Authors: Jianhua Wu, Bingzhao Gao, Jincheng Gao, Jianhao Yu, Hongqing Chu, Qiankun Yu, Xun Gong, Yi Chang, H. Eric Tseng, Hong Chen, Jie Chen
Abstract: With the development of artificial intelligence and breakthroughs in deep learning, large-scale Foundation Models (FMs), such as GPT, CLIP, etc., have achieved remarkable results in many fields including natural language processing and computer vision. The application of FMs in autonomous driving holds considerable promise. For example, they can contribute to enhance scene understanding and reasoning. By pre-training on rich linguistic and visual data, FMs can understand and interpret various elements in a driving scene, and provide cognitive reasoning to give linguistic and action commands for driving decisions and planning. Furthermore, FMs can augment data based on its understanding of driving scenarios to provide feasible scenes of those rare occurrences in the long tail distribution that are unlikely to be encountered during routine driving and data collection. The enhancement can subsequently lead to the improvement in the accuracy and reliability of autonomous driving systems. Another testament to the potential of FMs applications lies in the development of World Models, exemplified by the DREAMER series, which showcase the ability to comprehend physical laws and dynamics. Learning from massive data under the paradigm of self-supervised learning, World Model can generate unseen yet plausible driving environment, facilitating the enhancement in the prediction of road users behavior and the off-line training of driving strategies. In this paper, we synthesize the applications and future trends of FMs in autonomous driving. By utilizing the powerful capabilities of FMs, we strive to tackle the potential issues stemming from the long-tail distribution in autonomous driving, consequently advancing overall safety in this domain.
Authors: Arik Reuter, Anton Thielmann, Benjamin Saefken
Abstract: Understanding how images influence the world, interpreting which effects their semantics have on various quantities and exploring the reasons behind changes in image-based predictions are highly difficult yet extremely interesting problems. By adopting a holistic modeling approach utilizing Neural Additive Models in combination with Diffusion Autoencoders, we can effectively identify the latent hidden semantics of image effects and achieve full intelligibility of additional tabular effects. Our approach offers a high degree of flexibility, empowering us to comprehensively explore the impact of various image characteristics. We demonstrate that the proposed method can precisely identify complex image effects in an ablation study. To further showcase the practical applicability of our proposed model, we conduct a case study in which we investigate how the distinctive features and attributes captured within host images exert influence on the pricing of Airbnb rentals.
Authors: Prakash Chandra Chhipa, Meenakshi Subhash Chippa, Kanjar De, Rajkumar Saini, Marcus Liwicki, Mubarak Shah
Abstract: Perspective distortion (PD) causes unprecedented changes in shape, size, orientation, angles, and other spatial relationships of visual concepts in images. Precisely estimating camera intrinsic and extrinsic parameters is a challenging task that prevents synthesizing perspective distortion. Non-availability of dedicated training data poses a critical barrier to developing robust computer vision methods. Additionally, distortion correction methods make other computer vision tasks a multi-step approach and lack performance. In this work, we propose mitigating perspective distortion (MPD) by employing a fine-grained parameter control on a specific family of M\"obius transform to model real-world distortion without estimating camera intrinsic and extrinsic parameters and without the need for actual distorted data. Also, we present a dedicated perspectively distorted benchmark dataset, ImageNet-PD, to benchmark the robustness of deep learning models against this new dataset. The proposed method outperforms on existing benchmarks, ImageNet-E and ImageNet-X. Additionally, it significantly improves performance on ImageNet-PD while consistently performing on standard data distribution. Further, our method shows improved performance on three PD-affected real-world applications: crowd counting, fisheye image recognition, and person re-identification. We will release source code, dataset, and models for foster further research.
Authors: Maria Waheed, Michael Milford, Xiaojun Zhai, Maria Fasli, Klaus McDonald-Maier, Shoaib Ehsan
Abstract: Visual Place Recognition has been the subject of many endeavours utilizing different ensemble approaches to improve VPR performance. Ideas like multi-process fusion, Fly-Inspired Voting Units, SwitchHit or Switch-Fuse involve combining different VPR techniques together, utilizing different strategies. However, a major aspect often common to many of these strategies is voting. Voting is an extremely relevant topic to explore in terms of its application and significance for any ensemble VPR setup. This paper analyses several voting schemes to maximise the place detection accuracy of a VPR ensemble set up and determine the optimal voting schemes for selection. We take inspiration from a variety of voting schemes that are widely employed in fields such as politics and sociology and it is evident via empirical data that the selection of the voting method influences the results drastically. The paper tests a wide variety of voting schemes to present the improvement in the VPR results for several data sets. We aim to determine whether a single optimal voting scheme exists or, much like in other fields of research, the selection of a voting technique is relative to its application and environment. We propose a ranking of these different voting methods from best to worst which allows for better selection. While presenting our results in terms of voting method's performance bounds, in form of radar charts, PR curves to showcase the difference in performance and a comparison methodology using a McNemar test variant to determine the statistical significance of the differences. This test is performed to further confirm the reliability of outcomes and draw comparisons for better and informed selection a voting technique.
Authors: Pan Ting, Jianfeng Lin, Wenhao Yu, Wenlong Zhang, Xiaoying Chen, Jinlu Zhang, Binqiang Huang
Abstract: Object counting is a challenging task with broad application prospects in security surveillance, traffic management, and disease diagnosis. Existing object counting methods face a tri-fold challenge: achieving superior performance, maintaining high generalizability, and minimizing annotation costs. We develop a novel training-free class-agnostic object counter, TFCounter, which is prompt-context-aware via the cascade of the essential elements in large-scale foundation models. This approach employs an iterative counting framework with a dual prompt system to recognize a broader spectrum of objects varying in shape, appearance, and size. Besides, it introduces an innovative context-aware similarity module incorporating background context to enhance accuracy within messy scenes. To demonstrate cross-domain generalizability, we collect a novel counting dataset named BIKE-1000, including exclusive 1000 images of shared bicycles from Meituan. Extensive experiments on FSC-147, CARPK, and BIKE-1000 datasets demonstrate that TFCounter outperforms existing leading training-free methods and exhibits competitive results compared to trained counterparts.
Authors: Yannis Tevissen (ARMEDIA-SAMOVAR, ML), Khalil Guetari, Marine Tassel, Erwan Kerleroux, Fr\'ed\'eric Petitpont
Abstract: Image captioning models are widely used to describe recent and archived pictures with the objective of improving their accessibility and retrieval. Yet, these approaches tend to be inefficient and biased at retrieving people's names. In this work we introduce AstroCaptions, a dataset for the image captioning task. This dataset specifically contains thousands of public fig-ures that are complex to identify for a traditional model. We also propose a novel post-processing method to insert identified people's names inside the caption using explainable AI tools and the grounding capabilities of vi-sion-language models. The results obtained with this method show signifi-cant improvements of captions quality and a potential of reducing halluci-nations. Up to 93.2% of the persons detected can be inserted in the image captions leading to improvements in the BLEU, ROUGE, CIDEr and METEOR scores of each captioning model.
Authors: Mahmudul Islam Masum, Arif Sarwat, Hugo Riggs, Alicia Boymelgreen, Preyojon Dey
Abstract: This paper presents a comparative study of object detection using YOLOv5 and YOLOv8 for three distinct classes: artemia, cyst, and excrement. In this comparative study, we analyze the performance of these models in terms of accuracy, precision, recall, etc. where YOLOv5 often performed better in detecting Artemia and cysts with excellent precision and accuracy. However, when it came to detecting excrement, YOLOv5 faced notable challenges and limitations. This suggests that YOLOv8 offers greater versatility and adaptability in detection tasks while YOLOv5 may struggle in difficult situations and may need further fine-tuning or specialized training to enhance its performance. The results show insights into the suitability of YOLOv5 and YOLOv8 for detecting objects in challenging marine environments, with implications for applications such as ecological research.
Authors: Wenjing Shi, Phuong Tran, Sylvia Celed\'on-Pattichis, Marios S. Pattichis
Abstract: The paper develops datasets and methods to assess student participation in real-life collaborative learning environments. In collaborative learning environments, students are organized into small groups where they are free to interact within their group. Thus, students can move around freely causing issues with strong pose variation, move out and re-enter the camera scene, or face away from the camera. We formulate the problem of assessing student participation into two subproblems: (i) student group detection against strong background interference from other groups, and (ii) dynamic participant tracking within the group. A massive independent testing dataset of 12,518,250 student label instances, of total duration of 21 hours and 22 minutes of real-life videos, is used for evaluating the performance of our proposed method for student group detection. The proposed method of using multiple image representations is shown to perform equally or better than YOLO on all video instances. Over the entire dataset, the proposed method achieved an F1 score of 0.85 compared to 0.80 for YOLO. Following student group detection, the paper presents the development of a dynamic participant tracking system for assessing student group participation through long video sessions. The proposed dynamic participant tracking system is shown to perform exceptionally well, missing a student in just one out of 35 testing videos. In comparison, a state of the art method fails to track students in 14 out of the 35 testing videos. The proposed method achieves 82.3% accuracy on an independent set of long, real-life collaborative videos.
Authors: Adrien Le Coz, Houssem Ouertatani, St\'ephane Herbin, Faouzi Adjed
Abstract: Image classifiers should be used with caution in the real world. Performance evaluated on a validation set may not reflect performance in the real world. In particular, classifiers may perform well for conditions that are frequently encountered during training, but poorly for other infrequent conditions. In this study, we hypothesize that recent advances in text-to-image generative models make them valuable for benchmarking computer vision models such as image classifiers: they can generate images conditioned by textual prompts that cause classifier failures, allowing failure conditions to be described with textual attributes. However, their generation cost becomes an issue when a large number of synthetic images need to be generated, which is the case when many different attribute combinations need to be tested. We propose an image classifier benchmarking method as an iterative process that alternates image generation, classifier evaluation, and attribute selection. This method efficiently explores the attributes that ultimately lead to poor behavior detection.
Authors: Francesco Prinzi, Carmelo Militello, Calogero Zarcaro, Tommaso Vincenzo Bartolotta, Salvatore Gaglio, Salvatore Vitabile
Abstract: In the last years, artificial intelligence (AI) in clinical decision support systems (CDSS) played a key role in harnessing machine learning and deep learning architectures. Despite their promising capabilities, the lack of transparency and explainability of AI models poses significant challenges, particularly in medical contexts where reliability is a mandatory aspect. Achieving transparency without compromising predictive accuracy remains a key challenge. This paper presents a novel method, namely Rad4XCNN, to enhance the predictive power of CNN-derived features with the interpretability inherent in radiomic features. Rad4XCNN diverges from conventional methods based on saliency map, by associating intelligible meaning to CNN-derived features by means of Radiomics, offering new perspectives on explanation methods beyond visualization maps. Using a breast cancer classification task as a case study, we evaluated Rad4XCNN on ultrasound imaging datasets, including an online dataset and two in-house datasets for internal and external validation. Some key results are: i) CNN-derived features guarantee more robust accuracy when compared against ViT-derived and radiomic features; ii) conventional visualization map methods for explanation present several pitfalls; iii) Rad4XCNN does not sacrifice model accuracy for their explainability; iv) Rad4XCNN provides global explanation insights enabling the physician to analyze the model outputs and findings. In addition, we highlight the importance of integrating interpretability into AI models for enhanced trust and adoption in clinical practice, emphasizing how our method can mitigate some concerns related to explainable AI methods.
Authors: Yulin Luo, Ruichuan An, Bocheng Zou, Yiming Tang, Jiaming Liu, Shanghang Zhang
Abstract: The distribution of subpopulations is an important property hidden within a dataset. Uncovering and analyzing the subpopulation distribution within datasets provides a comprehensive understanding of the datasets, standing as a powerful tool beneficial to various downstream tasks, including Dataset Subpopulation Organization, Subpopulation Shift, and Slice Discovery. Despite its importance, there has been no work that systematically explores the subpopulation distribution of datasets to our knowledge. To address the limitation and solve all the mentioned tasks in a unified way, we introduce a novel concept of subpopulation structures to represent, analyze, and utilize subpopulation distributions within datasets. To characterize the structures in an interpretable manner, we propose the Subpopulation Structure Discovery with Large Language Models (SSD-LLM) framework, which employs world knowledge and instruction-following capabilities of Large Language Models (LLMs) to linguistically analyze informative image captions and summarize the structures. Furthermore, we propose complete workflows to address downstream tasks, named Task-specific Tuning, showcasing the application of the discovered structure to a spectrum of subpopulation-related tasks, including dataset subpopulation organization, subpopulation shift, and slice discovery. Furthermore, we propose complete workflows to address downstream tasks, named Task-specific Tuning, showcasing the application of the discovered structure to a spectrum of subpopulation-related tasks, including dataset subpopulation organization, subpopulation shift, and slice discovery.
Authors: Junchen Liu, Wenbo Hu, Zhuo Yang, Jianteng Chen, Guoliang Wang, Xiaoxue Chen, Yantong Cai, Huan-ang Gao, Hao Zhao
Abstract: Despite significant advancements in Neural Radiance Fields (NeRFs), the renderings may still suffer from aliasing and blurring artifacts, since it remains a fundamental challenge to effectively and efficiently characterize anisotropic areas induced by the cone-casting procedure. This paper introduces a Ripmap-Encoded Platonic Solid representation to precisely and efficiently featurize 3D anisotropic areas, achieving high-fidelity anti-aliasing renderings. Central to our approach are two key components: Platonic Solid Projection and Ripmap encoding. The Platonic Solid Projection factorizes the 3D space onto the unparalleled faces of a certain Platonic solid, such that the anisotropic 3D areas can be projected onto planes with distinguishable characterization. Meanwhile, each face of the Platonic solid is encoded by the Ripmap encoding, which is constructed by anisotropically pre-filtering a learnable feature grid, to enable featurzing the projected anisotropic areas both precisely and efficiently by the anisotropic area-sampling. Extensive experiments on both well-established synthetic datasets and a newly captured real-world dataset demonstrate that our Rip-NeRF attains state-of-the-art rendering quality, particularly excelling in the fine details of repetitive structures and textures, while maintaining relatively swift training times.
Authors: Stanislav Pidhorskyi, Tomas Simon, Gabriel Schwartz, He Wen, Yaser Sheikh, Jason Saragih
Abstract: Computing the gradients of a rendering process is paramount for diverse applications in computer vision and graphics. However, accurate computation of these gradients is challenging due to discontinuities and rendering approximations, particularly for surface-based representations and rasterization-based rendering. We present a novel method for computing gradients at visibility discontinuities for rasterization-based differentiable renderers. Our method elegantly simplifies the traditionally complex problem through a carefully designed approximation strategy, allowing for a straightforward, effective, and performant solution. We introduce a novel concept of micro-edges, which allows us to treat the rasterized images as outcomes of a differentiable, continuous process aligned with the inherently non-differentiable, discrete-pixel rasterization. This technique eliminates the necessity for rendering approximations or other modifications to the forward pass, preserving the integrity of the rendered image, which makes it applicable to rasterized masks, depth, and normals images where filtering is prohibitive. Utilizing micro-edges simplifies gradient interpretation at discontinuities and enables handling of geometry intersections, offering an advantage over the prior art. We showcase our method in dynamic human head scene reconstruction, demonstrating effective handling of camera images and segmentation masks.
Authors: Jiayang Shi, Junyi Zhu, Daniel M. Pelt, K. Joost Batenburg, Matthew B. Blaschko
Abstract: Computed Tomography (CT) is pivotal in industrial quality control and medical diagnostics. Sparse-view CT, offering reduced ionizing radiation, faces challenges due to its under-sampled nature, leading to ill-posed reconstruction problems. Recent advancements in Implicit Neural Representations (INRs) have shown promise in addressing sparse-view CT reconstruction. Recognizing that CT often involves scanning similar subjects, we propose a novel approach to improve reconstruction quality through joint reconstruction of multiple objects using INRs. This approach can potentially leverage both the strengths of INRs and the statistical regularities across multiple objects. While current INR joint reconstruction techniques primarily focus on accelerating convergence via meta-initialization, they are not specifically tailored to enhance reconstruction quality. To address this gap, we introduce a novel INR-based Bayesian framework integrating latent variables to capture the inter-object relationships. These variables serve as a dynamic reference throughout the optimization, thereby enhancing individual reconstruction fidelity. Our extensive experiments, which assess various key factors such as reconstruction quality, resistance to overfitting, and generalizability, demonstrate significant improvements over baselines in common numerical metrics. This underscores a notable advancement in CT reconstruction methods.
Authors: Yohei Nakayama, Jiawei Su
Abstract: Currently, the foundation models represented by large language models have made dramatic progress and are used in a very wide range of domains including 2D and 3D vision. As one of the important application domains of foundation models, earth observation has attracted attention and various approaches have been developed. When considering earth observation as a single image capture, earth observation imagery can be processed as an image with three or more channels, and when it comes with multiple image captures of different timestamps at one location, the temporal observation can be considered as a set of continuous image resembling video frames or medical SCAN slices. This paper presents Spatio-Temporal SwinMAE (ST-SwinMAE), an architecture which particularly focuses on representation learning for spatio-temporal image processing. Specifically, it uses a hierarchical Masked Auto-encoder (MAE) with Video Swin Transformer blocks. With the architecture, we present a pretrained model named Degas 100M as a geospatial foundation model. Also, we propose an approach for transfer learning with Degas 100M, which both pretrained encoder and decoder of MAE are utilized with skip connections added between them to achieve multi-scale information communication, forms an architecture named Spatio-Temporal SwinUNet (ST-SwinUNet). Our approach shows significant improvements of performance over existing state-of-the-art of foundation models. Specifically, for transfer learning of the land cover downstream task on the PhilEO Bench dataset, it shows 10.4\% higher accuracy compared with other geospatial foundation models on average.
Authors: Jiayang Shi, Daniel M. Pelt, K. Joost Batenburg
Abstract: Computed tomography (CT) is a widely used non-invasive medical imaging technique for disease diagnosis. The diagnostic accuracy is often affected by image resolution, which can be insufficient in practice. For medical CT images, the through-plane resolution is often worse than the in-plane resolution and there can be overlap between slices, causing difficulties in diagnoses. Self-supervised methods for through-plane resolution enhancement, which train on in-plane images and infer on through-plane images, have shown promise for both CT and MRI imaging. However, existing self-supervised methods either neglect overlap or can only handle specific cases with fixed combinations of resolution and overlap. To address these limitations, we propose a self-supervised method called SR4ZCT. It employs the same off-axis training approach while being capable of handling arbitrary combinations of resolution and overlap. Our method explicitly models the relationship between resolutions and voxel spacings of different planes to accurately simulate training images that match the original through-plane images. We highlight the significance of accurate modeling in self-supervised off-axis training and demonstrate the effectiveness of SR4ZCT using a real-world dataset.
Authors: Meiqi Cao, Rui Yan, Xiangbo Shu, Guangzhao Dai, Yazhou Yao, Guo-Sen Xie
Abstract: Panoramic Activity Recognition (PAR) aims to identify multi-granularity behaviors performed by multiple persons in panoramic scenes, including individual activities, group activities, and global activities. Previous methods 1) heavily rely on manually annotated detection boxes in training and inference, hindering further practical deployment; or 2) directly employ normal detectors to detect multiple persons with varying size and spatial occlusion in panoramic scenes, blocking the performance gain of PAR. To this end, we consider learning a detector adapting varying-size occluded persons, which is optimized along with the recognition module in the all-in-one framework. Therefore, we propose a novel Adapt-Focused bi-Propagating Prototype learning (AdaFPP) framework to jointly recognize individual, group, and global activities in panoramic activity scenes by learning an adapt-focused detector and multi-granularity prototypes as the pretext tasks in an end-to-end way. Specifically, to accommodate the varying sizes and spatial occlusion of multiple persons in crowed panoramic scenes, we introduce a panoramic adapt-focuser, achieving the size-adapting detection of individuals by comprehensively selecting and performing fine-grained detections on object-dense sub-regions identified through original detections. In addition, to mitigate information loss due to inaccurate individual localizations, we introduce a bi-propagation prototyper that promotes closed-loop interaction and informative consistency across different granularities by facilitating bidirectional information propagation among the individual, group, and global levels. Extensive experiments demonstrate the significant performance of AdaFPP and emphasize its powerful applicability for PAR.
Authors: Jordan A. James, Heather K. Manching, Amanda M. Hulse-Kemp, William J. Beksi
Abstract: Advancements in machine learning, computer vision, and robotics have paved the way for transformative solutions in various domains, particularly in agriculture. For example, accurate identification and segmentation of fruits from field images plays a crucial role in automating jobs such as harvesting, disease detection, and yield estimation. However, achieving robust and precise infield fruit segmentation remains a challenging task since large amounts of labeled data are required to handle variations in fruit size, shape, color, and occlusion. In this paper, we develop a few-shot semantic segmentation framework for infield fruits using transfer learning. Concretely, our work is aimed at addressing agricultural domains that lack publicly available labeled data. Motivated by similar success in urban scene parsing, we propose specialized pre-training using a public benchmark dataset for fruit transfer learning. By leveraging pre-trained neural networks, accurate semantic segmentation of fruit in the field is achieved with only a few labeled images. Furthermore, we show that models with pre-training learn to distinguish between fruit still on the trees and fruit that have fallen on the ground, and they can effectively transfer the knowledge to the target fruit dataset.
Authors: Zhenan Shao, Linjian Ma, Bo Li, Diane M. Beck
Abstract: Human object recognition exhibits remarkable resilience in cluttered and dynamic visual environments. In contrast, despite their unparalleled performance across numerous visual tasks, Deep Neural Networks (DNNs) remain far less robust than humans, showing, for example, a surprising susceptibility to adversarial attacks involving image perturbations that are (almost) imperceptible to humans. Human object recognition likely owes its robustness, in part, to the increasingly resilient representations that emerge along the hierarchy of the ventral visual cortex. Here we show that DNNs, when guided by neural representations from a hierarchical sequence of regions in the human ventral visual stream, display increasing robustness to adversarial attacks. These neural-guided models also exhibit a gradual shift towards more human-like decision-making patterns and develop hierarchically smoother decision surfaces. Importantly, the resulting representational spaces differ in important ways from those produced by conventional smoothing methods, suggesting that such neural-guidance may provide previously unexplored robustness solutions. Our findings support the gradual emergence of human robustness along the ventral visual hierarchy and suggest that the key to DNN robustness may lie in increasing emulation of the human brain.
Authors: Hyunseo Kim, Hyeonseo Yang, Taekyung Kim, YoonSung Kim, Jin-Hwa Kim, Byoung-Tak Zhang
Abstract: Active learning in 3D scene reconstruction has been widely studied, as selecting informative training views is critical for the reconstruction. Recently, Neural Radiance Fields (NeRF) variants have shown performance increases in active 3D reconstruction using image rendering or geometric uncertainty. However, the simultaneous consideration of both uncertainties in selecting informative views remains unexplored, while utilizing different types of uncertainty can reduce the bias that arises in the early training stage with sparse inputs. In this paper, we propose ActiveNeuS, which evaluates candidate views considering both uncertainties. ActiveNeuS provides a way to accumulate image rendering uncertainty while avoiding the bias that the estimated densities can introduce. ActiveNeuS computes the neural implicit surface uncertainty, providing the color uncertainty along with the surface information. It efficiently handles the bias by using the surface information and a grid, enabling the fast selection of diverse viewpoints. Our method outperforms previous works on popular datasets, Blender and DTU, showing that the views selected by ActiveNeuS significantly improve performance.
