the size of the pre-trained CLIP ViT-B/32 by 50%, while maintaining comparable zero-shot performance. While aiming for comparable performance, distillation with weight inheritance can speed up the training by 1.4 - 7.8 $\times$ compared to training from scratch. Moreover, our TinyCLIP ViT-8M/16, trained on YFCC-15M, achieves an impressive zero-shot top-1 accuracy of 41.1% on ImageNet, surpassing the original CLIP ViT-B/16 by 3.5% while utilizing only 8.9% parameters. Finally, we demonstrate the good transferability of TinyCLIP in various downstream tasks. Code and models will be open-sourced at https://aka.ms/tinyclip.
Authors: Luca Bartolomei, Matteo Poggi, Fabio Tosi, Andrea Conti, Stefano Mattoccia
This paper proposes a novel framework integrating the principles of active stereo in standard passive camera systems without a physical pattern projector. We virtually project a pattern over the left and right images according to the sparse measurements obtained from a depth sensor. Any such devices can be seamlessly plugged into our framework, allowing for the deployment of a virtual active stereo setup in any possible environment, overcoming the limitation of pattern projectors, such as limited working range or environmental conditions. Experiments on indoor/outdoor datasets, featuring both long and close-range, support the seamless effectiveness of our approach, boosting the accuracy of both stereo algorithms and deep networks.
Authors: Xiao Wang, Jianing Li, Lin Zhu, Zhipeng Zhang, Zhe Chen, Xin Li, Yaowei Wang, Yonghong Tian, Feng Wu
Different from visible cameras which record intensity images frame by frame, the biologically inspired event camera produces a stream of asynchronous and sparse events with much lower latency. In practice, visible cameras can better perceive texture details and slow motion, while event cameras can be free from motion blurs and have a larger dynamic range which enables them to work well under fast motion and low illumination. Therefore, the two sensors can cooperate with each other to achieve more reliable object tracking. In this work, we propose a large-scale Visible-Event benchmark (termed VisEvent) due to the lack of a realistic and scaled dataset for this task. Our dataset consists of 820 video pairs captured under low illumination, high speed, and background clutter scenarios, and it is divided into a training and a testing subset, each of which contains 500 and 320 videos, respectively. Based on VisEvent, we transform the event flows into event images and construct more than 30 baseline methods by extending current single-modality trackers into dual-modality versions. More importantly, we further build a simple but effective tracking algorithm by proposing a cross-modality transformer, to achieve more effective feature fusion between visible and event data. Extensive experiments on the proposed VisEvent dataset, FE108, COESOT, and two simulated datasets (i.e., OTB-DVS and VOT-DVS), validated the effectiveness of our model. The dataset and source code have been released on: \url{https://github.com/wangxiao5791509/VisEvent_SOT_Benchmark}.
Authors: Zhitong Xiong, Wei Huang, Jingtao Hu, Xiao Xiang Zhu
Generating 3D city models rapidly is crucial for many applications. Monocular height estimation is one of the most efficient and timely ways to obtain large-scale geometric information. However, existing works focus primarily on training and testing models using unbiased datasets, which does not align well with real-world applications. Therefore, we propose a new benchmark dataset to study the transferability of height estimation models in a cross-dataset setting. To this end, we first design and construct a large-scale benchmark dataset for cross-dataset transfer learning on the height estimation task. This benchmark dataset includes a newly proposed large-scale synthetic dataset, a newly collected real-world dataset, and four existing datasets from different cities. Next, a new experimental protocol, few-shot cross-dataset transfer, is designed. Furthermore, in this paper, we propose a scale-deformable convolution module to enhance the window-based Transformer for handling the scale-variation problem in the height estimation task. Experimental results have demonstrated the effectiveness of the proposed methods in the traditional and cross-dataset transfer settings. The datasets and codes are publicly available at https://mediatum.ub.tum.de/1662763 and https://thebenchmarkh.github.io/.
Authors: Jinglei Shi, Christine Guillemot
We propose in this paper a Quantized Distilled Low-Rank Neural Radiance Field (QDLR-NeRF) representation for the task of light field compression. While existing compression methods encode the set of light field sub-aperture images, our proposed method learns an implicit scene representation in the form of a Neural Radiance Field (NeRF), which also enables view synthesis. To reduce its size, the model is first learned under a Low-Rank (LR) constraint using a Tensor Train (TT) decomposition within an Alternating Direction Method of Multipliers (ADMM) optimization framework. To further reduce the model's size, the components of the tensor train decomposition need to be quantized. However, simultaneously considering the optimization of the NeRF model with both the low-rank constraint and rate-constrained weight quantization is challenging. To address this difficulty, we introduce a network distillation operation that separates the low-rank approximation and the weight quantization during network training. The information from the initial LR-constrained NeRF (LR-NeRF) is distilled into a model of much smaller dimension (DLR-NeRF) based on the TT decomposition of the LR-NeRF. We then learn an optimized global codebook to quantize all TT components, producing the final QDLR-NeRF. Experimental results show that our proposed method yields better compression efficiency compared to state-of-the-art methods, and it additionally has the advantage of allowing the synthesis of any light field view with high quality.
Authors: Ashish Rauniyar, Desta Haileselassie Hagos, Debesh Jha, Jan Erik Håkegård, Ulas Bagci, Danda B. Rawat, Vladimir Vlassov
With the advent of the IoT, AI and ML/DL algorithms, the landscape of data-driven medical applications has emerged as a promising avenue for designing robust and scalable diagnostic and prognostic models from medical data. Consequently, the realm of data-driven medical applications has garnered significant attention spanning academia and industry, ushering in marked enhancements in healthcare delivery quality. Despite these strides, the adoption of AI-driven medical applications remains hindered by formidable challenges, including the arduous task of meeting security, privacy, and quality of service (QoS) standards. Recent developments in federated learning have made it possible to train complex machine-learned models in a distributed manner and has become an active research domain, particularly processing the medical data at the edge of the network in a decentralized way to preserve privacy and address security concerns. To this end, this survey paper highlights the current and future of FL technology in medical applications where data sharing is a significant burden. We delve into the contemporary research trends and their outcomes, unravelling the intricacies of designing reliable and scalable FL models. Our survey outlines the foundational statistical predicaments of FL, confronts device-related obstacles, delves into security challenges, and navigates the intricate terrain of privacy concerns, all while spotlighting its transformative potential within the medical domain. A primary focus of our study rests on medical applications, where we underscore the weighty burden of global cancer and illuminate the potency of FL in engendering computer-aided diagnosis tools that address this challenge with heightened efficacy.