Authors: Soumyadeep Chandra, Sayeed Shafayet Chowdhury, Courtney Yong, Chandru P. Sundaram, Kaushik Roy
Abstract: Surgical action localization is a challenging computer vision problem. While it has promising applications including automated training of surgery procedures, surgical workflow optimization, etc., appropriate model design is pivotal to accomplishing this task. Moreover, the lack of suitable medical datasets adds an additional layer of complexity. To that effect, we introduce a new complex dataset of nephrectomy surgeries called UroSlice. To perform the action localization from these videos, we propose a novel model termed as `ViTALS' (Vision Transformer for Action Localization in Surgical Nephrectomy). Our model incorporates hierarchical dilated temporal convolution layers and inter-layer residual connections to capture the temporal correlations at finer as well as coarser granularities. The proposed approach achieves state-of-the-art performance on Cholec80 and UroSlice datasets (89.8% and 66.1% accuracy, respectively), validating its effectiveness.
Authors: Niccol\`o Biondi, Federico Pernici, Simone Ricci, Alberto Del Bimbo
Abstract: Learning compatible representations enables the interchangeable use of semantic features as models are updated over time. This is particularly relevant in search and retrieval systems where it is crucial to avoid reprocessing of the gallery images with the updated model. While recent research has shown promising empirical evidence, there is still a lack of comprehensive theoretical understanding about learning compatible representations. In this paper, we demonstrate that the stationary representations learned by the $d$-Simplex fixed classifier optimally approximate compatibility representation according to the two inequality constraints of its formal definition. This not only establishes a solid foundation for future works in this line of research but also presents implications that can be exploited in practical learning scenarios. An exemplary application is the now-standard practice of downloading and fine-tuning new pre-trained models. Specifically, we show the strengths and critical issues of stationary representations in the case in which a model undergoing sequential fine-tuning is asynchronously replaced by downloading a better-performing model pre-trained elsewhere. Such a representation enables seamless delivery of retrieval service (i.e., no reprocessing of gallery images) and offers improved performance without operational disruptions during model replacement. Code available at: https://github.com/miccunifi/iamcl2r.
Authors: Siyuan Yan, Cheng Luo, Zhen Yu, Zongyuan Ge
Abstract: Vision-language foundation models like CLIP have shown impressive zero-shot generalization, but finetuning on downstream datasets can cause overfitting and loss of its generalization ability on unseen domains. Although collecting additional data from new domains of interest is possible, this method is often impractical due to the challenges in obtaining annotated data. To address this, we propose a plug-and-play feature augmentation method called LDFS (Language-Guided Diverse Feature Synthesis) to synthesize new domain features and improve existing CLIP fine-tuning strategies. LDFS has three main contributions: 1) To synthesize novel domain features and promote diversity, we propose an instance-conditional feature augmentation strategy based on a textguided feature augmentation loss. 2) To maintain feature quality after augmenting, we introduce a pairwise regularizer to preserve augmented feature coherence within the CLIP feature space. 3) We propose to use stochastic text feature augmentation to reduce the modality gap and further facilitate the process of text-guided feature synthesis. Extensive experiments show LDFS superiority in improving CLIP generalization ability on unseen domains without collecting data from those domains. The code will be made publicly available.
Authors: Shuqi Shen, Junjie Yang
Abstract: Safety helmets play a crucial role in protecting workers from head injuries in construction sites, where potential hazards are prevalent. However, currently, there is no approach that can simultaneously achieve both model accuracy and performance in complex environments. In this study, we utilized a Yolo-based model for safety helmet detection, achieved a 2% improvement in mAP (mean Average Precision) performance while reducing parameters and Flops count by over 25%. YOLO(You Only Look Once) is a widely used, high-performance, lightweight model architecture that is well suited for complex environments. We presents a novel approach by incorporating a lightweight feature extraction network backbone based on GhostNetv2, integrating attention modules such as Spatial Channel-wise Attention Net(SCNet) and Coordination Attention Net(CANet), and adopting the Gradient Norm Aware optimizer (GAM) for improved generalization ability. In safety-critical environments, the accurate detection and speed of safety helmets plays a pivotal role in preventing occupational hazards and ensuring compliance with safety protocols. This work addresses the pressing need for robust and efficient helmet detection methods, offering a comprehensive framework that not only enhances accuracy but also improves the adaptability of detection models to real-world conditions. Our experimental results underscore the synergistic effects of GhostNetv2, attention modules, and the GAM optimizer, presenting a compelling solution for safety helmet detection that achieves superior performance in terms of accuracy, generalization, and efficiency.
Authors: Yanan Zhang, Jinqing Zhang, Zengran Wang, Junhao Xu, Di Huang
Abstract: In recent years, autonomous driving has garnered escalating attention for its potential to relieve drivers' burdens and improve driving safety. Vision-based 3D occupancy prediction, which predicts the spatial occupancy status and semantics of 3D voxel grids around the autonomous vehicle from image inputs, is an emerging perception task suitable for cost-effective perception system of autonomous driving. Although numerous studies have demonstrated the greater advantages of 3D occupancy prediction over object-centric perception tasks, there is still a lack of a dedicated review focusing on this rapidly developing field. In this paper, we first introduce the background of vision-based 3D occupancy prediction and discuss the challenges in this task. Secondly, we conduct a comprehensive survey of the progress in vision-based 3D occupancy prediction from three aspects: feature enhancement, deployment friendliness and label efficiency, and provide an in-depth analysis of the potentials and challenges of each category of methods. Finally, we present a summary of prevailing research trends and propose some inspiring future outlooks. To provide a valuable reference for researchers, a regularly updated collection of related papers, datasets, and codes is organized at https://github.com/zya3d/Awesome-3D-Occupancy-Prediction.
URLs: https://github.com/zya3d/Awesome-3D-Occupancy-Prediction.
Authors: Shuai Yuan, Lei Luo, Zhuo Hui, Can Pu, Xiaoyu Xiang, Rakesh Ranjan, Denis Demandolx
Abstract: Traditional unsupervised optical flow methods are vulnerable to occlusions and motion boundaries due to lack of object-level information. Therefore, we propose UnSAMFlow, an unsupervised flow network that also leverages object information from the latest foundation model Segment Anything Model (SAM). We first include a self-supervised semantic augmentation module tailored to SAM masks. We also analyze the poor gradient landscapes of traditional smoothness losses and propose a new smoothness definition based on homography instead. A simple yet effective mask feature module has also been added to further aggregate features on the object level. With all these adaptations, our method produces clear optical flow estimation with sharp boundaries around objects, which outperforms state-of-the-art methods on both KITTI and Sintel datasets. Our method also generalizes well across domains and runs very efficiently.
Authors: Joaquim Comas, Adria Ruiz, Federico Sukno
Abstract: Recent advancements in remote heart rate measurement (rPPG), motivated by data-driven approaches, have significantly improved accuracy. However, certain challenges, such as video compression, still remain: recovering the rPPG signal from highly compressed videos is particularly complex. Although several studies have highlighted the difficulties and impact of video compression for this, effective solutions remain limited. In this paper, we present a novel approach to address the impact of video compression on rPPG estimation, which leverages a pulse-signal magnification transformation to adapt compressed videos to an uncompressed data domain in which the rPPG signal is magnified. We validate the effectiveness of our model by exhaustive evaluations on two publicly available datasets, UCLA-rPPG and UBFC-rPPG, employing both intra- and cross-database performance at several compression rates. Additionally, we assess the robustness of our approach on two additional highly compressed and widely-used datasets, MAHNOB-HCI and COHFACE, which reveal outstanding heart rate estimation results.
Authors: Haoyu Hu, Xinyu Yi, Zhe Cao, Jun-Hai Yong, Feng Xu
Abstract: Hand manipulating objects is an important interaction motion in our daily activities. We faithfully reconstruct this motion with a single RGBD camera by a novel deep reinforcement learning method to leverage physics. Firstly, we propose object compensation control which establishes direct object control to make the network training more stable. Meanwhile, by leveraging the compensation force and torque, we seamlessly upgrade the simple point contact model to a more physical-plausible surface contact model, further improving the reconstruction accuracy and physical correctness. Experiments indicate that without involving any heuristic physical rules, this work still successfully involves physics in the reconstruction of hand-object interactions which are complex motions hard to imitate with deep reinforcement learning. Our code and data are available at https://github.com/hu-hy17/HOIC.
Authors: Yik San Cheng, Runkai Zhao, Heng Wang, Hanchuan Peng, Weidong Cai
Abstract: Neuron reconstruction, one of the fundamental tasks in neuroscience, rebuilds neuronal morphology from 3D light microscope imaging data. It plays a critical role in analyzing the structure-function relationship of neurons in the nervous system. However, due to the scarcity of neuron datasets and high-quality SWC annotations, it is still challenging to develop robust segmentation methods for single neuron reconstruction. To address this limitation, we aim to distill the consensus knowledge from massive natural image data to aid the segmentation model in learning the complex neuron structures. Specifically, in this work, we propose a novel training paradigm that leverages a 2D Vision Transformer model pre-trained on large-scale natural images to initialize our Transformer-based 3D neuron segmentation model with a tailored 2D-to-3D weight transferring strategy. Our method builds a knowledge sharing connection between the abundant natural and the scarce neuron image domains to improve the 3D neuron segmentation ability in a data-efficiency manner. Evaluated on a popular benchmark, BigNeuron, our method enhances neuron segmentation performance by 8.71% over the model trained from scratch with the same amount of training samples.
Authors: Yang Lei, Luke A. Matkovic, Justin Roper, Tonghe Wang, Jun Zhou, Beth Ghavidel, Mark McDonald, Pretesh Patel, Xiaofeng Yang
Abstract: This paper aims to create a deep learning framework that can estimate the deformation vector field (DVF) for directly registering abdominal MRI-CT images. The proposed method assumed a diffeomorphic deformation. By using topology-preserved deformation features extracted from the probabilistic diffeomorphic registration model, abdominal motion can be accurately obtained and utilized for DVF estimation. The model integrated Swin transformers, which have demonstrated superior performance in motion tracking, into the convolutional neural network (CNN) for deformation feature extraction. The model was optimized using a cross-modality image similarity loss and a surface matching loss. To compute the image loss, a modality-independent neighborhood descriptor (MIND) was used between the deformed MRI and CT images. The surface matching loss was determined by measuring the distance between the warped coordinates of the surfaces of contoured structures on the MRI and CT images. The deformed MRI image was assessed against the CT image using the target registration error (TRE), Dice similarity coefficient (DSC), and mean surface distance (MSD) between the deformed contours of the MRI image and manual contours of the CT image. When compared to only rigid registration, DIR with the proposed method resulted in an increase of the mean DSC values of the liver and portal vein from 0.850 and 0.628 to 0.903 and 0.763, a decrease of the mean MSD of the liver from 7.216 mm to 3.232 mm, and a decrease of the TRE from 26.238 mm to 8.492 mm. The proposed deformable image registration method based on a diffeomorphic transformer provides an effective and efficient way to generate an accurate DVF from an MRI-CT image pair of the abdomen. It could be utilized in the current treatment planning workflow for liver radiotherapy.
Authors: Andong Lu, Wanyu Wang, Chenglong Li, Jin Tang, Bin Luo
Abstract: Multi-modal feature fusion as a core investigative component of RGBT tracking emerges numerous fusion studies in recent years. However, existing RGBT tracking methods widely adopt fixed fusion structures to integrate multi-modal feature, which are hard to handle various challenges in dynamic scenarios. To address this problem, this work presents a novel \emph{A}ttention-based \emph{F}usion rou\emph{ter} called AFter, which optimizes the fusion structure to adapt to the dynamic challenging scenarios, for robust RGBT tracking. In particular, we design a fusion structure space based on the hierarchical attention network, each attention-based fusion unit corresponding to a fusion operation and a combination of these attention units corresponding to a fusion structure. Through optimizing the combination of attention-based fusion units, we can dynamically select the fusion structure to adapt to various challenging scenarios. Unlike complex search of different structures in neural architecture search algorithms, we develop a dynamic routing algorithm, which equips each attention-based fusion unit with a router, to predict the combination weights for efficient optimization of the fusion structure. Extensive experiments on five mainstream RGBT tracking datasets demonstrate the superior performance of the proposed AFter against state-of-the-art RGBT trackers. We release the code in https://github.com/Alexadlu/AFter.
Authors: Yuchuan Tian, Zhijun Tu, Hanting Chen, Jie Hu, Chao Xu, Yunhe Wang
Abstract: Diffusion Transformers (DiTs) introduce the transformer architecture to diffusion tasks for latent-space image generation. With an isotropic architecture that chains a series of transformer blocks, DiTs demonstrate competitive performance and good scalability; but meanwhile, the abandonment of U-Net by DiTs and their following improvements is worth rethinking. To this end, we conduct a simple toy experiment by comparing a U-Net architectured DiT with an isotropic one. It turns out that the U-Net architecture only gain a slight advantage amid the U-Net inductive bias, indicating potential redundancies within the U-Net-style DiT. Inspired by the discovery that U-Net backbone features are low-frequency-dominated, we perform token downsampling on the query-key-value tuple for self-attention and bring further improvements despite a considerable amount of reduction in computation. Based on self-attention with downsampled tokens, we propose a series of U-shaped DiTs (U-DiTs) in the paper and conduct extensive experiments to demonstrate the extraordinary performance of U-DiT models. The proposed U-DiT could outperform DiT-XL/2 with only 1/6 of its computation cost. Codes are available at https://github.com/YuchuanTian/U-DiT.
Authors: Eren Tahir, Mert Bal
Abstract: While image forensics is concerned with whether an image has been tampered with, image anti-forensics attempts to prevent image forensics methods from detecting tampered images. The competition between these two fields started long before the advancement of deep learning. JPEG compression, blurring and noising, which are simple methods by today's standards, have long been used for anti-forensics and have been the subject of much research in both forensics and anti-forensics. Although these traditional methods are old, they make it difficult to detect fake images and are used for data augmentation in training deep image forgery detection models. In addition to making the image difficult to detect, these methods leave traces on the image and consequently degrade the image quality. Separate image forensics methods have also been developed to detect these traces. In this study, we go one step further and improve the image quality after these methods with deep image restoration models and make it harder to detect the forged image. We evaluate the impact of these methods on image quality. We then test both our proposed methods with deep learning and methods without deep learning on the two best existing image manipulation detection models. In the obtained results, we show how existing image forgery detection models fail against the proposed methods. Code implementation will be publicly available at https://github.com/99eren99/DIRFIAF .
Authors: Christopher Maxey, Jaehoon Choi, Yonghan Lee, Hyungtae Lee, Dinesh Manocha, Heesung Kwon
Abstract: In this paper, we present a new approach to bridge the domain gap between synthetic and real-world data for un- manned aerial vehicle (UAV)-based perception. Our formu- lation is designed for dynamic scenes, consisting of moving objects or human actions, where the goal is to recognize the pose or actions. We propose an extension of K-Planes Neural Radiance Field (NeRF), wherein our algorithm stores a set of tiered feature vectors. The tiered feature vectors are generated to effectively model conceptual information about a scene as well as an image decoder that transforms output feature maps into RGB images. Our technique leverages the information amongst both static and dynamic objects within a scene and is able to capture salient scene attributes of high altitude videos. We evaluate its performance on challenging datasets, including Okutama Action and UG2, and observe considerable improvement in accuracy over state of the art aerial perception algorithms.
Authors: Vishal Nedungadi, Ankit Kariryaa, Stefan Oehmcke, Serge Belongie, Christian Igel, Nico Lang
Abstract: The volume of unlabelled Earth observation (EO) data is huge, but many important applications lack labelled training data. However, EO data offers the unique opportunity to pair data from different modalities and sensors automatically based on geographic location and time, at virtually no human labor cost. We seize this opportunity to create a diverse multi-modal pretraining dataset at global scale. Using this new corpus of 1.2 million locations, we propose a Multi-Pretext Masked Autoencoder (MP-MAE) approach to learn general-purpose representations for optical satellite images. Our approach builds on the ConvNeXt V2 architecture, a fully convolutional masked autoencoder (MAE). Drawing upon a suite of multi-modal pretext tasks, we demonstrate that our MP-MAE approach outperforms both MAEs pretrained on ImageNet and MAEs pretrained on domain-specific satellite images. This is shown on several downstream tasks including image classification and semantic segmentation. We find that multi-modal pretraining notably improves the linear probing performance, e.g. 4pp on BigEarthNet and 16pp on So2Sat, compared to pretraining on optical satellite images only. We show that this also leads to better label and parameter efficiency which are crucial aspects in global scale applications.
Authors: Di Liu, Bingbing Zhuang, Dimitris N. Metaxas, Manmohan Chandraker
Abstract: The perception of 3D motion of surrounding traffic participants is crucial for driving safety. While existing works primarily focus on general large motions, we contend that the instantaneous detection and quantification of subtle motions is equally important as they indicate the nuances in driving behavior that may be safety critical, such as behaviors near a stop sign of parking positions. We delve into this under-explored task, examining its unique challenges and developing our solution, accompanied by a carefully designed benchmark. Specifically, due to the lack of correspondences between consecutive frames of sparse Lidar point clouds, static objects might appear to be moving - the so-called swimming effect. This intertwines with the true object motion, thereby posing ambiguity in accurate estimation, especially for subtle motions. To address this, we propose to leverage local occupancy completion of object point clouds to densify the shape cue, and mitigate the impact of swimming artifacts. The occupancy completion is learned in an end-to-end fashion together with the detection of moving objects and the estimation of their motion, instantaneously as soon as objects start to move. Extensive experiments demonstrate superior performance compared to standard 3D motion estimation approaches, particularly highlighting our method's specialized treatment of subtle motions.
Authors: David A. Wood, Emily Guilhem, Sina Kafiabadi, Ayisha Al Busaidi, Kishan Dissanayake, Ahmed Hammam, Nina Mansoor, Matthew Townend, Siddharth Agarwal, Yiran Wei, Asif Mazumder, Gareth J. Barker, Peter Sasieni, Sebastien Ourselin, James H. Cole, Thomas C. Booth
Abstract: Artificial neural networks trained on large, expert-labelled datasets are considered state-of-the-art for a range of medical image recognition tasks. However, categorically labelled datasets are time-consuming to generate and constrain classification to a pre-defined, fixed set of classes. For neuroradiological applications in particular, this represents a barrier to clinical adoption. To address these challenges, we present a self-supervised text-vision framework that learns to detect clinically relevant abnormalities in brain MRI scans by directly leveraging the rich information contained in accompanying free-text neuroradiology reports. Our training approach consisted of two-steps. First, a dedicated neuroradiological language model - NeuroBERT - was trained to generate fixed-dimensional vector representations of neuroradiology reports (N = 50,523) via domain-specific self-supervised learning tasks. Next, convolutional neural networks (one per MRI sequence) learnt to map individual brain scans to their corresponding text vector representations by optimising a mean square error loss. Once trained, our text-vision framework can be used to detect abnormalities in unreported brain MRI examinations by scoring scans against suitable query sentences (e.g., 'there is an acute stroke', 'there is hydrocephalus' etc.), enabling a range of classification-based applications including automated triage. Potentially, our framework could also serve as a clinical decision support tool, not only by suggesting findings to radiologists and detecting errors in provisional reports, but also by retrieving and displaying examples of pathologies from historical examinations that could be relevant to the current case based on textual descriptors.
Authors: Junjiang Zhen, Bojun Xie
Abstract: Deep learning has had a significant impact on the identification and classification of mineral resources, especially playing a key role in efficiently and accurately identifying different minerals, which is important for improving the efficiency and accuracy of mining. However, traditional ore sorting meth- ods often suffer from inefficiency and lack of accuracy, especially in complex mineral environments. To address these challenges, this study proposes a method called OreYOLO, which incorporates an attentional mechanism and a multi-scale feature fusion strategy, based on ore data from gold and sul- fide ores. By introducing the progressive feature pyramid structure into YOLOv5 and embedding the attention mechanism in the feature extraction module, the detection performance and accuracy of the model are greatly improved. In order to adapt to the diverse ore sorting scenarios and the deployment requirements of edge devices, the network structure is designed to be lightweight, which achieves a low number of parameters (3.458M) and computational complexity (6.3GFLOPs) while maintaining high accuracy (99.3% and 99.2%, respectively). In the experimental part, a target detection dataset containing 6000 images of gold and sulfuric iron ore is constructed for gold and sulfuric iron ore classification training, and several sets of comparison experiments are set up, including the YOLO series, EfficientDet, Faster-RCNN, and CenterNet, etc., and the experiments prove that OreYOLO outperforms the commonly used high-performance object detection of these architectures
Authors: Javeria Shabbir, Muhammad Zeshan. Alam, M. Umair Mukati
Abstract: Light field (LF) imaging captures both angular and spatial light distributions, enabling advanced photographic techniques. However, micro-lens array (MLA)- based cameras face a spatial-angular resolution tradeoff due to a single shared sensor. We propose a novel light field framework for resolution enhancement, employing a modular approach. The first module generates a high-resolution, all-in-focus image. The second module, a texture transformer network, enhances the resolution of each light field perspective independently using the output of the first module as a reference image. The final module leverages light field regularity to jointly improve resolution across all LF image perspectives. Our approach demonstrates superior performance to existing methods in both qualitative and quantitative evaluations.
Authors: Mengxian Hu, Minghao Zhu, Xun Zhou, Qingqing Yan, Shu Li, Chengju Liu, Qijun Chen
Abstract: Motion diffusion models have recently proven successful for text-driven human motion generation. Despite their excellent generation performance, they are challenging to infer in real time due to the multi-step sampling mechanism that involves tens or hundreds of repeat function evaluation iterations. To this end, we investigate a motion latent consistency Training (MLCT) for motion generation to alleviate the computation and time consumption during iteration inference. It applies diffusion pipelines to low-dimensional motion latent spaces to mitigate the computational burden of each function evaluation. Explaining the diffusion process with probabilistic flow ordinary differential equation (PF-ODE) theory, the MLCT allows extremely few steps infer between the prior distribution to the motion latent representation distribution via maintaining consistency of the outputs over the trajectory of PF-ODE. Especially, we introduce a quantization constraint to optimize motion latent representations that are bounded, regular, and well-reconstructed compared to traditional variational constraints. Furthermore, we propose a conditional PF-ODE trajectory simulation method, which improves the conditional generation performance with minimal additional training costs. Extensive experiments on two human motion generation benchmarks show that the proposed model achieves state-of-the-art performance with less than 10\% time cost.
Authors: Muhammad Zeshan Alam
Abstract: This paper introduces a novel approach for enhanced lane detection by integrating spatial, angular, and temporal information through light field imaging and novel deep learning models. Utilizing lenslet-inspired 2D light field representations and LSTM networks, our method significantly improves lane detection in challenging conditions. We demonstrate the efficacy of this approach with modified CNN architectures, showing superior per- formance over traditional methods. Our findings suggest this integrated data approach could advance lane detection technologies and inspire new models that leverage these multidimensional insights for autonomous vehicle percep- tion.
Authors: Roopal Garg, Andrea Burns, Burcu Karagol Ayan, Yonatan Bitton, Ceslee Montgomery, Yasumasa Onoe, Andrew Bunner, Ranjay Krishna, Jason Baldridge, Radu Soricut
Abstract: Despite the longstanding adage "an image is worth a thousand words," creating accurate and hyper-detailed image descriptions for training Vision-Language models remains challenging. Current datasets typically have web-scraped descriptions that are short, low-granularity, and often contain details unrelated to the visual content. As a result, models trained on such data generate descriptions replete with missing information, visual inconsistencies, and hallucinations. To address these issues, we introduce ImageInWords (IIW), a carefully designed human-in-the-loop annotation framework for curating hyper-detailed image descriptions and a new dataset resulting from this process. We validate the framework through evaluations focused on the quality of the dataset and its utility for fine-tuning with considerations for readability, comprehensiveness, specificity, hallucinations, and human-likeness. Our dataset significantly improves across these dimensions compared to recently released datasets (+66%) and GPT-4V outputs (+48%). Furthermore, models fine-tuned with IIW data excel by +31% against prior work along the same human evaluation dimensions. Given our fine-tuned models, we also evaluate text-to-image generation and vision-language reasoning. Our model's descriptions can generate images closest to the original, as judged by both automated and human metrics. We also find our model produces more compositionally rich descriptions, outperforming the best baseline by up to 6% on ARO, SVO-Probes, and Winoground datasets.
Authors: Zhixiang Chi, Li Gu, Tao Zhong, Huan Liu, Yuanhao Yu, Konstantinos N Plataniotis, Yang Wang
Abstract: In this paper, we aim to adapt a model at test-time using a few unlabeled data to address distribution shifts. To tackle the challenges of extracting domain knowledge from a limited amount of data, it is crucial to utilize correlated information from pre-trained backbones and source domains. Previous studies fail to utilize recent foundation models with strong out-of-distribution generalization. Additionally, domain-centric designs are not flavored in their works. Furthermore, they employ the process of modelling source domains and the process of learning to adapt independently into disjoint training stages. In this work, we propose an approach on top of the pre-computed features of the foundation model. Specifically, we build a knowledge bank to learn the transferable knowledge from source domains. Conditioned on few-shot target data, we introduce a domain prompt generator to condense the knowledge bank into a domain-specific prompt. The domain prompt then directs the visual features towards a particular domain via a guidance module. Moreover, we propose a domain-aware contrastive loss and employ meta-learning to facilitate domain knowledge extraction. Extensive experiments are conducted to validate the domain knowledge extraction. The proposed method outperforms previous work on 5 large-scale benchmarks including WILDS and DomainNet.
Authors: Zhaoqi Leng, Pei Sun, Tong He, Dragomir Anguelov, Mingxing Tan
Abstract: 3D object detectors for point clouds often rely on a pooling-based PointNet to encode sparse points into grid-like voxels or pillars. In this paper, we identify that the common PointNet design introduces an information bottleneck that limits 3D object detection accuracy and scalability. To address this limitation, we propose PVTransformer: a transformer-based point-to-voxel architecture for 3D detection. Our key idea is to replace the PointNet pooling operation with an attention module, leading to a better point-to-voxel aggregation function. Our design respects the permutation invariance of sparse 3D points while being more expressive than the pooling-based PointNet. Experimental results show our PVTransformer achieves much better performance compared to the latest 3D object detectors. On the widely used Waymo Open Dataset, our PVTransformer achieves state-of-the-art 76.5 mAPH L2, outperforming the prior art of SWFormer by +1.7 mAPH L2.
Authors: Zhusi Zhong, Jie Li, Zhuoqi Ma, Scott Collins, Harrison Bai, Paul Zhang, Terrance Healey, Xinbo Gao, Michael K. Atalay, Zhicheng Jiao
Abstract: The COVID-19 pandemic has strained global public health, necessitating accurate diagnosis and intervention to control disease spread and reduce mortality rates. This paper introduces an interpretable deep survival prediction model designed specifically for improved understanding and trust in COVID-19 prognosis using chest X-ray (CXR) images. By integrating a large-scale pretrained image encoder, Risk-specific Grad-CAM, and anatomical region detection techniques, our approach produces regional interpretable outcomes that effectively capture essential disease features while focusing on rare but critical abnormal regions. Our model's predictive results provide enhanced clarity and transparency through risk area localization, enabling clinicians to make informed decisions regarding COVID-19 diagnosis with better understanding of prognostic insights. We evaluate the proposed method on a multi-center survival dataset and demonstrate its effectiveness via quantitative and qualitative assessments, achieving superior C-indexes (0.764 and 0.727) and time-dependent AUCs (0.799 and 0.691). These results suggest that our explainable deep survival prediction model surpasses traditional survival analysis methods in risk prediction, improving interpretability for clinical decision making and enhancing AI system trustworthiness.
Authors: Zhennan Chen, Xuying Zhang, Tian-Zhu Xiang, Ying Tai
Abstract: Camouflaged object detection (COD) aims to segment objects visually embedded in their surroundings, which is a very challenging task due to the high similarity between the objects and the background. To address it, most methods often incorporate additional information (e.g., boundary, texture, and frequency clues) to guide feature learning for better detecting camouflaged objects from the background. Although progress has been made, these methods are basically individually tailored to specific auxiliary cues, thus lacking adaptability and not consistently achieving high segmentation performance. To this end, this paper proposes an adaptive guidance learning network, dubbed \textit{AGLNet}, which is a unified end-to-end learnable model for exploring and adapting different additional cues in CNN models to guide accurate camouflaged feature learning. Specifically, we first design a straightforward additional information generation (AIG) module to learn additional camouflaged object cues, which can be adapted for the exploration of effective camouflaged features. Then we present a hierarchical feature combination (HFC) module to deeply integrate additional cues and image features to guide camouflaged feature learning in a multi-level fusion manner.Followed by a recalibration decoder (RD), different features are further aggregated and refined for accurate object prediction. Extensive experiments on three widely used COD benchmark datasets demonstrate that the proposed method achieves significant performance improvements under different additional cues, and outperforms the recent 20 state-of-the-art methods by a large margin. Our code will be made publicly available at: \textcolor{blue}{{https://github.com/ZNan-Chen/AGLNet}}.
Authors: Juntao Hu, Yuan Wu
Abstract: We propose a novel data augmentation method termed You Only Need hAlf (YONA), which simplifies the augmentation process. YONA bisects an image, substitutes one half with noise, and applies data augmentation techniques to the remaining half. This method reduces the redundant information in the original image, encourages neural networks to recognize objects from incomplete views, and significantly enhances neural networks' robustness. YONA is distinguished by its properties of parameter-free, straightforward application, enhancing various existing data augmentation strategies, and thereby bolstering neural networks' robustness without additional computational cost. To demonstrate YONA's efficacy, extensive experiments were carried out. These experiments confirm YONA's compatibility with diverse data augmentation methods and neural network architectures, yielding substantial improvements in CIFAR classification tasks, sometimes outperforming conventional image-level data augmentation methods. Furthermore, YONA markedly increases the resilience of neural networks to adversarial attacks. Additional experiments exploring YONA's variants conclusively show that masking half of an image optimizes performance. The code is available at https://github.com/HansMoe/YONA.
Authors: Tianxiang Cui, Huibing Wang, Jinjia Peng, Ruoxi Deng, Xianping Fu, Yang Wang
Abstract: Unsupervised person search aims to localize a particular target person from a gallery set of scene images without annotations, which is extremely challenging due to the unexpected variations of the unlabeled domains. However, most existing methods dedicate to developing multi-stage models to adapt domain variations while using clustering for iterative model training, which inevitably increases model complexity. To address this issue, we propose a Fast One-stage Unsupervised person Search (FOUS) which complementary integrates domain adaptaion with label adaptaion within an end-to-end manner without iterative clustering. To minimize the domain discrepancy, FOUS introduced an Attention-based Domain Alignment Module (ADAM) which can not only align various domains for both detection and ReID tasks but also construct an attention mechanism to reduce the adverse impacts of low-quality candidates resulting from unsupervised detection. Moreover, to avoid the redundant iterative clustering mode, FOUS adopts a prototype-guided labeling method which minimizes redundant correlation computations for partial samples and assigns noisy coarse label groups efficiently. The coarse label groups will be continuously refined via label-flexible training network with an adaptive selection strategy. With the adapted domains and labels, FOUS can achieve the state-of-the-art (SOTA) performance on two benchmark datasets, CUHK-SYSU and PRW. The code is available at https://github.com/whbdmu/FOUS.
Authors: Yimin Jiang, Huibing Wang, Jinjia Peng, Xianping Fu, Yang Wang
Abstract: Person search aims to localize specific a target person from a gallery set of images with various scenes. As the scene of moving pedestrian changes, the captured person image inevitably bring in lots of background noise and foreground noise on the person feature, which are completely unrelated to the person identity, leading to severe performance degeneration. To address this issue, we present a Scene-Adaptive Person Search (SEAS) model by introducing bilateral modulations to simultaneously eliminate scene noise and maintain a consistent person representation to adapt to various scenes. In SEAS, a Background Modulation Network (BMN) is designed to encode the feature extracted from the detected bounding box into a multi-granularity embedding, which reduces the input of background noise from multiple levels with norm-aware. Additionally, to mitigate the effect of foreground noise on the person feature, SEAS introduces a Foreground Modulation Network (FMN) to compute the clutter reduction offset for the person embedding based on the feature map of the scene image. By bilateral modulations on both background and foreground within an end-to-end manner, SEAS obtains consistent feature representations without scene noise. SEAS can achieve state-of-the-art (SOTA) performance on two benchmark datasets, CUHK-SYSU with 97.1\% mAP and PRW with 60.5\% mAP. The code is available at https://github.com/whbdmu/SEAS.
Authors: Xiaole Tang, Xin Hu, Xiang Gu, Jian Sun
Abstract: Deep learning-based image restoration methods have achieved promising performance. However, how to faithfully preserve the structure of the original image remains challenging. To address this challenge, we propose a novel Residual-Conditioned Optimal Transport (RCOT) approach, which models the image restoration as an optimal transport (OT) problem for both unpaired and paired settings, integrating the transport residual as a unique degradation-specific cue for both the transport cost and the transport map. Specifically, we first formalize a Fourier residual-guided OT objective by incorporating the degradation-specific information of the residual into the transport cost. Based on the dual form of the OT formulation, we design the transport map as a two-pass RCOT map that comprises a base model and a refinement process, in which the transport residual is computed by the base model in the first pass and then encoded as a degradation-specific embedding to condition the second-pass restoration. By duality, the RCOT problem is transformed into a minimax optimization problem, which can be solved by adversarially training neural networks. Extensive experiments on multiple restoration tasks show the effectiveness of our approach in terms of both distortion measures and perceptual quality. Particularly, RCOT restores images with more faithful structural details compared to state-of-the-art methods.
Authors: Ziyun Qian, Zeyu Xiao, Zhenyi Wu, Dingkang Yang, Mingcheng Li, Shunli Wang, Shuaibing Wang, Dongliang Kou, Lihua Zhang
Abstract: Motion style transfer is a significant research direction in multimedia applications. It enables the rapid switching of different styles of the same motion for virtual digital humans, thus vastly increasing the diversity and realism of movements. It is widely applied in multimedia scenarios such as movies, games, and the Metaverse. However, most of the current work in this field adopts the GAN, which may lead to instability and convergence issues, making the final generated motion sequence somewhat chaotic and unable to reflect a highly realistic and natural style. To address these problems, we consider style motion as a condition and propose the Style Motion Conditioned Diffusion (SMCD) framework for the first time, which can more comprehensively learn the style features of motion. Moreover, we apply Mamba model for the first time in the motion style transfer field, introducing the Motion Style Mamba (MSM) module to handle longer motion sequences. Thirdly, aiming at the SMCD framework, we propose Diffusion-based Content Consistency Loss and Content Consistency Loss to assist the overall framework's training. Finally, we conduct extensive experiments. The results reveal that our method surpasses state-of-the-art methods in both qualitative and quantitative comparisons, capable of generating more realistic motion sequences.
Authors: Honghua Chen, Chen Change Loy, Xingang Pan
Abstract: Despite the emergence of successful NeRF inpainting methods built upon explicit RGB and depth 2D inpainting supervisions, these methods are inherently constrained by the capabilities of their underlying 2D inpainters. This is due to two key reasons: (i) independently inpainting constituent images results in view-inconsistent imagery, and (ii) 2D inpainters struggle to ensure high-quality geometry completion and alignment with inpainted RGB images. To overcome these limitations, we propose a novel approach called MVIP-NeRF that harnesses the potential of diffusion priors for NeRF inpainting, addressing both appearance and geometry aspects. MVIP-NeRF performs joint inpainting across multiple views to reach a consistent solution, which is achieved via an iterative optimization process based on Score Distillation Sampling (SDS). Apart from recovering the rendered RGB images, we also extract normal maps as a geometric representation and define a normal SDS loss that motivates accurate geometry inpainting and alignment with the appearance. Additionally, we formulate a multi-view SDS score function to distill generative priors simultaneously from different view images, ensuring consistent visual completion when dealing with large view variations. Our experimental results show better appearance and geometry recovery than previous NeRF inpainting methods.
Authors: Baijun Ye, Caiyun Liu, Xiaoyu Ye, Yuantao Chen, Yuhai Wang, Zike Yan, Yongliang Shi, Hao Zhao, Guyue Zhou
Abstract: Due to the limited model capacity, leveraging distributed Neural Radiance Fields (NeRFs) for modeling extensive urban environments has become a necessity. However, current distributed NeRF registration approaches encounter aliasing artifacts, arising from discrepancies in rendering resolutions and suboptimal pose precision. These factors collectively deteriorate the fidelity of pose estimation within NeRF frameworks, resulting in occlusion artifacts during the NeRF blending stage. In this paper, we present a distributed NeRF system with tri-stage pose optimization. In the first stage, precise poses of images are achieved by bundle adjusting Mip-NeRF 360 with a coarse-to-fine strategy. In the second stage, we incorporate the inverting Mip-NeRF 360, coupled with the truncated dynamic low-pass filter, to enable the achievement of robust and precise poses, termed Frame2Model optimization. On top of this, we obtain a coarse transformation between NeRFs in different coordinate systems. In the third stage, we fine-tune the transformation between NeRFs by Model2Model pose optimization. After obtaining precise transformation parameters, we proceed to implement NeRF blending, showcasing superior performance metrics in both real-world and simulation scenarios. Codes and data will be publicly available at https://github.com/boilcy/Distributed-NeRF.
Authors: Chenhao Wu
Abstract: This work presents a drone detector with modified backbone and multiple pyramid feature maps enhancement structure (MDDPE). Novel feature maps improve modules that uses different levels of information to produce more robust and discriminatory features is proposed. These module includes the feature maps supplement function and the feature maps recombination enhancement function.To effectively handle the drone characteristics, auxiliary supervisions that are implemented in the early stages by employing tailored anchors designed are utilized. To further improve the modeling of real drone detection scenarios and initialization of the regressor, an updated anchor matching technique is introduced to match anchors and ground truth drone as closely as feasible. To show the proposed MDDPE's superiority over the most advanced detectors, extensive experiments are carried out using well-known drone detection benchmarks.
Authors: Kassaw Abraham Mulat, Zhengyong Feng, Tegegne Solomon Eshetie, Ahmed Endris Hasen
Abstract: Salient object detection (SOD) remains an important task in computer vision, with applications ranging from image segmentation to autonomous driving. Fully convolutional network (FCN)-based methods have made remarkable progress in visual saliency detection over the last few decades. However, these methods have limitations in accurately detecting salient objects, particularly in challenging scenes with multiple objects, small objects, or objects with low resolutions. To address this issue, we proposed a Saliency Fusion Attention U-Net (SalFAU-Net) model, which incorporates a saliency fusion module into each decoder block of the attention U-net model to generate saliency probability maps from each decoder block. SalFAU-Net employs an attention mechanism to selectively focus on the most informative regions of an image and suppress non-salient regions. We train SalFAU-Net on the DUTS dataset using a binary cross-entropy loss function. We conducted experiments on six popular SOD evaluation datasets to evaluate the effectiveness of the proposed method. The experimental results demonstrate that our method, SalFAU-Net, achieves competitive performance compared to other methods in terms of mean absolute error (MAE), F-measure, s-measure, and e-measure.
Authors: Zhenyu Lou, Qiongjie Cui, Haofan Wang, Xu Tang, Hong Zhou
Abstract: Predicting future human pose is a fundamental application for machine intelligence, which drives robots to plan their behavior and paths ahead of time to seamlessly accomplish human-robot collaboration in real-world 3D scenarios. Despite encouraging results, existing approaches rarely consider the effects of the external scene on the motion sequence, leading to pronounced artifacts and physical implausibilities in the predictions. To address this limitation, this work introduces a novel multi-modal sense-informed motion prediction approach, which conditions high-fidelity generation on two modal information: external 3D scene, and internal human gaze, and is able to recognize their salience for future human activity. Furthermore, the gaze information is regarded as the human intention, and combined with both motion and scene features, we construct a ternary intention-aware attention to supervise the generation to match where the human wants to reach. Meanwhile, we introduce semantic coherence-aware attention to explicitly distinguish the salient point clouds and the underlying ones, to ensure a reasonable interaction of the generated sequence with the 3D scene. On two real-world benchmarks, the proposed method achieves state-of-the-art performance both in 3D human pose and trajectory prediction.
Authors: Nikita Shvetsov, Anders Sildnes, Lill-Tove Rasmussen Busund, Stig Dalen, Kajsa M{\o}llersen, Lars Ailo Bongo, Thomas K. Kilvaer
Abstract: Addressing the critical need for accurate prognostic biomarkers in cancer treatment, quantifying tumor-infiltrating lymphocytes (TILs) in non-small cell lung cancer (NSCLC) presents considerable challenges. Manual TIL quantification in whole slide images (WSIs) is laborious and subject to variability, potentially undermining patient outcomes. Our study introduces an automated pipeline that utilizes semi-stochastic patch sampling, patch classification to retain prognostically relevant patches, and cell quantification using the HoVer-Net model to streamline the TIL evaluation process. This pipeline efficiently excludes approximately 70% of areas not relevant for prognosis and requires only 5% of the remaining patches to maintain prognostic accuracy (c-index 0.65 +- 0.01). The computational efficiency achieved does not sacrifice prognostic accuracy, as demonstrated by the TILs score's strong correlation with patient survival, which surpasses traditional CD8 IHC scoring methods. While the pipeline demonstrates potential for enhancing NSCLC prognostication and personalization of treatment, comprehensive clinical validation is still required. Future research should focus on verifying its broader clinical utility and investigating additional biomarkers to improve NSCLC prognosis.
Authors: Tobias Groot, Matias Valdenegro-Toro
Abstract: Language and Vision-Language Models (LLMs/VLMs) have revolutionized the field of AI by their ability to generate human-like text and understand images, but ensuring their reliability is crucial. This paper aims to evaluate the ability of LLMs (GPT4, GPT-3.5, LLaMA2, and PaLM 2) and VLMs (GPT4V and Gemini Pro Vision) to estimate their verbalized uncertainty via prompting. We propose the new Japanese Uncertain Scenes (JUS) dataset, aimed at testing VLM capabilities via difficult queries and object counting, and the Net Calibration Error (NCE) to measure direction of miscalibration. Results show that both LLMs and VLMs have a high calibration error and are overconfident most of the time, indicating a poor capability for uncertainty estimation. Additionally we develop prompts for regression tasks, and we show that VLMs have poor calibration when producing mean/standard deviation and 95% confidence intervals.
Authors: Yuanye Liu, Zheyao Gao, Nannan Shi, Fuping Wu, Yuxin Shi, Qingchao Chen, Xiahai Zhuang
Abstract: Accurate staging of liver fibrosis from magnetic resonance imaging (MRI) is crucial in clinical practice. While conventional methods often focus on a specific sub-region, multi-view learning captures more information by analyzing multiple patches simultaneously. However, previous multi-view approaches could not typically calculate uncertainty by nature, and they generally integrate features from different views in a black-box fashion, hence compromising reliability as well as interpretability of the resulting models. In this work, we propose a new multi-view method based on evidential learning, referred to as MERIT, which tackles the two challenges in a unified framework. MERIT enables uncertainty quantification of the predictions to enhance reliability, and employs a logic-based combination rule to improve interpretability. Specifically, MERIT models the prediction from each sub-view as an opinion with quantified uncertainty under the guidance of the subjective logic theory. Furthermore, a distribution-aware base rate is introduced to enhance performance, particularly in scenarios involving class distribution shifts. Finally, MERIT adopts a feature-specific combination rule to explicitly fuse multi-view predictions, thereby enhancing interpretability. Results have showcased the effectiveness of the proposed MERIT, highlighting the reliability and offering both ad-hoc and post-hoc interpretability. They also illustrate that MERIT can elucidate the significance of each view in the decision-making process for liver fibrosis staging.
Authors: Fares Abawi, Di Fu, Stefan Wermter
Abstract: Previous research on scanpath prediction has mainly focused on group models, disregarding the fact that the scanpaths and attentional behaviors of individuals are diverse. The disregard of these differences is especially detrimental to social human-robot interaction, whereby robots commonly emulate human gaze based on heuristics or predefined patterns. However, human gaze patterns are heterogeneous and varying behaviors can significantly affect the outcomes of such human-robot interactions. To fill this gap, we developed a deep learning-based social cue integration model for saliency prediction to instead predict scanpaths in videos. Our model learned scanpaths by recursively integrating fixation history and social cues through a gating mechanism and sequential attention. We evaluated our approach on gaze datasets of dynamic social scenes, observed under the free-viewing condition. The introduction of fixation history into our models makes it possible to train a single unified model rather than the resource-intensive approach of training individual models for each set of scanpaths. We observed that the late neural integration approach surpasses early fusion when training models on a large dataset, in comparison to a smaller dataset with a similar distribution. Results also indicate that a single unified model, trained on all the observers' scanpaths, performs on par or better than individually trained models. We hypothesize that this outcome is a result of the group saliency representations instilling universal attention in the model, while the supervisory signal guides it to learn personalized attentional behaviors, providing the unified model a benefit over individual models due to its implicit representation of universal attention.
Authors: Jinmin Li, Tao Dai, Jingyun Zhang, Kang Liu, Jun Wang, Shaoming Wang, Shu-Tao Xia, rizen guo
Abstract: Recently developed generative methods, including invertible rescaling network (IRN) based and generative adversarial network (GAN) based methods, have demonstrated exceptional performance in image rescaling. However, IRN-based methods tend to produce over-smoothed results, while GAN-based methods easily generate fake details, which thus hinders their real applications. To address this issue, we propose Boundary-aware Decoupled Flow Networks (BDFlow) to generate realistic and visually pleasing results. Unlike previous methods that model high-frequency information as standard Gaussian distribution directly, our BDFlow first decouples the high-frequency information into \textit{semantic high-frequency} that adheres to a Boundary distribution and \textit{non-semantic high-frequency} counterpart that adheres to a Gaussian distribution. Specifically, to capture semantic high-frequency parts accurately, we use Boundary-aware Mask (BAM) to constrain the model to produce rich textures, while non-semantic high-frequency part is randomly sampled from a Gaussian distribution.Comprehensive experiments demonstrate that our BDFlow significantly outperforms other state-of-the-art methods while maintaining lower complexity. Notably, our BDFlow improves the PSNR by $4.4$ dB and the SSIM by $0.1$ on average over GRAIN, utilizing only 74\% of the parameters and 20\% of the computation. The code will be available at https://github.com/THU-Kingmin/BAFlow.