Authors: Anjun Chen, Xiangyu Wang, Shaohao Zhu, Yanxu Li, Jiming Chen, Qi Ye
Millimeter Wave (mmWave) Radar is gaining popularity as it can work in adverse environments like smoke, rain, snow, poor lighting, etc. Prior work has explored the possibility of reconstructing 3D skeletons or meshes from the noisy and sparse mmWave Radar signals. However, it is unclear how accurately we can reconstruct the 3D body from the mmWave signals across scenes and how it performs compared with cameras, which are important aspects needed to be considered when either using mmWave radars alone or combining them with cameras. To answer these questions, an automatic 3D body annotation system is first designed and built up with multiple sensors to collect a large-scale dataset. The dataset consists of synchronized and calibrated mmWave radar point clouds and RGB(D) images in different scenes and skeleton/mesh annotations for humans in the scenes. With this dataset, we train state-of-the-art methods with inputs from different sensors and test them in various scenarios. The results demonstrate that 1) despite the noise and sparsity of the generated point clouds, the mmWave radar can achieve better reconstruction accuracy than the RGB camera but worse than the depth camera; 2) the reconstruction from the mmWave radar is affected by adverse weather conditions moderately while the RGB(D) camera is severely affected. Further, analysis of the dataset and the results shadow insights on improving the reconstruction from the mmWave radar and the combination of signals from different sensors.
Authors: Alireza Aghelan, Modjtaba Rouhani
In the field of medical image analysis, there is a substantial need for high-resolution (HR) images to improve diagnostic accuracy. However, It is a challenging task to obtain HR medical images, as it requires advanced instruments and significant time. Deep learning-based super-resolution methods can help to improve the resolution and perceptual quality of low-resolution (LR) medical images. Recently, Generative Adversarial Network (GAN) based methods have shown remarkable performance among deep learning-based super-resolution methods. Real-Enhanced Super-Resolution Generative Adversarial Network (Real-ESRGAN) is a practical model for recovering HR images from real-world LR images. In our proposed approach, we use transfer learning technique and fine-tune the pre-trained Real-ESRGAN model using medical image datasets. This technique helps in improving the performance of the model. The focus of this paper is on enhancing the resolution and perceptual quality of chest X-ray and retinal images. We use the Tuberculosis chest X-ray (Shenzhen) dataset and the STARE dataset of retinal images for fine-tuning the model. The proposed model achieves superior perceptual quality compared to the Real-ESRGAN model, effectively preserving fine details and generating images with more realistic textures.
Authors: Hengrong Lan, Lijie Huang, Zhiqiang Li, Jing Lv, Jianwen Luo
Accurate image reconstruction is crucial for photoacoustic (PA) computed tomography (PACT). Recently, deep learning has been used to reconstruct the PA image with a supervised scheme, which requires high-quality images as ground truth labels. In practice, there are inevitable trade-offs between cost and performance since the use of more channels is an expensive strategy to access more measurements. Here, we propose a cross-domain unsupervised reconstruction (CDUR) strategy with a pure transformer model, which overcomes the lack of ground truth labels from limited PA measurements. The proposed approach exploits the equivariance of PACT to achieve high performance with a smaller number of channels. We implement a self-supervised reconstruction in a model-based form. Meanwhile, we also leverage the self-supervision to enforce the measurement and image consistency on three partitions of measured PA data, by randomly masking different channels. We find that dynamically masking a high proportion of the channels, e.g., 80%, yields nontrivial self-supervisors in both image and signal domains, which decrease the multiplicity of the pseudo solution to efficiently reconstruct the image from fewer PA measurements with minimum error of the image. Experimental results on in-vivo PACT dataset of mice demonstrate the potential of our unsupervised framework. In addition, our method shows a high performance (0.83 structural similarity index (SSIM) in the extreme sparse case with 13 channels), which is close to that of supervised scheme (0.77 SSIM with 16 channels). On top of all the advantages, our method may be deployed on different trainable models in an end-to-end manner.
Authors: Yunpeng Bai, Zihan Zhong, Chao Dong, Weichen Zhang, Guowei Xu, Chun Yuan
The recent GAN inversion methods have been able to successfully invert the real image input to the corresponding editable latent code in StyleGAN. By combining with the language-vision model (CLIP), some text-driven image manipulation methods are proposed. However, these methods require extra costs to perform optimization for a certain image or a new attribute editing mode. To achieve a more efficient editing method, we propose a new Text-driven image Manipulation framework via Space Alignment (TMSA). The Space Alignment module aims to align the same semantic regions in CLIP and StyleGAN spaces. Then, the text input can be directly accessed into the StyleGAN space and be used to find the semantic shift according to the text description. The framework can support arbitrary image editing mode without additional cost. Our work provides the user with an interface to control the attributes of a given image according to text input and get the result in real time. Ex tensive experiments demonstrate our superior performance over prior works.
Authors: Daoan Zhang, Mingkai Chen, Chenming Li, Lingyun Huang, Jianguo Zhang
Domain Generalization (DG) is a fundamental challenge for machine learning models, which aims to improve model generalization on various domains. Previous methods focus on generating domain invariant features from various source domains. However, we argue that the domain variantions also contain useful information, ie, classification-aware information, for downstream tasks, which has been largely ignored. Different from learning domain invariant features from source domains, we decouple the input images into Domain Expert Features and noise. The proposed domain expert features lie in a learned latent space where the images in each domain can be classified independently, enabling the implicit use of classification-aware domain variations. Based on the analysis, we proposed a novel paradigm called Domain Disentanglement Network (DDN) to disentangle the domain expert features from the source domain images and aggregate the source domain expert features for representing the target test domain. We also propound a new contrastive learning method to guide the domain expert features to form a more balanced and separable feature space. Experiments on the widely-used benchmarks of PACS, VLCS, OfficeHome, DomainNet, and TerraIncognita demonstrate the competitive performance of our method compared to the recently proposed alternatives.
Authors: Nguyen Huu Phong, Bernardete Ribeiro
Recognizing human actions in video sequences, known as Human Action Recognition (HAR), is a challenging task in pattern recognition. While Convolutional Neural Networks (ConvNets) have shown remarkable success in image recognition, they are not always directly applicable to HAR, as temporal features are critical for accurate classification. In this paper, we propose a novel dynamic PSO-ConvNet model for learning actions in videos, building on our recent work in image recognition. Our approach leverages a framework where the weight vector of each neural network represents the position of a particle in phase space, and particles share their current weight vectors and gradient estimates of the Loss function. To extend our approach to video, we integrate ConvNets with state-of-the-art temporal methods such as Transformer and Recurrent Neural Networks. Our experimental results on the UCF-101 dataset demonstrate substantial improvements of up to 9% in accuracy, which confirms the effectiveness of our proposed method. In addition, we conducted experiments on larger and more variety of datasets including Kinetics-400 and HMDB-51 and obtained preference for Collaborative Learning in comparison with Non-Collaborative Learning (Individual Learning). Overall, our dynamic PSO-ConvNet model provides a promising direction for improving HAR by better capturing the spatio-temporal dynamics of human actions in videos. The code is available at https://github.com/leonlha/Video-Action-Recognition-Collaborative-Learning-with-Dynamics-via-PSO-ConvNet-Transformer.
Authors: Junwei Zheng, Jiaming Zhang, Kailun Yang, Kunyu Peng, Rainer Stiefelhagen
People with Visual Impairments (PVI) typically recognize objects through haptic perception. Knowing objects and materials before touching is desired by the target users but under-explored in the field of human-centered robotics. To fill this gap, in this work, a wearable vision-based robotic system, MateRobot, is established for PVI to recognize materials and object categories beforehand. To address the computational constraints of mobile platforms, we propose a lightweight yet accurate model MateViT to perform pixel-wise semantic segmentation, simultaneously recognizing both objects and materials. Our methods achieve respective 40.2% and 51.1% of mIoU on COCOStuff-10K and DMS datasets, surpassing the previous method with +5.7% and +7.0% gains. Moreover, on the field test with participants, our wearable system reaches a score of 28 in the NASA-Task Load Index, indicating low cognitive demands and ease of use. Our MateRobot demonstrates the feasibility of recognizing material property through visual cues and offers a promising step towards improving the functionality of wearable robots for PVI. The source code has been made publicly available at https://junweizheng93.github.io/publications/MATERobot/MATERobot.html.
Authors: Yifan Pu, Yiru Wang, Zhuofan Xia, Yizeng Han, Yulin Wang, Weihao Gan, Zidong Wang, Shiji Song, Gao Huang
Rotated object detection aims to identify and locate objects in images with arbitrary orientation. In this scenario, the oriented directions of objects vary considerably across different images, while multiple orientations of objects exist within an image. This intrinsic characteristic makes it challenging for standard backbone networks to extract high-quality features of these arbitrarily orientated objects. In this paper, we present Adaptive Rotated Convolution (ARC) module to handle the aforementioned challenges. In our ARC module, the convolution kernels rotate adaptively to extract object features with varying orientations in different images, and an efficient conditional computation mechanism is introduced to accommodate the large orientation variations of objects within an image. The two designs work seamlessly in rotated object detection problem. Moreover, ARC can conveniently serve as a plug-and-play module in various vision backbones to boost their representation ability to detect oriented objects accurately. Experiments on commonly used benchmarks (DOTA and HRSC2016) demonstrate that equipped with our proposed ARC module in the backbone network, the performance of multiple popular oriented object detectors is significantly improved (\eg +3.03\% mAP on Rotated RetinaNet and +4.16\% on CFA). Combined with the highly competitive method Oriented R-CNN, the proposed approach achieves state-of-the-art performance on the DOTA dataset with 81.77\% mAP. Code is available at \url{https://github.com/LeapLabTHU/ARC}.
Authors: Firas Al-Hindawi, Md Mahfuzur Rahman Siddiquee, Teresa Wu, Han Hu, Ying Sun
The ability to classify images is dependent on having access to large labeled datasets and testing on data from the same domain that the model can train on. Classification becomes more challenging when dealing with new data from a different domain, where gathering and especially labeling a larger image dataset for retraining a classification model requires a labor-intensive human effort. Cross-domain classification frameworks were developed to handle this data domain shift problem by utilizing unsupervised image-to-image translation models to translate an input image from the unlabeled domain to the labeled domain. The problem with these unsupervised models lies in their unsupervised nature. For lack of annotations, it is not possible to use the traditional supervised metrics to evaluate these translation models to pick the best-saved checkpoint model. This paper introduces a new method called Domain-knowledge Inspired Pseudo Supervision (DIPS) which utilizes domain-informed Gaussian Mixture Models to generate pseudo annotations to enable the use of traditional supervised metrics. This method was designed specifically to support cross-domain classification applications contrary to other typically used metrics such as the FID which were designed to evaluate the model in terms of the quality of the generated image from a human-eye perspective. DIPS proves its effectiveness by outperforming various GAN evaluation metrics, including FID, when selecting the optimal saved checkpoint model. It is also evaluated against truly supervised metrics. Furthermore, DIPS showcases its robustness and interpretability by demonstrating a strong correlation with truly supervised metrics, highlighting its superiority over existing state-of-the-art alternatives. The code and data to replicate the results can be found on the official Github repository: https://github.com/Hindawi91/DIPS
Authors: Aristeidis Panos, Yuriko Kobe, Daniel Olmeda Reino, Rahaf Aljundi, Richard E. Turner
In Class-Incremental Learning (CIL) an image classification system is exposed to new classes in each learning session and must be updated incrementally. Methods approaching this problem have updated both the classification head and the feature extractor body at each session of CIL. In this work, we develop a baseline method, First Session Adaptation (FSA), that sheds light on the efficacy of existing CIL approaches and allows us to assess the relative performance contributions from head and body adaption. FSA adapts a pre-trained neural network body only on the first learning session and fixes it thereafter; a head based on linear discriminant analysis (LDA), is then placed on top of the adapted body, allowing exact updates through CIL. FSA is replay-free i.e.~it does not memorize examples from previous sessions of continual learning. To empirically motivate FSA, we first consider a diverse selection of 22 image-classification datasets, evaluating different heads and body adaptation techniques in high/low-shot offline settings. We find that the LDA head performs well and supports CIL out-of-the-box. We also find that Featurewise Layer Modulation (FiLM) adapters are highly effective in the few-shot setting, and full-body adaption in the high-shot setting. Second, we empirically investigate various CIL settings including high-shot CIL and few-shot CIL, including settings that have previously been used in the literature. We show that FSA significantly improves over the state-of-the-art in 15 of the 16 settings considered. FSA with FiLM adapters is especially performant in the few-shot setting. These results indicate that current approaches to continuous body adaptation are not working as expected. Finally, we propose a measure that can be applied to a set of unlabelled inputs which is predictive of the benefits of body adaptation.