Authors: Xiwen Chen, Wenhui Zhu, Peijie Qiu, Abolfazl Razi
Abstract: Inverse imaging problems (IIPs) arise in various applications, with the main objective of reconstructing an image from its compressed measurements. This problem is often ill-posed for being under-determined with multiple interchangeably consistent solutions. The best solution inherently depends on prior knowledge or assumptions, such as the sparsity of the image. Furthermore, the reconstruction process for most IIPs relies significantly on the imaging (i.e. forward model) parameters, which might not be fully known, or the measurement device may undergo calibration drifts. These uncertainties in the forward model create substantial challenges, where inaccurate reconstructions usually happen when the postulated parameters of the forward model do not fully match the actual ones. In this work, we devoted to tackling accurate reconstruction under the context of a set of possible forward model parameters that exist. Here, we propose a novel Moment-Aggregation (MA) framework that is compatible with the popular IIP solution by using a neural network prior. Specifically, our method can reconstruct the signal by considering all candidate parameters of the forward model simultaneously during the update of the neural network. We theoretically demonstrate the convergence of the MA framework, which has a similar complexity with reconstruction under the known forward model parameters. Proof-of-concept experiments demonstrate that the proposed MA achieves performance comparable to the forward model with the known precise parameter in reconstruction across both compressive sensing and phase retrieval applications, with a PSNR gap of 0.17 to 1.94 over various datasets, including MNIST, X-ray, Glas, and MoNuseg. This highlights our method's significant potential in reconstruction under an uncertain forward model.
Authors: Jinmin Li, Tao Dai, Yaohua Zha, Yilu Luo, Longfei Lu, Bin Chen, Zhi Wang, Shu-Tao Xia, Jingyun Zhang
Abstract: Invertible Rescaling Networks (IRNs) and their variants have witnessed remarkable achievements in various image processing tasks like image rescaling. However, we observe that IRNs with deeper networks are difficult to train, thus hindering the representational ability of IRNs. To address this issue, we propose Invertible Residual Rescaling Models (IRRM) for image rescaling by learning a bijection between a high-resolution image and its low-resolution counterpart with a specific distribution. Specifically, we propose IRRM to build a deep network, which contains several Residual Downscaling Modules (RDMs) with long skip connections. Each RDM consists of several Invertible Residual Blocks (IRBs) with short connections. In this way, RDM allows rich low-frequency information to be bypassed by skip connections and forces models to focus on extracting high-frequency information from the image. Extensive experiments show that our IRRM performs significantly better than other state-of-the-art methods with much fewer parameters and complexity. Particularly, our IRRM has respectively PSNR gains of at least 0.3 dB over HCFlow and IRN in the $\times 4$ rescaling while only using 60\% parameters and 50\% FLOPs. The code will be available at https://github.com/THU-Kingmin/IRRM.
Authors: Lorenzo Agnolucci, Alberto Baldrati, Marco Bertini, Alberto Del Bimbo
Abstract: Given a query consisting of a reference image and a relative caption, Composed Image Retrieval (CIR) aims to retrieve target images visually similar to the reference one while incorporating the changes specified in the relative caption. The reliance of supervised methods on labor-intensive manually labeled datasets hinders their broad applicability. In this work, we introduce a new task, Zero-Shot CIR (ZS-CIR), that addresses CIR without the need for a labeled training dataset. We propose an approach named iSEARLE (improved zero-Shot composEd imAge Retrieval with textuaL invErsion) that involves mapping the visual information of the reference image into a pseudo-word token in CLIP token embedding space and combining it with the relative caption. To foster research on ZS-CIR, we present an open-domain benchmarking dataset named CIRCO (Composed Image Retrieval on Common Objects in context), the first CIR dataset where each query is labeled with multiple ground truths and a semantic categorization. The experimental results illustrate that iSEARLE obtains state-of-the-art performance on three different CIR datasets -- FashionIQ, CIRR, and the proposed CIRCO -- and two additional evaluation settings, namely domain conversion and object composition. The dataset, the code, and the model are publicly available at https://github.com/miccunifi/SEARLE.
Authors: Wenyu Zhang, Li Shen, Chuan-Sheng Foo
Abstract: Source-free domain adaptation (SFDA) aims to adapt a source model trained on a fully-labeled source domain to a related but unlabeled target domain. While the source model is a key avenue for acquiring target pseudolabels, the generated pseudolabels may exhibit source bias. In the conventional SFDA pipeline, a large data (e.g. ImageNet) pre-trained feature extractor is used to initialize the source model at the start of source training, and subsequently discarded. Despite having diverse features important for generalization, the pre-trained feature extractor can overfit to the source data distribution during source training and forget relevant target domain knowledge. Rather than discarding this valuable knowledge, we introduce an integrated framework to incorporate pre-trained networks into the target adaptation process. The proposed framework is flexible and allows us to plug modern pre-trained networks into the adaptation process to leverage their stronger representation learning capabilities. For adaptation, we propose the Co-learn algorithm to improve target pseudolabel quality collaboratively through the source model and a pre-trained feature extractor. Building on the recent success of the vision-language model CLIP in zero-shot image recognition, we present an extension Co-learn++ to further incorporate CLIP's zero-shot classification decisions. We evaluate on 3 benchmark datasets and include more challenging scenarios such as open-set, partial-set and open-partial SFDA. Experimental results demonstrate that our proposed strategy improves adaptation performance and can be successfully integrated with existing SFDA methods.
Authors: Xiaoyu Qiao, Weisheng Li, Yuping Huang, Lijian Yang
Abstract: Score matching with Langevin dynamics (SMLD) method has been successfully applied to accelerated MRI. However, the hyperparameters in the sampling process require subtle tuning, otherwise the results can be severely corrupted by hallucination artifacts, particularly with out-of-distribution test data. In this study, we propose a novel workflow in which SMLD results are regarded as additional priors to guide model-driven network training. First, we adopted a pretrained score network to obtain samples as preliminary guidance images (PGI) without the need for network retraining, parameter tuning and in-distribution test data. Although PGIs are corrupted by hallucination artifacts, we believe that they can provide extra information through effective denoising steps to facilitate reconstruction. Therefore, we designed a denoising module (DM) in the second step to improve the quality of PGIs. The features are extracted from the components of Langevin dynamics and the same score network with fine-tuning; hence, we can directly learn the artifact patterns. Third, we designed a model-driven network whose training is guided by denoised PGIs (DGIs). DGIs are densely connected with intermediate reconstructions in each cascade to enrich the features and are periodically updated to provide more accurate guidance. Our experiments on different sequences revealed that despite the low average quality of PGIs, the proposed workflow can effectively extract valuable information to guide the network training, even with severely reduced training data and sampling steps. Our method outperforms other cutting-edge techniques by effectively mitigating hallucination artifacts, yielding robust and high-quality reconstruction results.
Authors: Pietro Nardelli, Danilo Comminiello
Abstract: Due to the ever-increasing availability of video surveillance cameras and the growing need for crime prevention, the violence detection task is attracting greater attention from the research community. With respect to other action recognition tasks, violence detection in surveillance videos shows additional issues, such as the presence of a significant variety of real fight scenes. Unfortunately, available datasets seem to be very small compared with other action recognition datasets. Moreover, in surveillance applications, people in the scenes always differ for each video and the background of the footage differs for each camera. Also, violent actions in real-life surveillance videos must be detected quickly to prevent unwanted consequences, thus models would definitely benefit from a reduction in memory usage and computational costs. Such problems make classical action recognition methods difficult to be adopted. To tackle all these issues, we introduce JOSENet, a novel self-supervised framework that provides outstanding performance for violence detection in surveillance videos. The proposed model receives two spatiotemporal video streams, i.e., RGB frames and optical flows, and involves a new regularized self-supervised learning approach for videos. JOSENet provides improved performance compared to self-supervised state-of-the-art methods, while requiring one-fourth of the number of frames per video segment and a reduced frame rate. The source code and the instructions to reproduce our experiments are available at https://github.com/ispamm/JOSENet.
Authors: Juncheng Hu, Ximing Xing, Zhengqi Zhang, Jing Zhang, Qian Yu
Abstract: We propose a novel method, VectorPainter, for the task of stylized vector graphics synthesis. Given a text prompt and a reference style image, VectorPainter generates a vector graphic that aligns in content with the text prompt and remains faithful in style to the reference image. We recognize that the key to this task lies in fully leveraging the intrinsic properties of vector graphics. Innovatively, we conceptualize the stylization process as the rearrangement of vectorized strokes extracted from the reference image. VectorPainter employs an optimization-based pipeline. It begins by extracting vectorized strokes from the reference image, which are then used to initialize the synthesis process. To ensure fidelity to the reference style, a novel style preservation loss is introduced. Extensive experiments have been conducted to demonstrate that our method is capable of aligning with the text description while remaining faithful to the reference image.
Authors: Ali Emre Keskin, Hacer Yalim Keles
Abstract: Numerous sign language datasets exist, yet they typically cover only a limited selection of the thousands of signs used globally. Moreover, creating diverse sign language datasets is an expensive and challenging task due to the costs associated with gathering a varied group of signers. Motivated by these challenges, we aimed to develop a solution that addresses these limitations. In this context, we focused on textually describing body movements from skeleton keypoint sequences, leading to the creation of a new dataset. We structured this dataset around AUTSL, a comprehensive isolated Turkish sign language dataset. We also developed a baseline model, SkelCap, which can generate textual descriptions of body movements. This model processes the skeleton keypoints data as a vector, applies a fully connected layer for embedding, and utilizes a transformer neural network for sequence-to-sequence modeling. We conducted extensive evaluations of our model, including signer-agnostic and sign-agnostic assessments. The model achieved promising results, with a ROUGE-L score of 0.98 and a BLEU-4 score of 0.94 in the signer-agnostic evaluation. The dataset we have prepared, namely the AUTSL-SkelCap, will be made publicly available soon.
Authors: Xin Jin, Qianqian Qiao, Yi Lu, Shan Gao, Heng Huang, Guangdong Li
Abstract: Image aesthetic evaluation is a highly prominent research domain in the field of computer vision. In recent years, there has been a proliferation of datasets and corresponding evaluation methodologies for assessing the aesthetic quality of photographic works, leading to the establishment of a relatively mature research environment. However, in contrast to the extensive research in photographic aesthetics, the field of aesthetic evaluation for paintings and Drawings has seen limited attention until the introduction of the BAID dataset in March 2023. This dataset solely comprises overall scores for high-quality artistic images. Our research marks the pioneering introduction of a multi-attribute, multi-category dataset specifically tailored to the field of painting: Aesthetics of Paintings and Drawings Dataset (APDD). The construction of APDD received active participation from 28 professional artists worldwide, along with dozens of students specializing in the field of art. This dataset encompasses 24 distinct artistic categories and 10 different aesthetic attributes. Each image in APDD has been evaluated by six professionally trained experts in the field of art, including assessments for both total aesthetic scores and aesthetic attribute scores. The final APDD dataset comprises a total of 4985 images, with an annotation count exceeding 31100 entries. Concurrently, we propose an innovative approach: Art Assessment Network for Specific Painting Styles (AANSPS), designed for the assessment of aesthetic attributes in mixed-attribute art datasets. Through this research, our goal is to catalyze advancements in the field of aesthetic evaluation for paintings and drawings, while enriching the available resources and methodologies for its further development and application.
Authors: Viet-Thanh Nguyen, Van-Truong Pham, Thi-Thao Tran
Abstract: Skin lesion segmentation is a critical task in computer-aided diagnosis systems for dermatological diseases. Accurate segmentation of skin lesions from medical images is essential for early detection, diagnosis, and treatment planning. In this paper, we propose a new model for skin lesion segmentation namely AC-MambaSeg, an enhanced model that has the hybrid CNN-Mamba backbone, and integrates advanced components such as Convolutional Block Attention Module (CBAM), Attention Gate, and Selective Kernel Bottleneck. AC-MambaSeg leverages the Vision Mamba framework for efficient feature extraction, while CBAM and Selective Kernel Bottleneck enhance its ability to focus on informative regions and suppress background noise. We evaluate the performance of AC-MambaSeg on diverse datasets of skin lesion images including ISIC-2018 and PH2; then compare it against existing segmentation methods. Our model shows promising potential for improving computer-aided diagnosis systems and facilitating early detection and treatment of dermatological diseases. Our source code will be made available at: https://github.com/vietthanh2710/AC-MambaSeg.
Authors: Yu Gao, Jiancheng Huang, Xiaopeng Sun, Zequn Jie, Yujie Zhong, Lin Ma
Abstract: In this paper, we introduce Matten, a cutting-edge latent diffusion model with Mamba-Attention architecture for video generation. With minimal computational cost, Matten employs spatial-temporal attention for local video content modeling and bidirectional Mamba for global video content modeling. Our comprehensive experimental evaluation demonstrates that Matten has competitive performance with the current Transformer-based and GAN-based models in benchmark performance, achieving superior FVD scores and efficiency. Additionally, we observe a direct positive correlation between the complexity of our designed model and the improvement in video quality, indicating the excellent scalability of Matten.
Authors: Dong Chen, Arman Hosseini, Arik Smith, Amir Farzin Nikkhah, Arsalan Heydarian, Omid Shoghli, Bradford Campbell
Abstract: Electric scooters (e-scooters) have rapidly emerged as a popular mode of transportation in urban areas, yet they pose significant safety challenges. In the United States, the rise of e-scooters has been marked by a concerning increase in related injuries and fatalities. Recently, while deep-learning object detection holds paramount significance in autonomous vehicles to avoid potential collisions, its application in the context of e-scooters remains relatively unexplored. This paper addresses this gap by assessing the effectiveness and efficiency of cutting-edge object detectors designed for e-scooters. To achieve this, the first comprehensive benchmark involving 22 state-of-the-art YOLO object detectors, including five versions (YOLOv3, YOLOv5, YOLOv6, YOLOv7, and YOLOv8), has been established for real-time traffic object detection using a self-collected dataset featuring e-scooters. The detection accuracy, measured in terms of mAP@0.5, ranges from 27.4% (YOLOv7-E6E) to 86.8% (YOLOv5s). All YOLO models, particularly YOLOv3-tiny, have displayed promising potential for real-time object detection in the context of e-scooters. Both the traffic scene dataset (https://zenodo.org/records/10578641) and software program codes (https://github.com/DongChen06/ScooterDet) for model benchmarking in this study are publicly available, which will not only improve e-scooter safety with advanced object detection but also lay the groundwork for tailored solutions, promising a safer and more sustainable urban micromobility landscape.
URLs: https://zenodo.org/records/10578641), https://github.com/DongChen06/ScooterDet)
Authors: Zaedul Islam, A. Ben Hamza
Abstract: Accurate 3D human pose estimation is a challenging task due to occlusion and depth ambiguity. In this paper, we introduce a multi-hop graph transformer network designed for 2D-to-3D human pose estimation in videos by leveraging the strengths of multi-head self-attention and multi-hop graph convolutional networks with disentangled neighborhoods to capture spatio-temporal dependencies and handle long-range interactions. The proposed network architecture consists of a graph attention block composed of stacked layers of multi-head self-attention and graph convolution with learnable adjacency matrix, and a multi-hop graph convolutional block comprised of multi-hop convolutional and dilated convolutional layers. The combination of multi-head self-attention and multi-hop graph convolutional layers enables the model to capture both local and global dependencies, while the integration of dilated convolutional layers enhances the model's ability to handle spatial details required for accurate localization of the human body joints. Extensive experiments demonstrate the effectiveness and generalization ability of our model, achieving competitive performance on benchmark datasets.
Authors: Jinyin Wang, Xingchen Li, Yixuan Jin, Yihao Zhong, Keke Zhang, Chang Zhou
Abstract: This project investigates the human multi-modal behavior identification algorithm utilizing deep neural networks. According to the characteristics of different modal information, different deep neural networks are used to adapt to different modal video information. Through the integration of various deep neural networks, the algorithm successfully identifies behaviors across multiple modalities. In this project, multiple cameras developed by Microsoft Kinect were used to collect corresponding bone point data based on acquiring conventional images. In this way, the motion features in the image can be extracted. Ultimately, the behavioral characteristics discerned through both approaches are synthesized to facilitate the precise identification and categorization of behaviors. The performance of the suggested algorithm was evaluated using the MSR3D data set. The findings from these experiments indicate that the accuracy in recognizing behaviors remains consistently high, suggesting that the algorithm is reliable in various scenarios. Additionally, the tests demonstrate that the algorithm substantially enhances the accuracy of detecting pedestrian behaviors in video footage.
Authors: Adarsh Tiwari, Sanket Biswas, Josep Llad\'os
Abstract: We present SketchGPT, a flexible framework that employs a sequence-to-sequence autoregressive model for sketch generation, and completion, and an interpretation case study for sketch recognition. By mapping complex sketches into simplified sequences of abstract primitives, our approach significantly streamlines the input for autoregressive modeling. SketchGPT leverages the next token prediction objective strategy to understand sketch patterns, facilitating the creation and completion of drawings and also categorizing them accurately. This proposed sketch representation strategy aids in overcoming existing challenges of autoregressive modeling for continuous stroke data, enabling smoother model training and competitive performance. Our findings exhibit SketchGPT's capability to generate a diverse variety of drawings by adding both qualitative and quantitative comparisons with existing state-of-the-art, along with a comprehensive human evaluation study. The code and pretrained models will be released on our official GitHub.
Authors: Nil Biescas, Carlos Boned, Josep Llad\'os, Sanket Biswas
Abstract: This paper presents GeoContrastNet, a language-agnostic framework to structured document understanding (DU) by integrating a contrastive learning objective with graph attention networks (GATs), emphasizing the significant role of geometric features. We propose a novel methodology that combines geometric edge features with visual features within an overall two-staged GAT-based framework, demonstrating promising results in both link prediction and semantic entity recognition performance. Our findings reveal that combining both geometric and visual features could match the capabilities of large DU models that rely heavily on Optical Character Recognition (OCR) features in terms of performance accuracy and efficiency. This approach underscores the critical importance of relational layout information between the named text entities in a semi-structured layout of a page. Specifically, our results highlight the model's proficiency in identifying key-value relationships within the FUNSD dataset for forms and also discovering the spatial relationships in table-structured layouts for RVLCDIP business invoices. Our code and pretrained models will be accessible on our official GitHub.
Authors: Weihao Jiang, Chang Liu, Kun He
Abstract: Humans possess remarkable ability to accurately classify new, unseen images after being exposed to only a few examples. Such ability stems from their capacity to identify common features shared between new and previously seen images while disregarding distractions such as background variations. However, for artificial neural network models, determining the most relevant features for distinguishing between two images with limited samples presents a challenge. In this paper, we propose an intra-task mutual attention method for few-shot learning, that involves splitting the support and query samples into patches and encoding them using the pre-trained Vision Transformer (ViT) architecture. Specifically, we swap the class (CLS) token and patch tokens between the support and query sets to have the mutual attention, which enables each set to focus on the most useful information. This facilitates the strengthening of intra-class representations and promotes closer proximity between instances of the same class. For implementation, we adopt the ViT-based network architecture and utilize pre-trained model parameters obtained through self-supervision. By leveraging Masked Image Modeling as a self-supervised training task for pre-training, the pre-trained model yields semantically meaningful representations while successfully avoiding supervision collapse. We then employ a meta-learning method to fine-tune the last several layers and CLS token modules. Our strategy significantly reduces the num- ber of parameters that require fine-tuning while effectively uti- lizing the capability of pre-trained model. Extensive experiments show that our framework is simple, effective and computationally efficient, achieving superior performance as compared to the state-of-the-art baselines on five popular few-shot classification benchmarks under the 5-shot and 1-shot scenarios
Authors: Tao Liu, Feilong Chen, Shuai Fan, Chenpeng Du, Qi Chen, Xie Chen, Kai Yu
Abstract: The paper introduces AniTalker, an innovative framework designed to generate lifelike talking faces from a single portrait. Unlike existing models that primarily focus on verbal cues such as lip synchronization and fail to capture the complex dynamics of facial expressions and nonverbal cues, AniTalker employs a universal motion representation. This innovative representation effectively captures a wide range of facial dynamics, including subtle expressions and head movements. AniTalker enhances motion depiction through two self-supervised learning strategies: the first involves reconstructing target video frames from source frames within the same identity to learn subtle motion representations, and the second develops an identity encoder using metric learning while actively minimizing mutual information between the identity and motion encoders. This approach ensures that the motion representation is dynamic and devoid of identity-specific details, significantly reducing the need for labeled data. Additionally, the integration of a diffusion model with a variance adapter allows for the generation of diverse and controllable facial animations. This method not only demonstrates AniTalker's capability to create detailed and realistic facial movements but also underscores its potential in crafting dynamic avatars for real-world applications. Synthetic results can be viewed at https://github.com/X-LANCE/AniTalker.
Authors: Chengtao Lv, Hong Chen, Jinyang Guo, Yifu Ding, Xianglong Liu
Abstract: Segment Anything Model (SAM) has achieved impressive performance in many computer vision tasks. However, as a large-scale model, the immense memory and computation costs hinder its practical deployment. In this paper, we propose a post-training quantization (PTQ) framework for Segment Anything Model, namely PTQ4SAM. First, we investigate the inherent bottleneck of SAM quantization attributed to the bimodal distribution in post-Key-Linear activations. We analyze its characteristics from both per-tensor and per-channel perspectives, and propose a Bimodal Integration strategy, which utilizes a mathematically equivalent sign operation to transform the bimodal distribution into a relatively easy-quantized normal distribution offline. Second, SAM encompasses diverse attention mechanisms (i.e., self-attention and two-way cross-attention), resulting in substantial variations in the post-Softmax distributions. Therefore, we introduce an Adaptive Granularity Quantization for Softmax through searching the optimal power-of-two base, which is hardware-friendly. Extensive experimental results across various vision tasks (instance segmentation, semantic segmentation and object detection), datasets and model variants show the superiority of PTQ4SAM. For example, when quantizing SAM-L to 6-bit, we achieve lossless accuracy for instance segmentation, about 0.5\% drop with theoretical 3.9$\times$ acceleration. The code is available at \url{https://github.com/chengtao-lv/PTQ4SAM}.