Authors: Yi Lin, Yufan Chen, Kwang-Ting Cheng, Hao Chen
Medical image segmentation has made significant progress in recent years. Deep learning-based methods are recognized as data-hungry techniques, requiring large amounts of data with manual annotations. However, manual annotation is expensive in the field of medical image analysis, which requires domain-specific expertise. To address this challenge, few-shot learning has the potential to learn new classes from only a few examples. In this work, we propose a novel framework for few-shot medical image segmentation, termed CAT-Net, based on cross masked attention Transformer. Our proposed network mines the correlations between the support image and query image, limiting them to focus only on useful foreground information and boosting the representation capacity of both the support prototype and query features. We further design an iterative refinement framework that refines the query image segmentation iteratively and promotes the support feature in turn. We validated the proposed method on three public datasets: Abd-CT, Abd-MRI, and Card-MRI. Experimental results demonstrate the superior performance of our method compared to state-of-the-art methods and the effectiveness of each component. Code: https://github.com/hust-linyi/CAT-Net.
Authors: Chaitanya Devaguptapu, Samarth Sinha, K J Joseph, Vineeth N Balasubramanian, Animesh Garg
Models pre-trained on large-scale datasets are often fine-tuned to support newer tasks and datasets that arrive over time. This process necessitates storing copies of the model over time for each task that the pre-trained model is fine-tuned to. Building on top of recent model patching work, we propose $\Delta$-Patching for fine-tuning neural network models in an efficient manner, without the need to store model copies. We propose a simple and lightweight method called $\Delta$-Networks to achieve this objective. Our comprehensive experiments across setting and architecture variants show that $\Delta$-Networks outperform earlier model patching work while only requiring a fraction of parameters to be trained. We also show that this approach can be used for other problem settings such as transfer learning and zero-shot domain adaptation, as well as other tasks such as detection and segmentation.
Authors: Kai Yang, Hong Shang, Tianyang Shi, Xinghan Chen, Jingkai Zhou, Zhongqian Sun, Wei Yang
The research fields of parametric face model and 3D face reconstruction have been extensively studied. However, a critical question remains unanswered: how to tailor the face model for specific reconstruction settings. We argue that reconstruction with multi-view uncalibrated images demands a new model with stronger capacity. Our study shifts attention from data-dependent 3D Morphable Models (3DMM) to an understudied human-designed skinning model. We propose Adaptive Skinning Model (ASM), which redefines the skinning model with more compact and fully tunable parameters. With extensive experiments, we demonstrate that ASM achieves significantly improved capacity than 3DMM, with the additional advantage of model size and easy implementation for new topology. We achieve state-of-the-art performance with ASM for multi-view reconstruction on the Florence MICC Coop benchmark. Our quantitative analysis demonstrates the importance of a high-capacity model for fully exploiting abundant information from multi-view input in reconstruction. Furthermore, our model with physical-semantic parameters can be directly utilized for real-world applications, such as in-game avatar creation. As a result, our work opens up new research direction for parametric face model and facilitates future research on multi-view reconstruction.
Authors: Christopher Lang, Alexander Braun, Lars Schillingmann, Abhinav Valada
Self-supervised multi-object trackers have tremendous potential as they enable learning from raw domain-specific data. However, their re-identification accuracy still falls short compared to their supervised counterparts. We hypothesize that this drawback results from formulating self-supervised objectives that are limited to single frames or frame pairs. Such formulations do not capture sufficient visual appearance variations to facilitate learning consistent re-identification features for autonomous driving when the frame rate is low or object dynamics are high. In this work, we propose a training objective that enables self-supervised learning of re-identification features from multiple sequential frames by enforcing consistent association scores across short and long timescales. We perform extensive evaluations demonstrating that re-identification features trained from longer sequences significantly reduce ID switches on standard autonomous driving datasets compared to existing self-supervised learning methods, which are limited to training on frame pairs. Using our proposed SubCo loss function, we set the new state-of-the-art among self-supervised methods and even perform on par with fully supervised learning methods.
Authors: Kyle Vedder, Neehar Peri, Nathaniel Chodosh, Ishan Khatri, Eric Eaton, Dinesh Jayaraman, Yang Liu, Deva Ramanan, James Hays
Scene flow estimation is the task of describing the 3D motion field between temporally successive point clouds. State-of-the-art methods use strong priors and test-time optimization techniques, but require on the order of tens of seconds to process large-scale point clouds, making them unusable as computer vision primitives for real-time applications such as open world object detection. Feed forward methods are considerably faster, running on the order of tens to hundreds of milliseconds for large-scale point clouds, but require expensive human supervision. To address both limitations, we propose Scene Flow via Distillation, a simple, scalable distillation framework that uses a label-free optimization method to produce pseudo-labels to supervise a feed forward model. Our instantiation of this framework, ZeroFlow, achieves state-of-the-art performance on the Argoverse 2 Self-Supervised Scene Flow Challenge while using zero human labels by simply training on large-scale, diverse unlabeled data. At test-time, ZeroFlow is over 1000$\times$ faster than label-free state-of-the-art optimization-based methods on large-scale point clouds and over 1000$\times$ cheaper to train on unlabeled data compared to the cost of human annotation of that data. To facilitate further research, we will release our code, trained model weights, and high quality pseudo-labels for the Argoverse 2 and Waymo Open datasets.
Authors: Quoc-Huy Tran, Ahmed Mehmood, Muhammad Ahmed, Muhammad Naufil, Anas Zafar, Andrey Konin, M. Zeeshan Zia
This paper presents an unsupervised transformer-based framework for temporal activity segmentation which leverages not only frame-level cues but also segment-level cues. This is in contrast with previous methods which often rely on frame-level information only. Our approach begins with a frame-level prediction module which estimates framewise action classes via a transformer encoder. The frame-level prediction module is trained in an unsupervised manner via temporal optimal transport. To exploit segment-level information, we utilize a segment-level prediction module and a frame-to-segment alignment module. The former includes a transformer decoder for estimating video transcripts, while the latter matches frame-level features with segment-level features, yielding permutation-aware segmentation results. Moreover, inspired by temporal optimal transport, we introduce simple-yet-effective pseudo labels for unsupervised training of the above modules. Our experiments on four public datasets, i.e., 50 Salads, YouTube Instructions, Breakfast, and Desktop Assembly show that our approach achieves comparable or better performance than previous methods in unsupervised activity segmentation.
Authors: Quoc-Huy Tran, Muhammad Ahmed, M. Hassan Ahmed, Murad Popattia, Andrey Konin, M. Zeeshan Zia
This paper presents a self-supervised temporal video alignment framework which is useful for several fine-grained human activity understanding applications. In contrast with the state-of-the-art method of CASA, where sequences of 3D skeleton coordinates are taken directly as input, our key idea is to use sequences of 2D skeleton heatmaps as input. Unlike CASA which performs self-attention in the temporal domain only, we feed 2D skeleton heatmaps to a video transformer which performs self-attention both in the spatial and temporal domains for extracting effective spatiotemporal and contextual features. In addition, we introduce simple heatmap augmentation techniques based on 2D skeletons for self-supervised learning. Despite the lack of 3D information, our approach achieves not only higher accuracy but also better robustness against missing and noisy keypoints than CASA. Furthermore, extensive evaluations on three public datasets, i.e., Penn Action, IKEA ASM, and H2O, demonstrate that our approach outperforms previous methods in different fine-grained human activity understanding tasks. Finally, fusing 2D skeleton heatmaps with RGB videos yields the state-of-the-art on all metrics and datasets. To the best of our knowledge, our work is the first to utilize 2D skeleton heatmap inputs and the first to explore multi-modality fusion for temporal video alignment.
Authors: Leilei Ma, Dengdi Sun, Lei Wang, Haifeng Zhao, Bin Luo
Extracting image semantics effectively and assigning corresponding labels to multiple objects or attributes for natural images is challenging due to the complex scene contents and confusing label dependencies. Recent works have focused on modeling label relationships with graph and understanding object regions using class activation maps (CAM). However, these methods ignore the complex intra- and inter-category relationships among specific semantic features, and CAM is prone to generate noisy information. To this end, we propose a novel semantic-aware dual contrastive learning framework that incorporates sample-to-sample contrastive learning (SSCL) as well as prototype-to-sample contrastive learning (PSCL). Specifically, we leverage semantic-aware representation learning to extract category-related local discriminative features and construct category prototypes. Then based on SSCL, label-level visual representations of the same category are aggregated together, and features belonging to distinct categories are separated. Meanwhile, we construct a novel PSCL module to narrow the distance between positive samples and category prototypes and push negative samples away from the corresponding category prototypes. Finally, the discriminative label-level features related to the image content are accurately captured by the joint training of the above three parts. Experiments on five challenging large-scale public datasets demonstrate that our proposed method is effective and outperforms the state-of-the-art methods. Code and supplementary materials are released on https://github.com/yu-gi-oh-leilei/SADCL.
Authors: Annika Ries, Tina Dorosti, Johannes Thalhammer, Daniel Sasse, Andreas Sauter, Felix Meurer, Ashley Benne, Tobias Lasser, Franz Pfeiffer, Florian Schaff, Daniela Pfeiffer
Purpose: To improve the image quality of sparse-view computed tomography (CT) images with a U-Net for lung cancer detection and to determine the best trade-off between number of views, image quality, and diagnostic confidence.
Methods: CT images from 41 subjects (34 with lung cancer, seven healthy) were retrospectively selected (01.2016-12.2018) and forward projected onto 2048-view sinograms. Six corresponding sparse-view CT data subsets at varying levels of undersampling were reconstructed from sinograms using filtered backprojection with 16, 32, 64, 128, 256, and 512 views, respectively. A dual-frame U-Net was trained and evaluated for each subsampling level on 8,658 images from 22 diseased subjects. A representative image per scan was selected from 19 subjects (12 diseased, seven healthy) for a single-blinded reader study. The selected slices, for all levels of subsampling, with and without post-processing by the U-Net model, were presented to three readers. Image quality and diagnostic confidence were ranked using pre-defined scales. Subjective nodule segmentation was evaluated utilizing sensitivity (Se) and Dice Similarity Coefficient (DSC) with 95% confidence intervals (CI).
Results: The 64-projection sparse-view images resulted in Se = 0.89 and DSC = 0.81 [0.75,0.86] while their counterparts, post-processed with the U-Net, had improved metrics (Se = 0.94, DSC = 0.85 [0.82,0.87]). Fewer views lead to insufficient quality for diagnostic purposes. For increased views, no substantial discrepancies were noted between the sparse-view and post-processed images.
Conclusion: Projection views can be reduced from 2048 to 64 while maintaining image quality and the confidence of the radiologists on a satisfactory level.