Authors: Andrew Melnik, Michal Ljubljanac, Cong Lu, Qi Yan, Weiming Ren, Helge Ritter
Abstract: Diffusion generative models have recently become a robust technique for producing and modifying coherent, high-quality video. This survey offers a systematic overview of critical elements of diffusion models for video generation, covering applications, architectural choices, and the modeling of temporal dynamics. Recent advancements in the field are summarized and grouped into development trends. The survey concludes with an overview of remaining challenges and an outlook on the future of the field. Website: https://github.com/ndrwmlnk/Awesome-Video-Diffusion-Models
URLs: https://github.com/ndrwmlnk/Awesome-Video-Diffusion-Models
Authors: Wenxin Fan, Jian Cheng, Cheng Li, Xinrui Ma, Jing Yang, Juan Zou, Ruoyou Wu, Zan Chen, Yuanjing Feng, Hairong Zheng, Shanshan Wang
Abstract: Deep learning has emerged as a promising approach for learning the nonlinear mapping between diffusion-weighted MR images and tissue parameters, which enables automatic and deep understanding of the brain microstructures. However, the efficiency and accuracy in the multi-parametric estimations are still limited since previous studies tend to estimate multi-parametric maps with dense sampling and isolated signal modeling. This paper proposes DeepMpMRI, a unified framework for fast and high-fidelity multi-parametric estimation from various diffusion models using sparsely sampled q-space data. DeepMpMRI is equipped with a newly designed tensor-decomposition-based regularizer to effectively capture fine details by exploiting the correlation across parameters. In addition, we introduce a Nesterov-based adaptive learning algorithm that optimizes the regularization parameter dynamically to enhance the performance. DeepMpMRI is an extendable framework capable of incorporating flexible network architecture. Experimental results demonstrate the superiority of our approach over 5 state-of-the-art methods in simultaneously estimating multi-parametric maps for various diffusion models with fine-grained details both quantitatively and qualitatively, achieving 4.5 - 22.5$\times$ acceleration compared to the dense sampling of a total of 270 diffusion gradients.
Authors: Lin Yang, Shawn Xu, Andrew Sellergren, Timo Kohlberger, Yuchen Zhou, Ira Ktena, Atilla Kiraly, Faruk Ahmed, Farhad Hormozdiari, Tiam Jaroensri, Eric Wang, Ellery Wulczyn, Fayaz Jamil, Theo Guidroz, Chuck Lau, Siyuan Qiao, Yun Liu, Akshay Goel, Kendall Park, Arnav Agharwal, Nick George, Yang Wang, Ryutaro Tanno, David G. T. Barrett, Wei-Hung Weng, S. Sara Mahdavi, Khaled Saab, Tao Tu, Sreenivasa Raju Kalidindi, Mozziyar Etemadi, Jorge Cuadros, Gregory Sorensen, Yossi Matias, Katherine Chou, Greg Corrado, Joelle Barral, Shravya Shetty, David Fleet, S. M. Ali Eslami, Daniel Tse, Shruthi Prabhakara, Cory McLean, Dave Steiner, Rory Pilgrim, Christopher Kelly, Shekoofeh Azizi, Daniel Golden
Abstract: Many clinical tasks require an understanding of specialized data, such as medical images and genomics, which is not typically found in general-purpose large multimodal models. Building upon Gemini's multimodal models, we develop several models within the new Med-Gemini family that inherit core capabilities of Gemini and are optimized for medical use via fine-tuning with 2D and 3D radiology, histopathology, ophthalmology, dermatology and genomic data. Med-Gemini-2D sets a new standard for AI-based chest X-ray (CXR) report generation based on expert evaluation, exceeding previous best results across two separate datasets by an absolute margin of 1% and 12%, where 57% and 96% of AI reports on normal cases, and 43% and 65% on abnormal cases, are evaluated as "equivalent or better" than the original radiologists' reports. We demonstrate the first ever large multimodal model-based report generation for 3D computed tomography (CT) volumes using Med-Gemini-3D, with 53% of AI reports considered clinically acceptable, although additional research is needed to meet expert radiologist reporting quality. Beyond report generation, Med-Gemini-2D surpasses the previous best performance in CXR visual question answering (VQA) and performs well in CXR classification and radiology VQA, exceeding SoTA or baselines on 17 of 20 tasks. In histopathology, ophthalmology, and dermatology image classification, Med-Gemini-2D surpasses baselines across 18 out of 20 tasks and approaches task-specific model performance. Beyond imaging, Med-Gemini-Polygenic outperforms the standard linear polygenic risk score-based approach for disease risk prediction and generalizes to genetically correlated diseases for which it has never been trained. Although further development and evaluation are necessary in the safety-critical medical domain, our results highlight the potential of Med-Gemini across a wide range of medical tasks.
Authors: Yunfeng Li, Bo Wang, Ye Li, Zhiwen Yu, Liang Wang
Abstract: Complementary RGB and TIR modalities enable RGB-T tracking to achieve competitive performance in challenging scenarios. Therefore, how to better fuse cross-modal features is the core issue of RGB-T tracking. Some previous methods either insufficiently fuse RGB and TIR features, or depend on intermediaries containing information from both modalities to achieve cross-modal information interaction. The former does not fully exploit the potential of using only RGB and TIR information of the template or search region for channel and spatial feature fusion, and the latter lacks direct interaction between the template and search area, which limits the model's ability to fully exploit the original semantic information of both modalities. To alleviate these limitations, we explore how to improve the performance of a visual Transformer by using direct fusion of cross-modal channels and spatial features, and propose CSTNet. CSTNet uses ViT as a backbone and inserts cross-modal channel feature fusion modules (CFM) and cross-modal spatial feature fusion modules (SFM) for direct interaction between RGB and TIR features. The CFM performs parallel joint channel enhancement and joint multilevel spatial feature modeling of RGB and TIR features and sums the features, and then globally integrates the sum feature with the original features. The SFM uses cross-attention to model the spatial relationship of cross-modal features and then introduces a convolutional feedforward network for joint spatial and channel integration of multimodal features. Comprehensive experiments show that CSTNet achieves state-of-the-art performance on three public RGB-T tracking benchmarks. Code is available at https://github.com/LiYunfengLYF/CSTNet.
Authors: Jiacheng Cheng, Hijung Valentina Shin, Nuno Vasconcelos, Bryan Russell, Fabian Caba Heilbron
Abstract: In the recent years, the dual-encoder vision-language models (\eg CLIP) have achieved remarkable text-to-image retrieval performance. However, we discover that these models usually results in very different retrievals for a pair of paraphrased queries. Such behavior might render the retrieval system less predictable and lead to user frustration. In this work, we consider the task of paraphrased text-to-image retrieval where a model aims to return similar results given a pair of paraphrased queries. To start with, we collect a dataset of paraphrased image descriptions to facilitate quantitative evaluation for this task. We then hypothesize that the undesired behavior of existing dual-encoder model is due to their text towers which are trained on image-sentence pairs and lack the ability to capture the semantic similarity between paraphrased queries. To improve on this, we investigate multiple strategies for training a dual-encoder model starting from a language model pretrained on a large text corpus. Compared to public dual-encoder models such as CLIP and OpenCLIP, the model trained with our best adaptation strategy achieves a significantly higher ranking similarity for paraphrased queries while maintaining similar zero-shot classification and retrieval accuracy.
Authors: Juanjuan Weng, Zhiming Luo, Shaozi Li
Abstract: Recent studies have shown that Deep Neural Networks (DNNs) are susceptible to adversarial attacks, with frequency-domain analysis underscoring the significance of high-frequency components in influencing model predictions. Conversely, targeting low-frequency components has been effective in enhancing attack transferability on black-box models. In this study, we introduce a frequency decomposition-based feature mixing method to exploit these frequency characteristics in both clean and adversarial samples. Our findings suggest that incorporating features of clean samples into adversarial features extracted from adversarial examples is more effective in attacking normally-trained models, while combining clean features with the adversarial features extracted from low-frequency parts decomposed from the adversarial samples yields better results in attacking defense models. However, a conflict issue arises when these two mixing approaches are employed simultaneously. To tackle the issue, we propose a cross-frequency meta-optimization approach comprising the meta-train step, meta-test step, and final update. In the meta-train step, we leverage the low-frequency components of adversarial samples to boost the transferability of attacks against defense models. Meanwhile, in the meta-test step, we utilize adversarial samples to stabilize gradients, thereby enhancing the attack's transferability against normally trained models. For the final update, we update the adversarial sample based on the gradients obtained from both meta-train and meta-test steps. Our proposed method is evaluated through extensive experiments on the ImageNet-Compatible dataset, affirming its effectiveness in improving the transferability of attacks on both normally-trained CNNs and defense models. The source code is available at https://github.com/WJJLL/MetaSSA.
Authors: Zhizhao Duan, Hao Cheng, Duo Xu, Xi Wu, Xiangxie Zhang, Xi Ye, Zhen Xie
Abstract: In the vast and dynamic landscape of urban settings, Traffic Safety Description and Analysis plays a pivotal role in applications ranging from insurance inspection to accident prevention. This paper introduces CityLLaVA, a novel fine-tuning framework for Visual Language Models (VLMs) designed for urban scenarios. CityLLaVA enhances model comprehension and prediction accuracy through (1) employing bounding boxes for optimal visual data preprocessing, including video best-view selection and visual prompt engineering during both training and testing phases; (2) constructing concise Question-Answer sequences and designing textual prompts to refine instruction comprehension; (3) implementing block expansion to fine-tune large VLMs efficiently; and (4) advancing prediction accuracy via a unique sequential questioning-based prediction augmentation. Demonstrating top-tier performance, our method achieved a benchmark score of 33.4308, securing the leading position on the leaderboard. The code can be found: https://github.com/alibaba/AICITY2024_Track2_AliOpenTrek_CityLLaVA
URLs: https://github.com/alibaba/AICITY2024_Track2_AliOpenTrek_CityLLaVA
Authors: Zhiwei Wang, Xiaoyu Zeng, Chongwei Wu, Jinxin lv, Xu Zhang, Wei Fang, Qiang Li
Abstract: One-shot segmentation of brain tissue requires training registration-segmentation (reg-seg) dual-model iteratively, where reg-model aims to provide pseudo masks of unlabeled images for seg-model by warping a carefully-labeled atlas. However, the imperfect reg-model induces image-mask misalignment, poisoning the seg-model subsequently. Recent StyleSeg bypasses this bottleneck by replacing the unlabeled images with their warped copies of atlas, but needs to borrow the diverse image patterns via style transformation. Here, we present StyleSeg V2, inherited from StyleSeg but granted the ability of perceiving the registration errors. The motivation is that good registration behaves in a mirrored fashion for mirrored images. Therefore, almost at no cost, StyleSeg V2 can have reg-model itself "speak out" incorrectly-aligned regions by simply mirroring (symmetrically flipping the brain) its input, and the registration errors are symmetric inconsistencies between the outputs of original and mirrored inputs. Consequently, StyleSeg V2 allows the seg-model to make use of correctly-aligned regions of unlabeled images and also enhances the fidelity of style-transformed warped atlas image by weighting the local transformation strength according to registration errors. The experimental results on three public datasets demonstrate that our proposed StyleSeg V2 outperforms other state-of-the-arts by considerable margins, and exceeds StyleSeg by increasing the average Dice by at least 2.4%.
Authors: Haihong Hao, Shuo Wang, Huixia Ben, Yanbin Hao, Yansong Wang, Weiwei Wang
Abstract: Micro-expression recognition (MER) aims to recognize the short and subtle facial movements from the Micro-expression (ME) video clips, which reveal real emotions. Recent MER methods mostly only utilize special frames from ME video clips or extract optical flow from these special frames. However, they neglect the relationship between movements and space-time, while facial cues are hidden within these relationships. To solve this issue, we propose the Hierarchical Space-Time Attention (HSTA). Specifically, we first process ME video frames and special frames or data parallelly by our cascaded Unimodal Space-Time Attention (USTA) to establish connections between subtle facial movements and specific facial areas. Then, we design Crossmodal Space-Time Attention (CSTA) to achieve a higher-quality fusion for crossmodal data. Finally, we hierarchically integrate USTA and CSTA to grasp the deeper facial cues. Our model emphasizes temporal modeling without neglecting the processing of special data, and it fuses the contents in different modalities while maintaining their respective uniqueness. Extensive experiments on the four benchmarks show the effectiveness of our proposed HSTA. Specifically, compared with the latest method on the CASME3 dataset, it achieves about 3% score improvement in seven-category classification.
Authors: Tianlang He, Zhiqiu Xia, S. -H. Gary Chan
Abstract: Classifying a pedestrian in one of the three conveyor states of "elevator," "escalator" and "neither" is fundamental to many applications such as indoor localization and people flow analysis. We estimate, for the first time, the pedestrian conveyor state given the inertial navigation system (INS) readings of accelerometer, gyroscope and magnetometer sampled from the phone. Our problem is challenging because the INS signals of the conveyor state are coupled and perturbed by unpredictable arbitrary human actions, confusing the decision process. We propose ELESON, a novel, effective and lightweight INS-based deep learning approach to classify whether a pedestrian is in an elevator, escalator or neither. ELESON utilizes a motion feature extractor to decouple the conveyor state from human action in the feature space, and a magnetic feature extractor to account for the speed difference between elevator and escalator. Given the results of the extractors, it employs an evidential state classifier to estimate the confidence of the pedestrian states. Based on extensive experiments conducted on twenty hours of real pedestrian data, we demonstrate that ELESON outperforms significantly the state-of-the-art approaches (where combined INS signals of both the conveyor state and human actions are processed together), with 15% classification improvement in F1 score, stronger confidence discriminability with 10% increase in AUROC (Area Under the Receiver Operating Characteristics), and low computational and memory requirements on smartphones.
Authors: Zeyu Huang, Honghao Xu, Haibin Huang, Chongyang Ma, Hui Huang, Ruizhen Hu
Abstract: In this paper, we introduce a new method for the task of interaction transfer. Given an example interaction between a source object and an agent, our method can automatically infer both surface and spatial relationships for the agent and target objects within the same category, yielding more accurate and valid transfers. Specifically, our method characterizes the example interaction using a combined spatial and surface representation. We correspond the agent points and object points related to the representation to the target object space using a learned spatial and surface correspondence field, which represents objects as deformed and rotated signed distance fields. With the corresponded points, an optimization is performed under the constraints of our spatial and surface interaction representation and additional regularization. Experiments conducted on human-chair and hand-mug interaction transfer tasks show that our approach can handle larger geometry and topology variations between source and target shapes, significantly outperforming state-of-the-art methods.
Authors: Xuran Zhu
Abstract: Brain disorders are a major challenge to global health, causing millions of deaths each year. Accurate diagnosis of these diseases relies heavily on advanced medical imaging techniques such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT). However, the scarcity of annotated data poses a significant challenge in deploying machine learning models for medical diagnosis. To address this limitation, deep learning techniques have shown considerable promise. Domain adaptation techniques enhance a model's ability to generalize across imaging modalities by transferring knowledge from one domain (e.g., CT images) to another (e.g., MRI images). Such cross-modality adaptation is essential to improve the ability of models to consistently generalize across different imaging modalities. This study collected relevant resources from the Kaggle website and employed the Maximum Mean Difference (MMD) method - a popular domain adaptation method - to reduce the differences between imaging domains. By combining MMD with Convolutional Neural Networks (CNNs), the accuracy and utility of the model is obviously enhanced. The excellent experimental results highlight the great potential of data-driven domain adaptation techniques to improve diagnostic accuracy and efficiency, especially in resource-limited environments. By bridging the gap between different imaging modalities, the study aims to provide clinicians with more reliable diagnostic tools.
Authors: Leonhard Hennicke, Christian Medeiros Adriano, Holger Giese, Jan Mathias Koehler, Lukas Schott
Abstract: Generative foundation models like Stable Diffusion comprise a diverse spectrum of knowledge in computer vision with the potential for transfer learning, e.g., via generating data to train student models for downstream tasks. This could circumvent the necessity of collecting labeled real-world data, thereby presenting a form of data-free knowledge distillation. However, the resultant student models show a significant drop in accuracy compared to models trained on real data. We investigate possible causes for this drop and focus on the role of the different layers of the student model. By training these layers using either real or synthetic data, we reveal that the drop mainly stems from the model's final layers. Further, we briefly investigate other factors, such as differences in data-normalization between synthetic and real, the impact of data augmentations, texture vs.\ shape learning, and assuming oracle prompts. While we find that some of those factors can have an impact, they are not sufficient to close the gap towards real data. Building upon our insights that mainly later layers are responsible for the drop, we investigate the data-efficiency of fine-tuning a synthetically trained model with real data applied to only those last layers. Our results suggest an improved trade-off between the amount of real training data used and the model's accuracy. Our findings contribute to the understanding of the gap between synthetic and real data and indicate solutions to mitigate the scarcity of labeled real data.
Authors: Yuanhan Zhang, Kaichen Zhang, Bo Li, Fanyi Pu, Christopher Arif Setiadharma, Jingkang Yang, Ziwei Liu
Abstract: Multimodal information, together with our knowledge, help us to understand the complex and dynamic world. Large language models (LLM) and large multimodal models (LMM), however, still struggle to emulate this capability. In this paper, we present WorldQA, a video understanding dataset designed to push the boundaries of multimodal world models with three appealing properties: (1) Multimodal Inputs: The dataset comprises 1007 question-answer pairs and 303 videos, necessitating the analysis of both auditory and visual data for successful interpretation. (2) World Knowledge: We identify five essential types of world knowledge for question formulation. This approach challenges models to extend their capabilities beyond mere perception. (3) Long-Chain Reasoning: Our dataset introduces an average reasoning step of 4.45, notably surpassing other videoQA datasets. Furthermore, we introduce WorldRetriever, an agent designed to synthesize expert knowledge into a coherent reasoning chain, thereby facilitating accurate responses to WorldQA queries. Extensive evaluations of 13 prominent LLMs and LMMs reveal that WorldRetriever, although being the most effective model, achieved only 70% of humanlevel performance in multiple-choice questions. This finding highlights the necessity for further advancement in the reasoning and comprehension abilities of models. Our experiments also yield several key insights. For instance, while humans tend to perform better with increased frames, current LMMs, including WorldRetriever, show diminished performance under similar conditions. We hope that WorldQA,our methodology, and these insights could contribute to the future development of multimodal world models.
Authors: Yizhuo Lu, Changde Du, Chong Wang, Xuanliu Zhu, Liuyun Jiang, Huiguang He
Abstract: Reconstructing human dynamic vision from brain activity is a challenging task with great scientific significance. The difficulty stems from two primary issues: (1) vision-processing mechanisms in the brain are highly intricate and not fully revealed, making it challenging to directly learn a mapping between fMRI and video; (2) the temporal resolution of fMRI is significantly lower than that of natural videos. To overcome these issues, this paper propose a two-stage model named Mind-Animator, which achieves state-of-the-art performance on three public datasets. Specifically, during the fMRI-to-feature stage, we decouple semantic, structural, and motion features from fMRI through fMRI-vision-language tri-modal contrastive learning and sparse causal attention. In the feature-to-video stage, these features are merged to videos by an inflated Stable Diffusion. We substantiate that the reconstructed video dynamics are indeed derived from fMRI, rather than hallucinations of the generative model, through permutation tests. Additionally, the visualization of voxel-wise and ROI-wise importance maps confirms the neurobiological interpretability of our model.
Authors: William Lindskog, Valentin Spannagl, Christian Prehofer
Abstract: Ensuring driver readiness poses challenges, yet driver monitoring systems can assist in determining the driver's state. By observing visual cues, such systems recognize various behaviors and associate them with specific conditions. For instance, yawning or eye blinking can indicate driver drowsiness. Consequently, an abundance of distributed data is generated for driver monitoring. Employing machine learning techniques, such as driver drowsiness detection, presents a potential solution. However, transmitting the data to a central machine for model training is impractical due to the large data size and privacy concerns. Conversely, training on a single vehicle would limit the available data and likely result in inferior performance. To address these issues, we propose a federated learning framework for drowsiness detection within a vehicular network, leveraging the YawDD dataset. Our approach achieves an accuracy of 99.2%, demonstrating its promise and comparability to conventional deep learning techniques. Lastly, we show how our model scales using various number of federated clients
Authors: Maximilian Weber, Daniel Wild, Jens Kleesiek, Jan Egger, Christina Gsaxner
Abstract: Point cloud registration aligns 3D point clouds using spatial transformations. It is an important task in computer vision, with applications in areas such as augmented reality (AR) and medical imaging. This work explores the intersection of two research trends: the integration of AR into image-guided surgery and the use of deep learning for point cloud registration. The main objective is to evaluate the feasibility of applying deep learning-based point cloud registration methods for image-to-patient registration in augmented reality-guided surgery. We created a dataset of point clouds from medical imaging and corresponding point clouds captured with a popular AR device, the HoloLens 2. We evaluate three well-established deep learning models in registering these data pairs. While we find that some deep learning methods show promise, we show that a conventional registration pipeline still outperforms them on our challenging dataset.
Authors: Yingying Zhang, Chuangji Shi, Xin Guo, Jiangwei Lao, Jian Wang, Jiaotuan Wang, Jingdong Chen
Abstract: The design of the query is crucial for the performance of DETR and its variants. Each query consists of two components: a content part and a positional one. Traditionally, the content query is initialized with a zero or learnable embedding, lacking essential content information and resulting in sub-optimal performance. In this paper, we introduce a novel plug-and-play module, Self-Adaptive Content Query (SACQ), to address this limitation. The SACQ module utilizes features from the transformer encoder to generate content queries via self-attention pooling. This allows candidate queries to adapt to the input image, resulting in a more comprehensive content prior and better focus on target objects. However, this improved concentration poses a challenge for the training process that utilizes the Hungarian matching, which selects only a single candidate and suppresses other similar ones. To overcome this, we propose a query aggregation strategy to cooperate with SACQ. It merges similar predicted candidates from different queries, easing the optimization. Our extensive experiments on the COCO dataset demonstrate the effectiveness of our proposed approaches across six different DETR's variants with multiple configurations, achieving an average improvement of over 1.0 AP.
Authors: Lemuel Puglisi, Daniel C. Alexander, Daniele Rav\`i
Abstract: In this work, we introduce Brain Latent Progression (BrLP), a novel spatiotemporal disease progression model based on latent diffusion. BrLP is designed to predict the evolution of diseases at the individual level on 3D brain MRIs. Existing deep generative models developed for this task are primarily data-driven and face challenges in learning disease progressions. BrLP addresses these challenges by incorporating prior knowledge from disease models to enhance the accuracy of predictions. To implement this, we propose to integrate an auxiliary model that infers volumetric changes in various brain regions. Additionally, we introduce Latent Average Stabilization (LAS), a novel technique to improve spatiotemporal consistency of the predicted progression. BrLP is trained and evaluated on a large dataset comprising 11,730 T1-weighted brain MRIs from 2,805 subjects, collected from three publicly available, longitudinal Alzheimer's Disease (AD) studies. In our experiments, we compare the MRI scans generated by BrLP with the actual follow-up MRIs available from the subjects, in both cross-sectional and longitudinal settings. BrLP demonstrates significant improvements over existing methods, with an increase of 22% in volumetric accuracy across AD-related brain regions and 43% in image similarity to the ground-truth scans. The ability of BrLP to generate conditioned 3D scans at the subject level, along with the novelty of integrating prior knowledge to enhance accuracy, represents a significant advancement in disease progression modeling, opening new avenues for precision medicine. The code of BrLP is available at the following link: https://github.com/LemuelPuglisi/BrLP.