Authors: Yangyang Xu, Yibo Yang, Bernard Ghanem, Lefei Zhang, Du Bo, Dacheng Tao
CNNs and Transformers have their own advantages and both have been widely used for dense prediction in multi-task learning (MTL). Most of the current studies on MTL solely rely on CNN or Transformer. In this work, we present a novel MTL model by combining both merits of deformable CNN and query-based Transformer with shared gating for multi-task learning of dense prediction. This combination may offer a simple and efficient solution owing to its powerful and flexible task-specific learning and advantages of lower cost, less complexity and smaller parameters than the traditional MTL methods. We introduce deformable mixer Transformer with gating (DeMTG), a simple and effective encoder-decoder architecture up-to-date that incorporates the convolution and attention mechanism in a unified network for MTL. It is exquisitely designed to use advantages of each block, and provide deformable and comprehensive features for all tasks from local and global perspective. First, the deformable mixer encoder contains two types of operators: the channel-aware mixing operator leveraged to allow communication among different channels, and the spatial-aware deformable operator with deformable convolution applied to efficiently sample more informative spatial locations. Second, the task-aware gating transformer decoder is used to perform the task-specific predictions, in which task interaction block integrated with self-attention is applied to capture task interaction features, and the task query block integrated with gating attention is leveraged to select corresponding task-specific features. Further, the experiment results demonstrate that the proposed DeMTG uses fewer GFLOPs and significantly outperforms current Transformer-based and CNN-based competitive models on a variety of metrics on three dense prediction datasets. Our code and models are available at https://github.com/yangyangxu0/DeMTG.
Authors: Victor Gallego
Recently, large multimodal models, such as CLIP and Stable Diffusion have experimented tremendous successes in both foundations and applications. However, as these models increase in parameter size and computational requirements, it becomes more challenging for users to personalize them for specific tasks or preferences. In this work, we address the problem of adapting the previous models towards sets of particular human preferences, aligning the retrieved or generated images with the preferences of the user. We leverage the Bradley-Terry preference model to develop a fast adaptation method that efficiently fine-tunes the original model, with few examples and with minimal computing resources. Extensive evidence of the capabilities of this framework is provided through experiments in different domains related to multimodal text and image understanding, including preference prediction as a reward model, and generation tasks.
Authors: Feng Liu, Minchul Kim, ZiAng Gu, Anil Jain, Xiaoming Liu
Long-Term Person Re-Identification (LT-ReID) has become increasingly crucial in computer vision and biometrics. In this work, we aim to extend LT-ReID beyond pedestrian recognition to include a wider range of real-world human activities while still accounting for cloth-changing scenarios over large time gaps. This setting poses additional challenges due to the geometric misalignment and appearance ambiguity caused by the diversity of human pose and clothing. To address these challenges, we propose a new approach 3DInvarReID for (i) disentangling identity from non-identity components (pose, clothing shape, and texture) of 3D clothed humans, and (ii) reconstructing accurate 3D clothed body shapes and learning discriminative features of naked body shapes for person ReID in a joint manner. To better evaluate our study of LT-ReID, we collect a real-world dataset called CCDA, which contains a wide variety of human activities and clothing changes. Experimentally, we show the superior performance of our approach for person ReID.
Authors: Yuanyou Xu, Zongxin Yang, Yi Yang
Tracking any given object(s) spatially and temporally is a common purpose in Visual Object Tracking (VOT) and Video Object Segmentation (VOS). Joint tracking and segmentation have been attempted in some studies but they often lack full compatibility of both box and mask in initialization and prediction, and mainly focus on single-object scenarios. To address these limitations, this paper proposes a Multi-object Mask-box Integrated framework for unified Tracking and Segmentation, dubbed MITS. Firstly, the unified identification module is proposed to support both box and mask reference for initialization, where detailed object information is inferred from boxes or directly retained from masks. Additionally, a novel pinpoint box predictor is proposed for accurate multi-object box prediction, facilitating target-oriented representation learning. All target objects are processed simultaneously from encoding to propagation and decoding, as a unified pipeline for VOT and VOS. Experimental results show MITS achieves state-of-the-art performance on both VOT and VOS benchmarks. Notably, MITS surpasses the best prior VOT competitor by around 6% on the GOT-10k test set, and significantly improves the performance of box initialization on VOS benchmarks. The code is available at https://github.com/yoxu515/MITS.
Authors: Lingtong Kong, Boyuan Jiang, Donghao Luo, Wenqing Chu, Ying Tai, Chengjie Wang, Jie Yang
Video frame interpolation is an important low-level vision task, which can increase frame rate for more fluent visual experience. Existing methods have achieved great success by employing advanced motion models and synthesis networks. However, the spatial redundancy when synthesizing the target frame has not been fully explored, that can result in lots of inefficient computation. On the other hand, the computation compression degree in frame interpolation is highly dependent on both texture distribution and scene motion, which demands to understand the spatial-temporal information of each input frame pair for a better compression degree selection. In this work, we propose a novel two-stage frame interpolation framework termed WaveletVFI to address above problems. It first estimates intermediate optical flow with a lightweight motion perception network, and then a wavelet synthesis network uses flow aligned context features to predict multi-scale wavelet coefficients with sparse convolution for efficient target frame reconstruction, where the sparse valid masks that control computation in each scale are determined by a crucial threshold ratio. Instead of setting a fixed value like previous methods, we find that embedding a classifier in the motion perception network to learn a dynamic threshold for each sample can achieve more computation reduction with almost no loss of accuracy. On the common high resolution and animation frame interpolation benchmarks, proposed WaveletVFI can reduce computation up to 40% while maintaining similar accuracy, making it perform more efficiently against other state-of-the-arts. Code is available at https://github.com/ltkong218/WaveletVFI.
Authors: Dongwoo Lee, Jeongtaek Oh, Jaesung Rim, Sunghyun Cho, Kyoung Mu Lee
We present ExBluRF, a novel view synthesis method for extreme motion blurred images based on efficient radiance fields optimization. Our approach consists of two main components: 6-DOF camera trajectory-based motion blur formulation and voxel-based radiance fields. From extremely blurred images, we optimize the sharp radiance fields by jointly estimating the camera trajectories that generate the blurry images. In training, multiple rays along the camera trajectory are accumulated to reconstruct single blurry color, which is equivalent to the physical motion blur operation. We minimize the photo-consistency loss on blurred image space and obtain the sharp radiance fields with camera trajectories that explain the blur of all images. The joint optimization on the blurred image space demands painfully increasing computation and resources proportional to the blur size. Our method solves this problem by replacing the MLP-based framework to low-dimensional 6-DOF camera poses and voxel-based radiance fields. Compared with the existing works, our approach restores much sharper 3D scenes from challenging motion blurred views with the order of 10 times less training time and GPU memory consumption.