Authors: Xunchu Zhou, Xiaohong Liu, Yunlong Dong, Tengchuan Kou, Yixuan Gao, Zicheng Zhang, Chunyi Li, Haoning Wu, Guangtao Zhai
Abstract: Recently, User-Generated Content (UGC) videos have gained popularity in our daily lives. However, UGC videos often suffer from poor exposure due to the limitations of photographic equipment and techniques. Therefore, Video Exposure Correction (VEC) algorithms have been proposed, Low-Light Video Enhancement (LLVE) and Over-Exposed Video Recovery (OEVR) included. Equally important to the VEC is the Video Quality Assessment (VQA). Unfortunately, almost all existing VQA models are built generally, measuring the quality of a video from a comprehensive perspective. As a result, Light-VQA, trained on LLVE-QA, is proposed for assessing LLVE. We extend the work of Light-VQA by expanding the LLVE-QA dataset into Video Exposure Correction Quality Assessment (VEC-QA) dataset with over-exposed videos and their corresponding corrected versions. In addition, we propose Light-VQA+, a VQA model specialized in assessing VEC. Light-VQA+ differs from Light-VQA mainly from the usage of the CLIP model and the vision-language guidance during the feature extraction, followed by a new module referring to the Human Visual System (HVS) for more accurate assessment. Extensive experimental results show that our model achieves the best performance against the current State-Of-The-Art (SOTA) VQA models on the VEC-QA dataset and other public datasets.
Authors: Jiesong Bai, Yuhao Yin, Qiyuan He
Abstract: In the field of low-light image enhancement, both traditional Retinex methods and advanced deep learning techniques such as Retinexformer have shown distinct advantages and limitations. Traditional Retinex methods, designed to mimic the human eye's perception of brightness and color, decompose images into illumination and reflection components but struggle with noise management and detail preservation under low light conditions. Retinexformer enhances illumination estimation through traditional self-attention mechanisms, but faces challenges with insufficient interpretability and suboptimal enhancement effects. To overcome these limitations, this paper introduces the RetinexMamba architecture. RetinexMamba not only captures the physical intuitiveness of traditional Retinex methods but also integrates the deep learning framework of Retinexformer, leveraging the computational efficiency of State Space Models (SSMs) to enhance processing speed. This architecture features innovative illumination estimators and damage restorer mechanisms that maintain image quality during enhancement. Moreover, RetinexMamba replaces the IG-MSA (Illumination-Guided Multi-Head Attention) in Retinexformer with a Fused-Attention mechanism, improving the model's interpretability. Experimental evaluations on the LOL dataset show that RetinexMamba outperforms existing deep learning approaches based on Retinex theory in both quantitative and qualitative metrics, confirming its effectiveness and superiority in enhancing low-light images.
Authors: Nianchang Huang, Yang Yang, Qiang Zhang, Jungong Han, Jin Huang
Abstract: This paper delves into the task of arbitrary modality salient object detection (AM SOD), aiming to detect salient objects from arbitrary modalities, eg RGB images, RGB-D images, and RGB-D-T images. A novel modality-adaptive Transformer (MAT) will be proposed to investigate two fundamental challenges of AM SOD, ie more diverse modality discrepancies caused by varying modality types that need to be processed, and dynamic fusion design caused by an uncertain number of modalities present in the inputs of multimodal fusion strategy. Specifically, inspired by prompt learning's ability of aligning the distributions of pre-trained models to the characteristic of downstream tasks by learning some prompts, MAT will first present a modality-adaptive feature extractor (MAFE) to tackle the diverse modality discrepancies by introducing a modality prompt for each modality. In the training stage, a new modality translation contractive (MTC) loss will be further designed to assist MAFE in learning those modality-distinguishable modality prompts. Accordingly, in the testing stage, MAFE can employ those learned modality prompts to adaptively adjust its feature space according to the characteristics of the input modalities, thus being able to extract discriminative unimodal features. Then, MAFE will present a channel-wise and spatial-wise fusion hybrid (CSFH) strategy to meet the demand for dynamic fusion. For that, CSFH dedicates a channel-wise dynamic fusion module (CDFM) and a novel spatial-wise dynamic fusion module (SDFM) to fuse the unimodal features from varying numbers of modalities and meanwhile effectively capture cross-modal complementary semantic and detail information, respectively. Moreover, CSFH will carefully align CDFM and SDFM to different levels of unimodal features based on their characteristics for more effective complementary information exploitation.
Authors: Nianchang Huang, Yang Yang, Ruida Xi, Qiang Zhang, Jungong Han, Jin Huang
Abstract: Toward desirable saliency prediction, the types and numbers of inputs for a salient object detection (SOD) algorithm may dynamically change in many real-life applications. However, existing SOD algorithms are mainly designed or trained for one particular type of inputs, failing to be generalized to other types of inputs. Consequentially, more types of SOD algorithms need to be prepared in advance for handling different types of inputs, raising huge hardware and research costs. Differently, in this paper, we propose a new type of SOD task, termed Arbitrary Modality SOD (AM SOD). The most prominent characteristics of AM SOD are that the modality types and modality numbers will be arbitrary or dynamically changed. The former means that the inputs to the AM SOD algorithm may be arbitrary modalities such as RGB, depths, or even any combination of them. While, the latter indicates that the inputs may have arbitrary modality numbers as the input type is changed, e.g. single-modality RGB image, dual-modality RGB-Depth (RGB-D) images or triple-modality RGB-Depth-Thermal (RGB-D-T) images. Accordingly, a preliminary solution to the above challenges, \i.e. a modality switch network (MSN), is proposed in this paper. In particular, a modality switch feature extractor (MSFE) is first designed to extract discriminative features from each modality effectively by introducing some modality indicators, which will generate some weights for modality switching. Subsequently, a dynamic fusion module (DFM) is proposed to adaptively fuse features from a variable number of modalities based on a novel Transformer structure. Finally, a new dataset, named AM-XD, is constructed to facilitate research on AM SOD. Extensive experiments demonstrate that our AM SOD method can effectively cope with changes in the type and number of input modalities for robust salient object detection.
Authors: Li Mi, Xianjie Dai, Javiera Castillo-Navarro, Devis Tuia
Abstract: Image-based retrieval in large Earth observation archives is challenging because one needs to navigate across thousands of candidate matches only with the query image as a guide. By using text as information supporting the visual query, the retrieval system gains in usability, but at the same time faces difficulties due to the diversity of visual signals that cannot be summarized by a short caption only. For this reason, as a matching-based task, cross-modal text-image retrieval often suffers from information asymmetry between texts and images. To address this challenge, we propose a Knowledge-aware Text-Image Retrieval (KTIR) method for remote sensing images. By mining relevant information from an external knowledge graph, KTIR enriches the text scope available in the search query and alleviates the information gaps between texts and images for better matching. Moreover, by integrating domain-specific knowledge, KTIR also enhances the adaptation of pre-trained vision-language models to remote sensing applications. Experimental results on three commonly used remote sensing text-image retrieval benchmarks show that the proposed knowledge-aware method leads to varied and consistent retrievals, outperforming state-of-the-art retrieval methods.
Authors: Virgile Foy (IMT), Fabrice Gamboa (IMT), Reda Chhaibi (IMT)
Abstract: In the field of computer vision, the numerical encoding of 3D surfaces is crucial. It is classical to represent surfaces with their Signed Distance Functions (SDFs) or Unsigned Distance Functions (UDFs). For tasks like representation learning, surface classification, or surface reconstruction, this function can be learned by a neural network, called Neural Distance Function. This network, and in particular its weights, may serve as a parametric and implicit representation for the surface. The network must represent the surface as accurately as possible. In this paper, we propose a method for learning UDFs that improves the fidelity of the obtained Neural UDF to the original 3D surface. The key idea of our method is to concentrate the learning effort of the Neural UDF on surface edges. More precisely, we show that sampling more training points around surface edges allows better local accuracy of the trained Neural UDF, and thus improves the global expressiveness of the Neural UDF in terms of Hausdorff distance. To detect surface edges, we propose a new statistical method based on the calculation of a $p$-value at each point on the surface. Our method is shown to detect surface edges more accurately than a commonly used local geometric descriptor.
Authors: Xingguang Zhong, Yue Pan, Cyrill Stachniss, Jens Behley
Abstract: Building accurate maps is a key building block to enable reliable localization, planning, and navigation of autonomous vehicles. We propose a novel approach for building accurate maps of dynamic environments utilizing a sequence of LiDAR scans. To this end, we propose encoding the 4D scene into a novel spatio-temporal implicit neural map representation by fitting a time-dependent truncated signed distance function to each point. Using our representation, we extract the static map by filtering the dynamic parts. Our neural representation is based on sparse feature grids, a globally shared decoder, and time-dependent basis functions, which we jointly optimize in an unsupervised fashion. To learn this representation from a sequence of LiDAR scans, we design a simple yet efficient loss function to supervise the map optimization in a piecewise way. We evaluate our approach on various scenes containing moving objects in terms of the reconstruction quality of static maps and the segmentation of dynamic point clouds. The experimental results demonstrate that our method is capable of removing the dynamic part of the input point clouds while reconstructing accurate and complete 3D maps, outperforming several state-of-the-art methods. Codes are available at: https://github.com/PRBonn/4dNDF
Authors: Anurag Dalal, Daniel Hagen, Kjell G. Robbersmyr, Kristian Muri Knausg{\aa}rd
Abstract: Image-based 3D reconstruction is a challenging task that involves inferring the 3D shape of an object or scene from a set of input images. Learning-based methods have gained attention for their ability to directly estimate 3D shapes. This review paper focuses on state-of-the-art techniques for 3D reconstruction, including the generation of novel, unseen views. An overview of recent developments in the Gaussian Splatting method is provided, covering input types, model structures, output representations, and training strategies. Unresolved challenges and future directions are also discussed. Given the rapid progress in this domain and the numerous opportunities for enhancing 3D reconstruction methods, a comprehensive examination of algorithms appears essential. Consequently, this study offers a thorough overview of the latest advancements in Gaussian Splatting.
Authors: Emil Benedykciuk, Marcin Denkowski, Grzegorz W\'ojcik
Abstract: This paper introduces a novel approach to enhance the performance of pre-trained neural networks in medical image segmentation using Neural Architecture Search (NAS) methods, specifically Differentiable Architecture Search (DARTS). We present the concept of Implantable Adaptive Cell (IAC), small but powerful modules identified through Partially-Connected DARTS, designed to be injected into the skip connections of an existing and already trained U-shaped model. Our strategy allows for the seamless integration of the IAC into the pre-existing architecture, thereby enhancing its performance without necessitating a complete retraining from scratch. The empirical studies, focusing on medical image segmentation tasks, demonstrate the efficacy of this method. The integration of specialized IAC cells into various configurations of the U-Net model increases segmentation accuracy by almost 2\% points on average for the validation dataset and over 3\% points for the training dataset. The findings of this study not only offer a cost-effective alternative to the complete overhaul of complex models for performance upgrades but also indicate the potential applicability of our method to other architectures and problem domains.
Authors: Chengxin Zhao, Hefei Ling, Sijing Xie, Nan Sun, Zongyi Li, Yuxuan Shi, Jiazhong Chen
Abstract: Embedding invisible hyperlinks or hidden codes in images to replace QR codes has become a hot topic recently. This technology requires first localizing the embedded region in the captured photos before decoding. Existing methods that train models to find the invisible embedded region struggle to obtain accurate localization results, leading to degraded decoding accuracy. This limitation is primarily because the CNN network is sensitive to low-frequency signals, while the embedded signal is typically in the high-frequency form. Based on this, this paper proposes a Dual-Branch Dual-Head (DBDH) neural network tailored for the precise localization of invisible embedded regions. Specifically, DBDH uses a low-level texture branch containing 62 high-pass filters to capture the high-frequency signals induced by embedding. A high-level context branch is used to extract discriminative features between the embedded and normal regions. DBDH employs a detection head to directly detect the four vertices of the embedding region. In addition, we introduce an extra segmentation head to segment the mask of the embedding region during training. The segmentation head provides pixel-level supervision for model learning, facilitating better learning of the embedded signals. Based on two state-of-the-art invisible offline-to-online messaging methods, we construct two datasets and augmentation strategies for training and testing localization models. Extensive experiments demonstrate the superior performance of the proposed DBDH over existing methods.
Authors: Chengxin Zhao, Hefei Ling, Sijing Xie, Han Fang, Yaokun Fang, Nan Sun
Abstract: Modern image processing tools have made it easy for attackers to crop the region or object of interest in images and paste it into other images. The challenge this cropping-paste attack poses to the watermarking technology is that it breaks the synchronization of the image watermark, introducing multiple superimposed desynchronization distortions, such as rotation, scaling, and translation. However, current watermarking methods can only resist a single type of desynchronization and cannot be applied to protect the object's copyright under the cropping-paste attack. With the finding that the key to resisting the cropping-paste attack lies in robust features of the object to protect, this paper proposes a self-synchronizing object-aligned watermarking method, called SSyncOA. Specifically, we first constrain the watermarked region to be aligned with the protected object, and then synchronize the watermark's translation, rotation, and scaling distortions by normalizing the object invariant features, i.e., its centroid, principal orientation, and minimum bounding square, respectively. To make the watermark embedded in the protected object, we introduce the object-aligned watermarking model, which incorporates the real cropping-paste attack into the encoder-noise layer-decoder pipeline and is optimized end-to-end. Besides, we illustrate the effect of different desynchronization distortions on the watermark training, which confirms the necessity of the self-synchronization process. Extensive experiments demonstrate the superiority of our method over other SOTAs.
Authors: Lunchen Xie, Eugenio Lomurno, Matteo Gambella, Danilo Ardagna, Manuel Roveri, Matteo Matteucci, Qingjiang Shi
Abstract: Accurate classification of medical images is essential for modern diagnostics. Deep learning advancements led clinicians to increasingly use sophisticated models to make faster and more accurate decisions, sometimes replacing human judgment. However, model development is costly and repetitive. Neural Architecture Search (NAS) provides solutions by automating the design of deep learning architectures. This paper presents ZO-DARTS+, a differentiable NAS algorithm that improves search efficiency through a novel method of generating sparse probabilities by bi-level optimization. Experiments on five public medical datasets show that ZO-DARTS+ matches the accuracy of state-of-the-art solutions while reducing search times by up to three times.
Authors: Haowen Sun, Ruikun Zheng, Haibin Huang, Chongyang Ma, Hui Huang, Ruizhen Hu
Abstract: In this paper, we introduce LGTM, a novel Local-to-Global pipeline for Text-to-Motion generation. LGTM utilizes a diffusion-based architecture and aims to address the challenge of accurately translating textual descriptions into semantically coherent human motion in computer animation. Specifically, traditional methods often struggle with semantic discrepancies, particularly in aligning specific motions to the correct body parts. To address this issue, we propose a two-stage pipeline to overcome this challenge: it first employs large language models (LLMs) to decompose global motion descriptions into part-specific narratives, which are then processed by independent body-part motion encoders to ensure precise local semantic alignment. Finally, an attention-based full-body optimizer refines the motion generation results and guarantees the overall coherence. Our experiments demonstrate that LGTM gains significant improvements in generating locally accurate, semantically-aligned human motion, marking a notable advancement in text-to-motion applications. Code and data for this paper are available at https://github.com/L-Sun/LGTM
Authors: Pengpeng Li, Haowei Gu, Yang Yang
Abstract: In this competition we employed a model fusion approach to achieve object detection results close to those of real images. Our method is based on the CO-DETR model, which was trained on two sets of data: one containing images under dark conditions and another containing images enhanced with low-light conditions. We used various enhancement techniques on the test data to generate multiple sets of prediction results. Finally, we applied a clustering aggregation method guided by IoU thresholds to select the optimal results.
Authors: Zheng Zhu, Xiaofeng Wang, Wangbo Zhao, Chen Min, Nianchen Deng, Min Dou, Yuqi Wang, Botian Shi, Kai Wang, Chi Zhang, Yang You, Zhaoxiang Zhang, Dawei Zhao, Liang Xiao, Jian Zhao, Jiwen Lu, Guan Huang
Abstract: General world models represent a crucial pathway toward achieving Artificial General Intelligence (AGI), serving as the cornerstone for various applications ranging from virtual environments to decision-making systems. Recently, the emergence of the Sora model has attained significant attention due to its remarkable simulation capabilities, which exhibits an incipient comprehension of physical laws. In this survey, we embark on a comprehensive exploration of the latest advancements in world models. Our analysis navigates through the forefront of generative methodologies in video generation, where world models stand as pivotal constructs facilitating the synthesis of highly realistic visual content. Additionally, we scrutinize the burgeoning field of autonomous-driving world models, meticulously delineating their indispensable role in reshaping transportation and urban mobility. Furthermore, we delve into the intricacies inherent in world models deployed within autonomous agents, shedding light on their profound significance in enabling intelligent interactions within dynamic environmental contexts. At last, we examine challenges and limitations of world models, and discuss their potential future directions. We hope this survey can serve as a foundational reference for the research community and inspire continued innovation. This survey will be regularly updated at: https://github.com/GigaAI-research/General-World-Models-Survey.
URLs: https://github.com/GigaAI-research/General-World-Models-Survey.
Authors: Thennarasi Balakrishnan, Sandeep Singh Sengar
Abstract: Object detection algorithms particularly those based on YOLO have demonstrated remarkable efficiency in balancing speed and accuracy. However, their application in brain tumour detection remains underexplored. This study proposes RepVGG-GELAN, a novel YOLO architecture enhanced with RepVGG, a reparameterized convolutional approach for object detection tasks particularly focusing on brain tumour detection within medical images. RepVGG-GELAN leverages the RepVGG architecture to improve both speed and accuracy in detecting brain tumours. Integrating RepVGG into the YOLO framework aims to achieve a balance between computational efficiency and detection performance. This study includes a spatial pyramid pooling-based Generalized Efficient Layer Aggregation Network (GELAN) architecture which further enhances the capability of RepVGG. Experimental evaluation conducted on a brain tumour dataset demonstrates the effectiveness of RepVGG-GELAN surpassing existing RCS-YOLO in terms of precision and speed. Specifically, RepVGG-GELAN achieves an increased precision of 4.91% and an increased AP50 of 2.54% over the latest existing approach while operating at 240.7 GFLOPs. The proposed RepVGG-GELAN with GELAN architecture presents promising results establishing itself as a state-of-the-art solution for accurate and efficient brain tumour detection in medical images. The implementation code is publicly available at https://github.com/ThensiB/RepVGG-GELAN.
Authors: Amit Moryossef
Abstract: This paper addresses a critical flaw in MediaPipe Holistic's hand Region of Interest (ROI) prediction, which struggles with non-ideal hand orientations, affecting sign language recognition accuracy. We propose a data-driven approach to enhance ROI estimation, leveraging an enriched feature set including additional hand keypoints and the z-dimension. Our results demonstrate better estimates, with higher Intersection-over-Union compared to the current method. Our code and optimizations are available at https://github.com/sign-language-processing/mediapipe-hand-crop-fix.
URLs: https://github.com/sign-language-processing/mediapipe-hand-crop-fix.
Authors: Xin Ding, Yongwei Wang, Kao Zhang, Z. Jane Wang
Abstract: Continuous Conditional Generative Modeling (CCGM) aims to estimate the distribution of high-dimensional data, typically images, conditioned on scalar continuous variables known as regression labels. While Continuous conditional Generative Adversarial Networks (CcGANs) were initially designed for this task, their adversarial training mechanism remains vulnerable to extremely sparse or imbalanced data, resulting in suboptimal outcomes. To enhance the quality of generated images, a promising alternative is to replace CcGANs with Conditional Diffusion Models (CDMs), renowned for their stable training process and ability to produce more realistic images. However, existing CDMs encounter challenges when applied to CCGM tasks due to several limitations such as inadequate U-Net architectures and deficient model fitting mechanisms for handling regression labels. In this paper, we introduce Continuous Conditional Diffusion Models (CCDMs), the first CDM designed specifically for the CCGM task. CCDMs address the limitations of existing CDMs by introducing specially designed conditional diffusion processes, a modified denoising U-Net with a custom-made conditioning mechanism, a novel hard vicinal loss for model fitting, and an efficient conditional sampling procedure. With comprehensive experiments on four datasets with varying resolutions ranging from 64x64 to 192x192, we demonstrate the superiority of the proposed CCDM over state-of-the-art CCGM models, establishing new benchmarks in CCGM. Extensive ablation studies validate the model design and implementation configuration of the proposed CCDM. Our code is publicly available at https://github.com/UBCDingXin/CCDM.
Authors: Han Liu, Siyang Zhao, Xiaotong Zhang, Feng Zhang, Wei Wang, Fenglong Ma, Hongyang Chen, Hong Yu, Xianchao Zhang
Abstract: Few-shot and zero-shot text classification aim to recognize samples from novel classes with limited labeled samples or no labeled samples at all. While prevailing methods have shown promising performance via transferring knowledge from seen classes to unseen classes, they are still limited by (1) Inherent dissimilarities among classes make the transformation of features learned from seen classes to unseen classes both difficult and inefficient. (2) Rare labeled novel samples usually cannot provide enough supervision signals to enable the model to adjust from the source distribution to the target distribution, especially for complicated scenarios. To alleviate the above issues, we propose a simple and effective strategy for few-shot and zero-shot text classification. We aim to liberate the model from the confines of seen classes, thereby enabling it to predict unseen categories without the necessity of training on seen classes. Specifically, for mining more related unseen category knowledge, we utilize a large pre-trained language model to generate pseudo novel samples, and select the most representative ones as category anchors. After that, we convert the multi-class classification task into a binary classification task and use the similarities of query-anchor pairs for prediction to fully leverage the limited supervision signals. Extensive experiments on six widely used public datasets show that our proposed method can outperform other strong baselines significantly in few-shot and zero-shot tasks, even without using any seen class samples.
Authors: Jinwei Han, Yingguo Gao, Zhiwen Lin, Ke Yan, Shouhong Ding, Yuan Gao, Gui-Song Xia
Abstract: Zero-shot learning (ZSL) aims to recognize novel classes through transferring shared semantic knowledge (e.g., attributes) from seen classes to unseen classes. Recently, attention-based methods have exhibited significant progress which align visual features and attributes via a spatial attention mechanism. However, these methods only explore visual-semantic relationship in the spatial dimension, which can lead to classification ambiguity when different attributes share similar attention regions, and semantic relationship between attributes is rarely discussed. To alleviate the above problems, we propose a Dual Relation Mining Network (DRMN) to enable more effective visual-semantic interactions and learn semantic relationship among attributes for knowledge transfer. Specifically, we introduce a Dual Attention Block (DAB) for visual-semantic relationship mining, which enriches visual information by multi-level feature fusion and conducts spatial attention for visual to semantic embedding. Moreover, an attribute-guided channel attention is utilized to decouple entangled semantic features. For semantic relationship modeling, we utilize a Semantic Interaction Transformer (SIT) to enhance the generalization of attribute representations among images. Additionally, a global classification branch is introduced as a complement to human-defined semantic attributes, and we then combine the results with attribute-based classification. Extensive experiments demonstrate that the proposed DRMN leads to new state-of-the-art performances on three standard ZSL benchmarks, i.e., CUB, SUN, and AwA2.
Authors: Leonard Bruns, Jun Zhang, Patric Jensfelt
Abstract: Existing neural field-based SLAM methods typically employ a single monolithic field as their scene representation. This prevents efficient incorporation of loop closure constraints and limits scalability. To address these shortcomings, we propose a neural mapping framework which anchors lightweight neural fields to the pose graph of a sparse visual SLAM system. Our approach shows the ability to integrate large-scale loop closures, while limiting necessary reintegration. Furthermore, we verify the scalability of our approach by demonstrating successful building-scale mapping taking multiple loop closures into account during the optimization, and show that our method outperforms existing state-of-the-art approaches on large scenes in terms of quality and runtime. Our code is available at https://kth-rpl.github.io/neural_graph_mapping/.
Authors: Matina Mahdizadeh Sani, Ali Royat, Mahdieh Soleymani Baghshah
Abstract: Deep neural networks have reached remarkable achievements in medical image processing tasks, specifically classifying and detecting various diseases. However, when confronted with limited data, these networks face a critical vulnerability, often succumbing to overfitting by excessively memorizing the limited information available. This work addresses the challenge mentioned above by improving the supervised contrastive learning method to reduce the impact of false positives. Unlike most existing methods that rely predominantly on fully supervised learning, our approach leverages the advantages of self-supervised learning in conjunction with employing the available labeled data. We evaluate our method on the BreakHis dataset, which consists of breast cancer histopathology images, and demonstrate an increase in classification accuracy by 1.45% at the image level and 1.42% at the patient level compared to the state-of-the-art method. This improvement corresponds to 93.63% absolute accuracy, highlighting our approach's effectiveness in leveraging data properties to learn more appropriate representation space.
Authors: Samreen Anjum, Suyog Jain, Danna Gurari
Abstract: We propose a hybrid framework for consistently producing high-quality object tracks by combining an automated object tracker with little human input. The key idea is to tailor a module for each dataset to intelligently decide when an object tracker is failing and so humans should be brought in to re-localize an object for continued tracking. Our approach leverages self-supervised learning on unlabeled videos to learn a tailored representation for a target object that is then used to actively monitor its tracked region and decide when the tracker fails. Since labeled data is not needed, our approach can be applied to novel object categories. Experiments on three datasets demonstrate our method outperforms existing approaches, especially for small, fast moving, or occluded objects.
Authors: Mahdi Naseri, Jiayan Qiu, Zhou Wang
Abstract: In this paper, we investigate a novel artificial intelligence generation task, termed as generated contents enrichment (GCE). Different from conventional artificial intelligence contents generation task that enriches the given textual description implicitly with limited semantics for generating visually real content, our proposed GCE strives to perform content enrichment explicitly on both the visual and textual domain, from which the enriched contents are visually real, structurally reasonable, and semantically abundant. Towards to solve GCE, we propose a deep end-to-end method that explicitly explores the semantics and inter-semantic relationships during the enrichment. Specifically, we first model the input description as a semantic graph, wherein each node represents an object and each edge corresponds to the inter-object relationship. We then adopt Graph Convolutional Networks on top of the input scene description to predict the enriching objects and their relationships with the input objects. Finally, the enriched graph is fed into an image synthesis model to carry out the visual contents generation. Our experiments conducted on the Visual Genome dataset exhibit promising and visually plausible results.
Authors: Chenyu Gao, Shunxing Bao, Michael Kim, Nancy Newlin, Praitayini Kanakaraj, Tianyuan Yao, Gaurav Rudravaram, Yuankai Huo, Daniel Moyer, Kurt Schilling, Walter Kukull, Arthur Toga, Derek Archer, Timothy Hohman, Bennett Landman, Zhiyuan Li
Abstract: Purpose: In diffusion MRI (dMRI), the volumetric and bundle analyses of whole-brain tissue microstructure and connectivity can be severely impeded by an incomplete field-of-view (FOV). This work aims to develop a method for imputing the missing slices directly from existing dMRI scans with an incomplete FOV. We hypothesize that the imputed image with complete FOV can improve the whole-brain tractography for corrupted data with incomplete FOV. Therefore, our approach provides a desirable alternative to discarding the valuable dMRI data, enabling subsequent tractography analyses that would otherwise be challenging or unattainable with corrupted data. Approach: We propose a framework based on a deep generative model that estimates the absent brain regions in dMRI scans with incomplete FOV. The model is capable of learning both the diffusion characteristics in diffusion-weighted images (DWI) and the anatomical features evident in the corresponding structural images for efficiently imputing missing slices of DWI outside of incomplete FOV. Results: For evaluating the imputed slices, on the WRAP dataset the proposed framework achieved PSNRb0=22.397, SSIMb0=0.905, PSNRb1300=22.479, SSIMb1300=0.893; on the NACC dataset it achieved PSNRb0=21.304, SSIMb0=0.892, PSNRb1300=21.599, SSIMb1300= 0.877. The proposed framework improved the tractography accuracy, as demonstrated by an increased average Dice score for 72 tracts (p < 0.001) on both the WRAP and NACC datasets. Conclusions: Results suggest that the proposed framework achieved sufficient imputation performance in dMRI data with incomplete FOV for improving whole-brain tractography, thereby repairing the corrupted data. Our approach achieved more accurate whole-brain tractography results with extended and complete FOV and reduced the uncertainty when analyzing bundles associated with Alzheimer's Disease.
Authors: Kaiwen Jiang, Yang Fu, Mukund Varma T, Yash Belhe, Xiaolong Wang, Hao Su, Ravi Ramamoorthi
Abstract: Novel view synthesis from a sparse set of input images is a challenging problem of great practical interest, especially when camera poses are absent or inaccurate. Direct optimization of camera poses and usage of estimated depths in neural radiance field algorithms usually do not produce good results because of the coupling between poses and depths, and inaccuracies in monocular depth estimation. In this paper, we leverage the recent 3D Gaussian splatting method to develop a novel construct-and-optimize method for sparse view synthesis without camera poses. Specifically, we construct a solution progressively by using monocular depth and projecting pixels back into the 3D world. During construction, we optimize the solution by detecting 2D correspondences between training views and the corresponding rendered images. We develop a unified differentiable pipeline for camera registration and adjustment of both camera poses and depths, followed by back-projection. We also introduce a novel notion of an expected surface in Gaussian splatting, which is critical to our optimization. These steps enable a coarse solution, which can then be low-pass filtered and refined using standard optimization methods. We demonstrate results on the Tanks and Temples and Static Hikes datasets with as few as three widely-spaced views, showing significantly better quality than competing methods, including those with approximate camera pose information. Moreover, our results improve with more views and outperform previous InstantNGP and Gaussian Splatting algorithms even when using half the dataset.
Authors: Sankalp Sinha, Muhammad Saif Ullah Khan, Talha Uddin Sheikh, Didier Stricker, Muhammad Zeshan Afzal
Abstract: Zero-shot learning has been extensively investigated in the broader field of visual recognition, attracting significant interest recently. However, the current work on zero-shot learning in document image classification remains scarce. The existing studies either focus exclusively on zero-shot inference, or their evaluation does not align with the established criteria of zero-shot evaluation in the visual recognition domain. We provide a comprehensive document image classification analysis in Zero-Shot Learning (ZSL) and Generalized Zero-Shot Learning (GZSL) settings to address this gap. Our methodology and evaluation align with the established practices of this domain. Additionally, we propose zero-shot splits for the RVL-CDIP dataset. Furthermore, we introduce CICA (pronounced 'ki-ka'), a framework that enhances the zero-shot learning capabilities of CLIP. CICA consists of a novel 'content module' designed to leverage any generic document-related textual information. The discriminative features extracted by this module are aligned with CLIP's text and image features using a novel 'coupled-contrastive' loss. Our module improves CLIP's ZSL top-1 accuracy by 6.7% and GZSL harmonic mean by 24% on the RVL-CDIP dataset. Our module is lightweight and adds only 3.3% more parameters to CLIP. Our work sets the direction for future research in zero-shot document classification.
Authors: Dong Lao, Congli Wang, Alex Wong, Stefano Soatto
Abstract: We describe a method for recovering the irradiance underlying a collection of images corrupted by atmospheric turbulence. Since supervised data is often technically impossible to obtain, assumptions and biases have to be imposed to solve this inverse problem, and we choose to model them explicitly. Rather than initializing a latent irradiance ("template") by heuristics to estimate deformation, we select one of the images as a reference, and model the deformation in this image by the aggregation of the optical flow from it to other images, exploiting a prior imposed by Central Limit Theorem. Then with a novel flow inversion module, the model registers each image TO the template but WITHOUT the template, avoiding artifacts related to poor template initialization. To illustrate the robustness of the method, we simply (i) select the first frame as the reference and (ii) use the simplest optical flow to estimate the warpings, yet the improvement in registration is decisive in the final reconstruction, as we achieve state-of-the-art performance despite its simplicity. The method establishes a strong baseline that can be further improved by integrating it seamlessly into more sophisticated pipelines, or with domain-specific methods if so desired.
Authors: Qianning Wang, He Hu, Yucheng Zhou
Abstract: As automation advances in manufacturing, the demand for precise and sophisticated defect detection technologies grows. Existing vision models for defect recognition methods are insufficient for handling the complexities and variations of defects in contemporary manufacturing settings. These models especially struggle in scenarios involving limited or imbalanced defect data. In this work, we introduce MemoryMamba, a novel memory-augmented state space model (SSM), designed to overcome the limitations of existing defect recognition models. MemoryMamba integrates the state space model with the memory augmentation mechanism, enabling the system to maintain and retrieve essential defect-specific information in training. Its architecture is designed to capture dependencies and intricate defect characteristics, which are crucial for effective defect detection. In the experiments, MemoryMamba was evaluated across four industrial datasets with diverse defect types and complexities. The model consistently outperformed other methods, demonstrating its capability to adapt to various defect recognition scenarios.
Authors: Mira Slavcheva, Dave Gausebeck, Kevin Chen, David Buchhofer, Azwad Sabik, Chen Ma, Sachal Dhillon, Olaf Brandt, Alan Dolhasz
Abstract: We propose a pipeline that leverages Stable Diffusion to improve inpainting results in the context of defurnishing -- the removal of furniture items from indoor panorama images. Specifically, we illustrate how increased context, domain-specific model fine-tuning, and improved image blending can produce high-fidelity inpaints that are geometrically plausible without needing to rely on room layout estimation. We demonstrate qualitative and quantitative improvements over other furniture removal techniques.
Authors: Jang Hyun Cho, Boris Ivanovic, Yulong Cao, Edward Schmerling, Yue Wang, Xinshuo Weng, Boyi Li, Yurong You, Philipp Kr\"ahenb\"uhl, Yan Wang, Marco Pavone
Abstract: Multi-modal large language models (MLLMs) have shown incredible capabilities in a variety of 2D vision and language tasks. We extend MLLMs' perceptual capabilities to ground and reason about images in 3-dimensional space. To that end, we first develop a large-scale pre-training dataset for 2D and 3D called LV3D by combining multiple existing 2D and 3D recognition datasets under a common task formulation: as multi-turn question-answering. Next, we introduce a new MLLM named Cube-LLM and pre-train it on LV3D. We show that pure data scaling makes a strong 3D perception capability without 3D specific architectural design or training objective. Cube-LLM exhibits intriguing properties similar to LLMs: (1) Cube-LLM can apply chain-of-thought prompting to improve 3D understanding from 2D context information. (2) Cube-LLM can follow complex and diverse instructions and adapt to versatile input and output formats. (3) Cube-LLM can be visually prompted such as 2D box or a set of candidate 3D boxes from specialists. Our experiments on outdoor benchmarks demonstrate that Cube-LLM significantly outperforms existing baselines by 21.3 points of AP-BEV on the Talk2Car dataset for 3D grounded reasoning and 17.7 points on the DriveLM dataset for complex reasoning about driving scenarios, respectively. Cube-LLM also shows competitive results in general MLLM benchmarks such as refCOCO for 2D grounding with (87.0) average score, as well as visual question answering benchmarks such as VQAv2, GQA, SQA, POPE, etc. for complex reasoning. Our project is available at https://janghyuncho.github.io/Cube-LLM.
Authors: Sanjay Subramanian, Evonne Ng, Lea M\"uller, Dan Klein, Shiry Ginosar, Trevor Darrell
Abstract: We present a zero-shot pose optimization method that enforces accurate physical contact constraints when estimating the 3D pose of humans. Our central insight is that since language is often used to describe physical interaction, large pretrained text-based models can act as priors on pose estimation. We can thus leverage this insight to improve pose estimation by converting natural language descriptors, generated by a large multimodal model (LMM), into tractable losses to constrain the 3D pose optimization. Despite its simplicity, our method produces surprisingly compelling pose reconstructions of people in close contact, correctly capturing the semantics of the social and physical interactions. We demonstrate that our method rivals more complex state-of-the-art approaches that require expensive human annotation of contact points and training specialized models. Moreover, unlike previous approaches, our method provides a unified framework for resolving self-contact and person-to-person contact.
Authors: Muhammad Uzair Khattak, Muhammad Ferjad Naeem, Jameel Hassan, Muzammal Naseer, Federico Tombari, Fahad Shahbaz Khan, Salman Khan
Abstract: Recent advancements in Large Language Models (LLMs) have led to the development of Video Large Multi-modal Models (Video-LMMs) that can handle a wide range of video understanding tasks. These models have the potential to be deployed in real-world applications such as robotics, AI assistants, medical imaging, and autonomous vehicles. The widespread adoption of Video-LMMs in our daily lives underscores the importance of ensuring and evaluating their robust performance in mirroring human-like reasoning and interaction capabilities in complex, real-world contexts. However, existing benchmarks for Video-LMMs primarily focus on general video comprehension abilities and neglect assessing their reasoning capabilities over complex videos in the real-world context, and robustness of these models through the lens of user prompts as text queries. In this paper, we present the Complex Video Reasoning and Robustness Evaluation Suite (CVRR-ES), a novel benchmark that comprehensively assesses the performance of Video-LMMs across 11 diverse real-world video dimensions. We evaluate 9 recent models, including both open-source and closed-source variants, and find that most of the Video-LMMs, {especially open-source ones,} struggle with robustness and reasoning when dealing with complex videos. Based on our analysis, we develop a training-free Dual-Step Contextual Prompting (DSCP) technique to enhance the performance of existing Video-LMMs. Our findings provide valuable insights for building the next generation of human-centric AI systems with advanced robustness and reasoning capabilities. Our dataset and code are publicly available at: https://mbzuai-oryx.github.io/CVRR-Evaluation-Suite/.
Authors: Dahyun Jeong, Hyelim Son, Yunjin Choi, Keunwoo Kim
Abstract: Our study presents a framework for predicting image-based social media content popularity that focuses on addressing complex image information and a hierarchical data structure. We utilize the Google Cloud Vision API to effectively extract key image and color information from users' postings, achieving 6.8\% higher accuracy compared to using non-image covariates alone. For prediction, we explore a wide range of prediction models, including Linear Mixed Model, Support Vector Regression, Multi-layer Perceptron, Random Forest, and XGBoost, with linear regression as the benchmark. Our comparative study demonstrates that models that are capable of capturing the underlying nonlinear interactions between covariates outperform other methods.
Authors: Anna Hedstr\"om, Leander Weber, Sebastian Lapuschkin, Marina H\"ohne
Abstract: The Model Parameter Randomisation Test (MPRT) is highly recognised in the eXplainable Artificial Intelligence (XAI) community due to its fundamental evaluative criterion: explanations should be sensitive to the parameters of the model they seek to explain. However, recent studies have raised several methodological concerns for the empirical interpretation of MPRT. In response, we propose two modifications to the original test: Smooth MPRT and Efficient MPRT. The former reduces the impact of noise on evaluation outcomes via sampling, while the latter avoids the need for biased similarity measurements by re-interpreting the test through the increase in explanation complexity after full model randomisation. Our experiments show that these modifications enhance the metric reliability, facilitating a more trustworthy deployment of explanation methods.
Authors: Neil Dizon, Jyrki Jauhiainen, Tuomo Valkonen
Abstract: Online optimisation facilitates the solution of dynamic inverse problems, such as image stabilisation, fluid flow monitoring, and dynamic medical imaging. In this paper, we improve upon previous work on predictive online primal-dual methods on two fronts. Firstly, we provide a more concise analysis that symmetrises previously unsymmetric regret bounds, and relaxes previous restrictive conditions on the dual predictor. Secondly, based on the latter, we develop several improved dual predictors. We numerically demonstrate their efficacy in image stabilisation and dynamic positron emission tomography.
Authors: Minhui Yu, Mengqi Wu, Ling Yue, Andrea Bozoki, Mingxia Liu
Abstract: Magnetic resonance imaging (MRI) and positron emission tomography (PET) are increasingly used in multimodal analysis of neurodegenerative disorders. While MRI is broadly utilized in clinical settings, PET is less accessible. Many studies have attempted to use deep generative models to synthesize PET from MRI scans. However, they often suffer from unstable training and inadequately preserve brain functional information conveyed by PET. To this end, we propose a functional imaging constrained diffusion (FICD) framework for 3D brain PET image synthesis with paired structural MRI as input condition, through a new constrained diffusion model (CDM). The FICD introduces noise to PET and then progressively removes it with CDM, ensuring high output fidelity throughout a stable training phase. The CDM learns to predict denoised PET with a functional imaging constraint introduced to ensure voxel-wise alignment between each denoised PET and its ground truth. Quantitative and qualitative analyses conducted on 293 subjects with paired T1-weighted MRI and 18F-fluorodeoxyglucose (FDG)-PET scans suggest that FICD achieves superior performance in generating FDG-PET data compared to state-of-the-art methods. We further validate the effectiveness of the proposed FICD on data from a total of 1,262 subjects through three downstream tasks, with experimental results suggesting its utility and generalizability.
Authors: Coby Penso, Jacob Goldberger
Abstract: Conformal Prediction (CP) quantifies network uncertainty by building a small prediction set with a pre-defined probability that the correct class is within this set. In this study we tackle the problem of CP calibration based on a validation set with noisy labels. We introduce a conformal score that is robust to label noise. The noise-free conformal score is estimated using the noisy labeled data and the noise level. In the test phase the noise-free score is used to form the prediction set. We applied the proposed algorithm to several standard medical imaging classification datasets. We show that our method outperforms current methods by a large margin, in terms of the average size of the prediction set, while maintaining the required coverage.
Authors: M. Saquib Sarfraz, Mei-Yen Chen, Lukas Layer, Kunyu Peng, Marios Koulakis
Abstract: The current state of machine learning scholarship in Timeseries Anomaly Detection (TAD) is plagued by the persistent use of flawed evaluation metrics, inconsistent benchmarking practices, and a lack of proper justification for the choices made in novel deep learning-based model designs. Our paper presents a critical analysis of the status quo in TAD, revealing the misleading track of current research and highlighting problematic methods, and evaluation practices. Our position advocates for a shift in focus from pursuing only the novelty in model design to improving benchmarking practices, creating non-trivial datasets, and placing renewed emphasis on studying the utility of model architectures for specific tasks. Our findings demonstrate the need for rigorous evaluation protocols, the creation of simple baselines, and the revelation that state-of-the-art deep anomaly detection models effectively learn linear mappings. These findings suggest the need for more exploration and development of simple and interpretable TAD methods. The increment of model complexity in the state-of-the-art deep-learning based models unfortunately offers very little improvement. We offer insights and suggestions for the field to move forward.
Authors: Eugenio Lomurno, Matteo D'Oria, Matteo Matteucci
Abstract: Recent advances in generative artificial intelligence have enabled the creation of high-quality synthetic data that closely mimics real-world data. This paper explores the adaptation of the Stable Diffusion 2.0 model for generating synthetic datasets, using Transfer Learning, Fine-Tuning and generation parameter optimisation techniques to improve the utility of the dataset for downstream classification tasks. We present a class-conditional version of the model that exploits a Class-Encoder and optimisation of key generation parameters. Our methodology led to synthetic datasets that, in a third of cases, produced models that outperformed those trained on real datasets.
Authors: Jingwei Zhang, Mohammad Jalali, Cheuk Ting Li, Farzan Farnia
Abstract: An interpretable comparison of generative models requires the identification of sample types produced more frequently by each of the involved models. While several quantitative scores have been proposed in the literature to rank different generative models, such score-based evaluations do not reveal the nuanced differences between the generative models in capturing various sample types. In this work, we propose a method called Fourier-based Identification of Novel Clusters (FINC) to identify modes produced by a generative model with a higher frequency in comparison to a reference distribution. FINC provides a scalable stochastic algorithm based on random Fourier features to estimate the eigenspace of kernel covariance matrices of two generative models and utilize the principal eigendirections to detect the sample types present more dominantly in each model. We demonstrate the application of the FINC method to standard computer vision datasets and generative model frameworks. Our numerical results suggest the scalability and efficiency of the developed Fourier-based method in highlighting the sample types captured with different frequencies by widely-used generative models.
Authors: Fahad Sarfraz, Bahram Zonooz, Elahe Arani
Abstract: While humans excel at continual learning (CL), deep neural networks (DNNs) exhibit catastrophic forgetting. A salient feature of the brain that allows effective CL is that it utilizes multiple modalities for learning and inference, which is underexplored in DNNs. Therefore, we study the role and interactions of multiple modalities in mitigating forgetting and introduce a benchmark for multimodal continual learning. Our findings demonstrate that leveraging multiple views and complementary information from multiple modalities enables the model to learn more accurate and robust representations. This makes the model less vulnerable to modality-specific regularities and considerably mitigates forgetting. Furthermore, we observe that individual modalities exhibit varying degrees of robustness to distribution shift. Finally, we propose a method for integrating and aligning the information from different modalities by utilizing the relational structural similarities between the data points in each modality. Our method sets a strong baseline that enables both single- and multimodal inference. Our study provides a promising case for further exploring the role of multiple modalities in enabling CL and provides a standard benchmark for future research.
Authors: Chaojie Zhang, Shengjia Chen, Ozkan Cigdem, Haresh Rengaraj Rajamohan, Kyunghyun Cho, Richard Kijowski, Cem M. Deniz
Abstract: A transformer-based deep learning model, MR-Transformer, was developed for total knee replacement (TKR) prediction using magnetic resonance imaging (MRI). The model incorporates the ImageNet pre-training and captures three-dimensional (3D) spatial correlation from the MR images. The performance of the proposed model was compared to existing state-of-the-art deep learning models for knee injury diagnosis using MRI. Knee MR scans of four different tissue contrasts from the Osteoarthritis Initiative and Multicenter Osteoarthritis Study databases were utilized in the study. Experimental results demonstrated the state-of-the-art performance of the proposed model on TKR prediction using MRI.