Authors: Tianyi Song, Jiuxin Cao, Kun Wang, Bo Liu, Xiaofeng Zhang
The excellent text-to-image synthesis capability of diffusion models has driven progress in synthesizing coherent visual stories. The current state-of-the-art method combines the features of historical captions, historical frames, and the current captions as conditions for generating the current frame. However, this method treats each historical frame and caption as the same contribution. It connects them in order with equal weights, ignoring that not all historical conditions are associated with the generation of the current frame. To address this issue, we propose Causal-Story. This model incorporates a local causal attention mechanism that considers the causal relationship between previous captions, frames, and current captions. By assigning weights based on this relationship, Causal-Story generates the current frame, thereby improving the global consistency of story generation. We evaluated our model on the PororoSV and FlintstonesSV datasets and obtained state-of-the-art FID scores, and the generated frames also demonstrate better storytelling in visuals.
Authors: Hongbo Zhao, Yikang Zhang, Qijun Chen, Rui Fan
Accurate estimation of stereo camera extrinsic parameters is the key to guarantee the performance of stereo matching algorithms. In prior arts, the online self-calibration of stereo cameras has commonly been formulated as a specialized visual odometry problem, without taking into account the principles of stereo rectification. In this paper, we first delve deeply into the concept of rectifying homography, which serves as the cornerstone for the development of our novel stereo camera online self-calibration algorithm, for cases where only a single pair of images is available. Furthermore, we introduce a simple yet effective solution for global optimum extrinsic parameter estimation in the presence of stereo video sequences. Additionally, we emphasize the impracticality of using three Euler angles and three components in the translation vectors for performance quantification. Instead, we introduce four new evaluation metrics to quantify the robustness and accuracy of extrinsic parameter estimation, applicable to both single-pair and multi-pair cases. Extensive experiments conducted across indoor and outdoor environments using various experimental setups validate the effectiveness of our proposed algorithm. The comprehensive evaluation results demonstrate its superior performance in comparison to the baseline algorithm. Our source code, demo video, and supplement are publicly available at mias.group/StereoCalibrator.
Authors: ShahRukh Athar, Zhixin Shu, Zexiang Xu, Fujun Luan, Sai Bi, Kalyan Sunkavalli, Dimitris Samaras
Recent advances in Neural Radiance Fields (NeRFs) have made it possible to reconstruct and reanimate dynamic portrait scenes with control over head-pose, facial expressions and viewing direction. However, training such models assumes photometric consistency over the deformed region e.g. the face must be evenly lit as it deforms with changing head-pose and facial expression. Such photometric consistency across frames of a video is hard to maintain, even in studio environments, thus making the created reanimatable neural portraits prone to artifacts during reanimation. In this work, we propose CoDyNeRF, a system that enables the creation of fully controllable 3D portraits in real-world capture conditions. CoDyNeRF learns to approximate illumination dependent effects via a dynamic appearance model in the canonical space that is conditioned on predicted surface normals and the facial expressions and head-pose deformations. The surface normals prediction is guided using 3DMM normals that act as a coarse prior for the normals of the human head, where direct prediction of normals is hard due to rigid and non-rigid deformations induced by head-pose and facial expression changes. Using only a smartphone-captured short video of a subject for training, we demonstrate the effectiveness of our method on free view synthesis of a portrait scene with explicit head pose and expression controls, and realistic lighting effects. The project page can be found here: this http URL
Authors: Minsu Kim, Giseop Kim, Kyong Hwan Jin, Sunwook Choi
A recent sensor fusion in a Bird's Eye View (BEV) space has shown its utility in various tasks such as 3D detection, map segmentation, etc. However, the approach struggles with inaccurate camera BEV estimation, and a perception of distant areas due to the sparsity of LiDAR points. In this paper, we propose a broad BEV fusion (BroadBEV) that addresses the problems with a spatial synchronization approach of cross-modality. Our strategy aims to enhance camera BEV estimation for a broad-sighted perception while simultaneously improving the completion of LiDAR's sparsity in the entire BEV space. Toward that end, we devise Point-scattering that scatters LiDAR BEV distribution to camera depth distribution. The method boosts the learning of depth estimation of the camera branch and induces accurate location of dense camera features in BEV space. For an effective BEV fusion between the spatially synchronized features, we suggest ColFusion that applies self-attention weights of LiDAR and camera BEV features to each other. Our extensive experiments demonstrate that BroadBEV provides a broad-sighted BEV perception with remarkable performance gains.
Authors: Weibin Yang, Longwei Xu, Pengwei Wang, Dehua Geng, Yusong Li, Mingyuan Xu, Zhiqi Dong
U-Net and its variants have been widely used in medical image segmentation. However, most current U-Net variants confine their improvement strategies to building more complex encoder, while leaving the decoder unchanged or adopting a simple symmetric structure. These approaches overlook the true functionality of the decoder: receiving low-resolution feature maps from the encoder and restoring feature map resolution and lost information through upsampling. As a result, the decoder, especially its upsampling component, plays a crucial role in enhancing segmentation outcomes. However, in 3D medical image segmentation, the commonly used transposed convolution can result in visual artifacts. This issue stems from the absence of direct relationship between adjacent pixels in the output feature map. Furthermore, plain encoder has already possessed sufficient feature extraction capability because downsampling operation leads to the gradual expansion of the receptive field, but the loss of information during downsampling process is unignorable. To address the gap in relevant research, we extend our focus beyond the encoder and introduce neU-Net (i.e., not complex encoder U-Net), which incorporates a novel Sub-pixel Convolution for upsampling to construct a powerful decoder. Additionally, we introduce multi-scale wavelet inputs module on the encoder side to provide additional information. Our model design achieves excellent results, surpassing other state-of-the-art methods on both the Synapse and ACDC datasets.
Authors: Haolong Li, Joerg Stueckler
Wheeled mobile robots need the ability to estimate their motion and the effect of their control actions for navigation planning. In this paper, we present ST-VIO, a novel approach which tightly fuses a single-track dynamics model for wheeled ground vehicles with visual inertial odometry. Our method calibrates and adapts the dynamics model online and facilitates accurate forward prediction conditioned on future control inputs. The single-track dynamics model approximates wheeled vehicle motion under specific control inputs on flat ground using ordinary differential equations. We use a singularity-free and differentiable variant of the single-track model to enable seamless integration as dynamics factor into VIO and to optimize the model parameters online together with the VIO state variables. We validate our method with real-world data in both indoor and outdoor environments with different terrain types and wheels. In our experiments, we demonstrate that our ST-VIO can not only adapt to the change of the environments and achieve accurate prediction under new control inputs, but even improves the tracking accuracy. Supplementary video: https://youtu.be/BuGY1L1FRa4.
Authors: Md Shazid Islam, Arindam Dutta, Calvin-Khang Ta, Kevin Rodriguez, Christian Michael, Mark Alber, G. Venugopala Reddy, Amit K. Roy-Chowdhury
Tracking of plant cells in images obtained by microscope is a challenging problem due to biological phenomena such as large number of cells, non-uniform growth of different layers of the tightly packed plant cells and cell division. Moreover, images in deeper layers of the tissue being noisy and unavoidable systemic errors inherent in the imaging process further complicates the problem. In this paper, we propose a novel learning-based method that exploits the tightly packed three-dimensional cell structure of plant cells to create a three-dimensional graph in order to perform accurate cell tracking. We further propose novel algorithms for cell division detection and effective three-dimensional registration, which improve upon the state-of-the-art algorithms. We demonstrate the efficacy of our algorithm in terms of tracking accuracy and inference-time on a benchmark dataset.
Authors: Chengcheng Wang, Wei He, Ying Nie, Jianyuan Guo, Chuanjian Liu, Kai Han, Yunhe Wang
In the past years, YOLO-series models have emerged as the leading approaches in the area of real-time object detection. Many studies pushed up the baseline to a higher level by modifying the architecture, augmenting data and designing new losses. However, we find previous models still suffer from information fusion problem, although Feature Pyramid Network (FPN) and Path Aggregation Network (PANet) have alleviated this. Therefore, this study provides an advanced Gatherand-Distribute mechanism (GD) mechanism, which is realized with convolution and self-attention operations. This new designed model named as Gold-YOLO, which boosts the multi-scale feature fusion capabilities and achieves an ideal balance between latency and accuracy across all model scales. Additionally, we implement MAE-style pretraining in the YOLO-series for the first time, allowing YOLOseries models could be to benefit from unsupervised pretraining. Gold-YOLO-N attains an outstanding 39.9% AP on the COCO val2017 datasets and 1030 FPS on a T4 GPU, which outperforms the previous SOTA model YOLOv6-3.0-N with similar FPS by +2.4%. The PyTorch code is available at https://github.com/huawei-noah/Efficient-Computing/tree/master/Detection/Gold-YOLO, and the MindSpore code is available at https://gitee.com/mindspore/models/tree/master/research/cv/Gold_YOLO.
Authors: Steven Stalder, Michele Volpi, Nicolas Büttner, Stephen Law, Kenneth Harttgen, Esra Suel
Cities around the world face a critical shortage of affordable and decent housing. Despite its critical importance for policy, our ability to effectively monitor and track progress in urban housing is limited. Deep learning-based computer vision methods applied to street-level images have been successful in the measurement of socioeconomic and environmental inequalities but did not fully utilize temporal images to track urban change as time-varying labels are often unavailable. We used self-supervised methods to measure change in London using 15 million street images taken between 2008 and 2021. Our novel adaptation of Barlow Twins, Street2Vec, embeds urban structure while being invariant to seasonal and daily changes without manual annotations. It outperformed generic embeddings, successfully identified point-level change in London's housing supply from street-level images, and distinguished between major and minor change. This capability can provide timely information for urban planning and policy decisions toward more liveable, equitable, and sustainable cities.
Authors: Ruining Deng, Can Cui, Lucas W. Remedios, Shunxing Bao, R. Michael Womick, Sophie Chiron, Jia Li, Joseph T. Roland, Ken S. Lau, Qi Liu, Keith T. Wilson, Yaohong Wang, Lori A. Coburn, Bennett A. Landman, Yuankai Huo
Analyzing high resolution whole slide images (WSIs) with regard to information across multiple scales poses a significant challenge in digital pathology. Multi-instance learning (MIL) is a common solution for working with high resolution images by classifying bags of objects (i.e. sets of smaller image patches). However, such processing is typically performed at a single scale (e.g., 20x magnification) of WSIs, disregarding the vital inter-scale information that is key to diagnoses by human pathologists. In this study, we propose a novel cross-scale MIL algorithm to explicitly aggregate inter-scale relationships into a single MIL network for pathological image diagnosis. The contribution of this paper is three-fold: (1) A novel cross-scale MIL (CS-MIL) algorithm that integrates the multi-scale information and the inter-scale relationships is proposed; (2) A toy dataset with scale-specific morphological features is created and released to examine and visualize differential cross-scale attention; (3) Superior performance on both in-house and public datasets is demonstrated by our simple cross-scale MIL strategy. The official implementation is publicly available at https://github.com/hrlblab/CS-MIL.
Authors: Zhimin Sun, Shen Chen, Taiping Yao, Bangjie Yin, Ran Yi, Shouhong Ding, Lizhuang Ma
The challenge in sourcing attribution for forgery faces has gained widespread attention due to the rapid development of generative techniques. While many recent works have taken essential steps on GAN-generated faces, more threatening attacks related to identity swapping or expression transferring are still overlooked. And the forgery traces hidden in unknown attacks from the open-world unlabeled faces still remain under-explored. To push the related frontier research, we introduce a new benchmark called Open-World DeepFake Attribution (OW-DFA), which aims to evaluate attribution performance against various types of fake faces under open-world scenarios. Meanwhile, we propose a novel framework named Contrastive Pseudo Learning (CPL) for the OW-DFA task through 1) introducing a Global-Local Voting module to guide the feature alignment of forged faces with different manipulated regions, 2) designing a Confidence-based Soft Pseudo-label strategy to mitigate the pseudo-noise caused by similar methods in unlabeled set. In addition, we extend the CPL framework with a multi-stage paradigm that leverages pre-train technique and iterative learning to further enhance traceability performance. Extensive experiments verify the superiority of our proposed method on the OW-DFA and also demonstrate the interpretability of deepfake attribution task and its impact on improving the security of deepfake detection area.