Authors: Leye Zhang, Xiangxiang Tian, Hongjun Zhang
Abstract: Attempt to use convolutional neural network to achieve kinematic analysis of plane bar structure. Through 3dsMax animation software and OpenCV module, self-build image dataset of geometrically stable system and geometrically unstable system. we construct and train convolutional neural network model based on the TensorFlow and Keras deep learning platform framework. The model achieves 100% accuracy on the training set, validation set, and test set. The accuracy on the additional test set is 93.7%, indicating that convolutional neural network can learn and master the relevant knowledge of kinematic analysis of structural mechanics. In the future, the generalization ability of the model can be improved through the diversity of dataset, which has the potential to surpass human experts for complex structures. Convolutional neural network has certain practical value in the field of kinematic analysis of structural mechanics. Using visualization technology, we reveal how convolutional neural network learns and recognizes structural features. Using pre-trained VGG16 model for feature extraction and fine-tuning, we found that the generalization ability is inferior to the self-built model.
Authors: Fadillah Maani, Anees Ur Rehman Hashmi, Numan Saeed, Mohammad Yaqub
Abstract: Brain tumor segmentation is a fundamental step in assessing a patient's cancer progression. However, manual segmentation demands significant expert time to identify tumors in 3D multimodal brain MRI scans accurately. This reliance on manual segmentation makes the process prone to intra- and inter-observer variability. This work proposes a brain tumor segmentation method as part of the BraTS-GoAT challenge. The task is to segment tumors in brain MRI scans automatically from various populations, such as adults, pediatrics, and underserved sub-Saharan Africa. We employ a recent CNN architecture for medical image segmentation, namely MedNeXt, as our baseline, and we implement extensive model ensembling and postprocessing for inference. Our experiments show that our method performs well on the unseen validation set with an average DSC of 85.54% and HD95 of 27.88. The code is available on https://github.com/BioMedIA-MBZUAI/BraTS2024_BioMedIAMBZ.
URLs: https://github.com/BioMedIA-MBZUAI/BraTS2024_BioMedIAMBZ.
Authors: Haofei Song, Xintian Mao, Jing Yu, Qingli Li, Yan Wang
Abstract: Medical imaging is limited by acquisition time and scanning equipment. CT and MR volumes, reconstructed with thicker slices, are anisotropic with high in-plane resolution and low through-plane resolution. We reveal an intriguing phenomenon that due to the mentioned nature of data, performing slice-wise interpolation from the axial view can yield greater benefits than performing super-resolution from other views. Based on this observation, we propose an Inter-Intra-slice Interpolation Network (I$^3$Net), which fully explores information from high in-plane resolution and compensates for low through-plane resolution. The through-plane branch supplements the limited information contained in low through-plane resolution from high in-plane resolution and enables continual and diverse feature learning. In-plane branch transforms features to the frequency domain and enforces an equal learning opportunity for all frequency bands in a global context learning paradigm. We further propose a cross-view block to take advantage of the information from all three views online. Extensive experiments on two public datasets demonstrate the effectiveness of I$^3$Net, and noticeably outperforms state-of-the-art super-resolution, video frame interpolation and slice interpolation methods by a large margin. We achieve 43.90dB in PSNR, with at least 1.14dB improvement under the upscale factor of $\times$2 on MSD dataset with faster inference. Code is available at https://github.com/DeepMed-Lab-ECNU/Medical-Image-Reconstruction.
URLs: https://github.com/DeepMed-Lab-ECNU/Medical-Image-Reconstruction.
Authors: Shaohua Gao, Qi Jiang, Yiqi Liao, Yi Qiu, Wanglei Ying, Kailun Yang, Kaiwei Wang, Benhao Zhang, Jian Bai
Abstract: We propose a high-performance glass-plastic hybrid minimalist aspheric panoramic annular lens (ASPAL) to solve several major limitations of the traditional panoramic annular lens (PAL), such as large size, high weight, and complex system. The field of view (FoV) of the ASPAL is 360{\deg}x(35{\deg}~110{\deg}) and the imaging quality is close to the diffraction limit. This large FoV ASPAL is composed of only 4 lenses. Moreover, we establish a physical structure model of PAL using the ray tracing method and study the influence of its physical parameters on compactness ratio. In addition, for the evaluation of local tolerances of annular surfaces, we propose a tolerance analysis method suitable for ASPAL. This analytical method can effectively analyze surface irregularities on annular surfaces and provide clear guidance on manufacturing tolerances for ASPAL. Benefiting from high-precision glass molding and injection molding aspheric lens manufacturing techniques, we finally manufactured 20 ASPALs in small batches. The weight of an ASPAL prototype is only 8.5 g. Our framework provides promising insights for the application of panoramic systems in space and weight-constrained environmental sensing scenarios such as intelligent security, micro-UAVs, and micro-robots.
Authors: \c{S}\"ukr\"u \"Ozt\"urk, Hacer Yalim Keles
Abstract: This study introduces the continuous Educational Turkish Sign Language (E-TSL) dataset, collected from online Turkish language lessons for 5th, 6th, and 8th grades. The dataset comprises 1,410 videos totaling nearly 24 hours and includes performances from 11 signers. Turkish, an agglutinative language, poses unique challenges for sign language translation, particularly with a vocabulary where 64% are singleton words and 85% are rare words, appearing less than five times. We developed two baseline models to address these challenges: the Pose to Text Transformer (P2T-T) and the Graph Neural Network based Transformer (GNN-T) models. The GNN-T model achieved 19.13% BLEU-1 score and 3.28% BLEU-4 score, presenting a significant challenge compared to existing benchmarks. The P2T-T model, while demonstrating slightly lower performance in BLEU scores, achieved a higher ROUGE-L score of 22.09%. Additionally, we benchmarked our model using the well-known PHOENIX-Weather 2014T dataset to validate our approach.
Authors: Xiaoyan Lei, Wenlong ZHang, Weifeng Cao
Abstract: Efficient Image Super-Resolution (SR) aims to accelerate SR network inference by minimizing computational complexity and network parameters while preserving performance. Existing state-of-the-art Efficient Image Super-Resolution methods are based on convolutional neural networks. Few attempts have been made with Mamba to harness its long-range modeling capability and efficient computational complexity, which have shown impressive performance on high-level vision tasks. In this paper, we propose DVMSR, a novel lightweight Image SR network that incorporates Vision Mamba and a distillation strategy. The network of DVMSR consists of three modules: feature extraction convolution, multiple stacked Residual State Space Blocks (RSSBs), and a reconstruction module. Specifically, the deep feature extraction module is composed of several residual state space blocks (RSSB), each of which has several Vision Mamba Moudles(ViMM) together with a residual connection. To achieve efficiency improvement while maintaining comparable performance, we employ a distillation strategy to the vision Mamba network for superior performance. Specifically, we leverage the rich representation knowledge of teacher network as additional supervision for the output of lightweight student networks. Extensive experiments have demonstrated that our proposed DVMSR can outperform state-of-the-art efficient SR methods in terms of model parameters while maintaining the performance of both PSNR and SSIM. The source code is available at https://github.com/nathan66666/DVMSR.git
Authors: Jordan Dotzel, Yuzong Chen, Bahaa Kotb, Sushma Prasad, Gang Wu, Sheng Li, Mohamed S. Abdelfattah, Zhiru Zhang
Abstract: Large language models (LLMs) have recently achieved state-of-the-art performance across various tasks, yet due to their large computational requirements, they struggle with strict latency and power demands. Deep neural network (DNN) quantization has traditionally addressed these limitations by converting models to low-precision integer formats. Yet recently alternative formats, such as Normal Float (NF4), have been shown to consistently increase model accuracy, albeit at the cost of increased chip area. In this work, we first conduct a large-scale analysis of LLM weights and activations across 30 networks to conclude most distributions follow a Student's t-distribution. We then derive a new theoretically optimal format, Student Float (SF4), with respect to this distribution, that improves over NF4 across modern LLMs, for example increasing the average accuracy on LLaMA2-7B by 0.76% across tasks. Using this format as a high-accuracy reference, we then propose augmenting E2M1 with two variants of supernormal support for higher model accuracy. Finally, we explore the quality and performance frontier across 11 datatypes, including non-traditional formats like Additive-Powers-of-Two (APoT), by evaluating their model accuracy and hardware complexity. We discover a Pareto curve composed of INT4, E2M1, and E2M1 with supernormal support, which offers a continuous tradeoff between model accuracy and chip area. For example, E2M1 with supernormal support increases the accuracy of Phi-2 by up to 2.19% with 1.22% area overhead, enabling more LLM-based applications to be run at four bits.
Authors: Yihao Zhou, Timothy Tin-Yan Lee, Kelly Ka-Lee Lai, Chonglin Wu, Hin Ting Lau, De Yang, Chui-Yi Chan, Winnie Chiu-Wing Chu, Jack Chun-Yiu Cheng, Tsz-Ping Lam, Yong-Ping Zheng
Abstract: The current clinical gold standard for evaluating adolescent idiopathic scoliosis (AIS) is X-ray radiography, using Cobb angle measurement. However, the frequent monitoring of the AIS progression using X-rays poses a challenge due to the cumulative radiation exposure. Although 3D ultrasound has been validated as a reliable and radiation-free alternative for scoliosis assessment, the process of measuring spinal curvature is still carried out manually. Consequently, there is a considerable demand for a fully automatic system that can locate bony landmarks and perform angle measurements. To this end, we introduce an estimation model for automatic ultrasound curve angle (UCA) measurement. The model employs a dual-branch network to detect candidate landmarks and perform vertebra segmentation on ultrasound coronal images. An affinity clustering strategy is utilized within the vertebral segmentation area to illustrate the affinity relationship between candidate landmarks. Subsequently, we can efficiently perform line delineation from a clustered affinity map for UCA measurement. As our method is specifically designed for UCA calculation, this method outperforms other state-of-the-art methods for landmark and line detection tasks. The high correlation between the automatic UCA and Cobb angle (R$^2$=0.858) suggests that our proposed method can potentially replace manual UCA measurement in ultrasound scoliosis assessment.
Authors: Ransalu Senanayake
Abstract: Uncertainty has long been a critical area of study in robotics, particularly when robots are equipped with analytical models. As we move towards the widespread use of deep neural networks in robots, which have demonstrated remarkable performance in research settings, understanding the nuances of uncertainty becomes crucial for their real-world deployment. This guide offers an overview of the importance of uncertainty and provides methods to quantify and evaluate it from an applications perspective.
Authors: Matteo Bianchi, Antonio De Santis, Andrea Tocchetti, Marco Brambilla
Abstract: Transparency and explainability in image classification are essential for establishing trust in machine learning models and detecting biases and errors. State-of-the-art explainability methods generate saliency maps to show where a specific class is identified, without providing a detailed explanation of the model's decision process. Striving to address such a need, we introduce a post-hoc method that explains the entire feature extraction process of a Convolutional Neural Network. These explanations include a layer-wise representation of the features the model extracts from the input. Such features are represented as saliency maps generated by clustering and merging similar feature maps, to which we associate a weight derived by generalizing Grad-CAM for the proposed methodology. To further enhance these explanations, we include a set of textual labels collected through a gamified crowdsourcing activity and processed using NLP techniques and Sentence-BERT. Finally, we show an approach to generate global explanations by aggregating labels across multiple images.
Authors: Hangyu Lin, Chen Liu, Chengming Xu, Zhengqi Gao, Yanwei Fu, Yuan Yao
Abstract: Cross-modality distillation arises as an important topic for data modalities containing limited knowledge such as depth maps and high-quality sketches. Such techniques are of great importance, especially for memory and privacy-restricted scenarios where labeled training data is generally unavailable. To solve the problem, existing label-free methods leverage a few pairwise unlabeled data to distill the knowledge by aligning features or statistics between the source and target modalities. For instance, one typically aims to minimize the L2 distance or contrastive loss between the learned features of pairs of samples in the source (e.g. image) and the target (e.g. sketch) modalities. However, most algorithms in this domain only focus on the experimental results but lack theoretical insight. To bridge the gap between the theory and practical method of cross-modality distillation, we first formulate a general framework of cross-modality contrastive distillation (CMCD), built upon contrastive learning that leverages both positive and negative correspondence, towards a better distillation of generalizable features. Furthermore, we establish a thorough convergence analysis that reveals that the distance between source and target modalities significantly impacts the test error on downstream tasks within the target modality which is also validated by the empirical results. Extensive experimental results show that our algorithm outperforms existing algorithms consistently by a margin of 2-3\% across diverse modalities and tasks, covering modalities of image, sketch, depth map, and audio and tasks of recognition and segmentation.
Authors: Tao Han, zhenghao Chen, Song Guo, Wanghan Xu, Lei Bai
Abstract: The advent of data-driven weather forecasting models, which learn from hundreds of terabytes (TB) of reanalysis data, has significantly advanced forecasting capabilities. However, the substantial costs associated with data storage and transmission present a major challenge for data providers and users, affecting resource-constrained researchers and limiting their accessibility to participate in AI-based meteorological research. To mitigate this issue, we introduce an efficient neural codec, the Variational Autoencoder Transformer (VAEformer), for extreme compression of climate data to significantly reduce data storage cost, making AI-based meteorological research portable to researchers. Our approach diverges from recent complex neural codecs by utilizing a low-complexity Auto-Encoder transformer. This encoder produces a quantized latent representation through variance inference, which reparameterizes the latent space as a Gaussian distribution. This method improves the estimation of distributions for cross-entropy coding. Extensive experiments demonstrate that our VAEformer outperforms existing state-of-the-art compression methods in the context of climate data. By applying our VAEformer, we compressed the most popular ERA5 climate dataset (226 TB) into a new dataset, CRA5 (0.7 TB). This translates to a compression ratio of over 300 while retaining the dataset's utility for accurate scientific analysis. Further, downstream experiments show that global weather forecasting models trained on the compact CRA5 dataset achieve forecasting accuracy comparable to the model trained on the original dataset. Code, the CRA5 dataset, and the pre-trained model are available at https://github.com/taohan10200/CRA5.
Authors: Peng Jia, Yu Song, Jiameng Lv, Runyu Ning
Abstract: With the growing amount of astronomical data, there is an increasing need for automated data processing pipelines, which can extract scientific information from observation data without human interventions. A critical aspect of these pipelines is the image quality evaluation and masking algorithm, which evaluates image qualities based on various factors such as cloud coverage, sky brightness, scattering light from the optical system, point spread function size and shape, and read-out noise. Occasionally, the algorithm requires masking of areas severely affected by noise. However, the algorithm often necessitates significant human interventions, reducing data processing efficiency. In this study, we present a deep learning based image quality evaluation algorithm that uses an autoencoder to learn features of high quality astronomical images. The trained autoencoder enables automatic evaluation of image quality and masking of noise affected areas. We have evaluated the performance of our algorithm using two test cases: images with point spread functions of varying full width half magnitude, and images with complex backgrounds. In the first scenario, our algorithm could effectively identify variations of the point spread functions, which can provide valuable reference information for photometry. In the second scenario, our method could successfully mask regions affected by complex regions, which could significantly increase the photometry accuracy. Our algorithm can be employed to automatically evaluate image quality obtained by different sky surveying projects, further increasing the speed and robustness of data processing pipelines.
Authors: Yiting Qu, Xinyue Shen, Yixin Wu, Michael Backes, Savvas Zannettou, Yang Zhang
Abstract: Image safety classifiers play an important role in identifying and mitigating the spread of unsafe images online (e.g., images including violence, hateful rhetoric, etc.). At the same time, with the advent of text-to-image models and increasing concerns about the safety of AI models, developers are increasingly relying on image safety classifiers to safeguard their models. Yet, the performance of current image safety classifiers remains unknown for real-world and AI-generated images. To bridge this research gap, in this work, we propose UnsafeBench, a benchmarking framework that evaluates the effectiveness and robustness of image safety classifiers. First, we curate a large dataset of 10K real-world and AI-generated images that are annotated as safe or unsafe based on a set of 11 unsafe categories of images (sexual, violent, hateful, etc.). Then, we evaluate the effectiveness and robustness of five popular image safety classifiers, as well as three classifiers that are powered by general-purpose visual language models. Our assessment indicates that existing image safety classifiers are not comprehensive and effective enough in mitigating the multifaceted problem of unsafe images. Also, we find that classifiers trained only on real-world images tend to have degraded performance when applied to AI-generated images. Motivated by these findings, we design and implement a comprehensive image moderation tool called PerspectiveVision, which effectively identifies 11 categories of real-world and AI-generated unsafe images. The best PerspectiveVision model achieves an overall F1-Score of 0.810 on six evaluation datasets, which is comparable with closed-source and expensive state-of-the-art models like GPT-4V. UnsafeBench and PerspectiveVision can aid the research community in better understanding the landscape of image safety classification in the era of generative AI.
Authors: Yuefeng Zhang
Abstract: In lossy image compression, the objective is to achieve minimal signal distortion while compressing images to a specified bit rate. The increasing demand for visual analysis applications, particularly in classification tasks, has emphasized the significance of considering semantic distortion in compressed images. To bridge the gap between image compression and visual analysis, we propose a Rate-Distortion-Classification (RDC) model for lossy image compression, offering a unified framework to optimize the trade-off between rate, distortion, and classification accuracy. The RDC model is extensively analyzed both statistically on a multi-distribution source and experimentally on the widely used MNIST dataset. The findings reveal that the RDC model exhibits desirable properties, including monotonic non-increasing and convex functions, under certain conditions. This work provides insights into the development of human-machine friendly compression methods and Video Coding for Machine (VCM) approaches, paving the way for end-to-end image compression techniques in real-world applications.
Authors: Yanxi Chen, Chunxiao Li, Xinyang Dai, Jinhuan Li, Weiyu Sun, Yiming Wang, Renyuan Zhang, Tinghe Zhang, Bo Wang
Abstract: Multi-label learning (MLL) requires comprehensive multi-semantic annotations that is hard to fully obtain, thus often resulting in missing labels scenarios. In this paper, we investigate Single Positive Multi-label Learning (SPML), where each image is associated with merely one positive label. Existing SPML methods only focus on designing losses using mechanisms such as hard pseudo-labeling and robust losses, mostly leading to unacceptable false negatives. To address this issue, we first propose a generalized loss framework based on expected risk minimization to provide soft pseudo labels, and point out that the former losses can be seamlessly converted into our framework. In particular, we design a novel robust loss based on our framework, which enjoys flexible coordination between false positives and false negatives, and can additionally deal with the imbalance between positive and negative samples. Extensive experiments show that our approach can significantly improve SPML performance and outperform the vast majority of state-of-the-art methods on all the four benchmarks.
Authors: Guangtao Zheng, Wenqian Ye, Aidong Zhang
Abstract: Deep neural classifiers tend to rely on spurious correlations between spurious attributes of inputs and targets to make predictions, which could jeopardize their generalization capability. Training classifiers robust to spurious correlations typically relies on annotations of spurious correlations in data, which are often expensive to get. In this paper, we tackle an annotation-free setting and propose a self-guided spurious correlation mitigation framework. Our framework automatically constructs fine-grained training labels tailored for a classifier obtained with empirical risk minimization to improve its robustness against spurious correlations. The fine-grained training labels are formulated with different prediction behaviors of the classifier identified in a novel spuriousness embedding space. We construct the space with automatically detected conceptual attributes and a novel spuriousness metric which measures how likely a class-attribute correlation is exploited for predictions. We demonstrate that training the classifier to distinguish different prediction behaviors reduces its reliance on spurious correlations without knowing them a priori and outperforms prior methods on five real-world datasets.
Authors: Ximing Yang, Zhibo Zhang, Zhengfu He, Cheng Jin
Abstract: As details are missing in most representations of structures, the lack of controllability to more information is one of the major weaknesses in structure-based controllable point cloud generation. It is observable that definitions of details and structures are subjective. Details can be treated as structures on small scales. To represent structures in different scales at the same time, we present a graph-based representation of structures called the Multiscale Structure Graph (MSG). Given structures in multiple scales, similar patterns of local structures can be found at different scales, positions, and angles. The knowledge learned from a regional structure pattern shall be transferred to other similar patterns. An encoding and generation mechanism, namely the Multiscale Structure-based Point Cloud Generator (MSPCG) is proposed, which can simultaneously learn point cloud generation from local patterns with miscellaneous spatial properties. The proposed method supports multiscale editions on point clouds by editing the MSG. By generating point clouds from local structures and learning simultaneously in multiple scales, our MSPCG has better generalization ability and scalability. Trained on the ShapeNet, our MSPCG can generate point clouds from a given structure for unseen categories and indoor scenes. The experimental results show that our method significantly outperforms baseline methods.
Authors: Mohammed Hassanin, Saeed Anwar, Ibrahim Radwan, Fahad S Khan, Ajmal Mian
Abstract: Inspired by the human cognitive system, attention is a mechanism that imitates the human cognitive awareness about specific information, amplifying critical details to focus more on the essential aspects of data. Deep learning has employed attention to boost performance for many applications. Interestingly, the same attention design can suit processing different data modalities and can easily be incorporated into large networks. Furthermore, multiple complementary attention mechanisms can be incorporated into one network. Hence, attention techniques have become extremely attractive. However, the literature lacks a comprehensive survey on attention techniques to guide researchers in employing attention in their deep models. Note that, besides being demanding in terms of training data and computational resources, transformers only cover a single category in self-attention out of the many categories available. We fill this gap and provide an in-depth survey of 50 attention techniques, categorizing them by their most prominent features. We initiate our discussion by introducing the fundamental concepts behind the success of the attention mechanism. Next, we furnish some essentials such as the strengths and limitations of each attention category, describe their fundamental building blocks, basic formulations with primary usage, and applications specifically for computer vision. We also discuss the challenges and general open questions related to attention mechanisms. Finally, we recommend possible future research directions for deep attention. All the information about visual attention methods in deep learning is provided at \href{https://github.com/saeed-anwar/VisualAttention}{https://github.com/saeed-anwar/VisualAttention}
URLs: https://github.com/saeed-anwar/VisualAttention, https://github.com/saeed-anwar/VisualAttention
Authors: Wei-Chih Hung, Vincent Casser, Henrik Kretzschmar, Jyh-Jing Hwang, Dragomir Anguelov
Abstract: The 3D Average Precision (3D AP) relies on the intersection over union between predictions and ground truth objects. However, camera-only detectors have limited depth accuracy, which may cause otherwise reasonable predictions that suffer from such longitudinal localization errors to be treated as false positives. We therefore propose variants of the 3D AP metric to be more permissive with respect to depth estimation errors. Specifically, our novel longitudinal error tolerant metrics, LET-3D-AP and LET-3D-APL, allow longitudinal localization errors of the prediction boxes up to a given tolerance. To evaluate the proposed metrics, we also construct a new test set for the Waymo Open Dataset, tailored to camera-only 3D detection methods. Surprisingly, we find that state-of-the-art camera-based detectors can outperform popular LiDAR-based detectors with our new metrics past at 10% depth error tolerance, suggesting that existing camera-based detectors already have the potential to surpass LiDAR-based detectors in downstream applications. We believe the proposed metrics and the new benchmark dataset will facilitate advances in the field of camera-only 3D detection by providing more informative signals that can better indicate the system-level performance.
Authors: Feng Sun, Ming-Kun Xie, Sheng-Jun Huang
Abstract: In this paper, we study the partial multi-label (PML) image classification problem, where each image is annotated with a candidate label set consists of multiple relevant labels and other noisy labels. Existing PML methods typically design a disambiguation strategy to filter out noisy labels by utilizing prior knowledge with extra assumptions, which unfortunately is unavailable in many real tasks. Furthermore, because the objective function for disambiguation is usually elaborately designed on the whole training set, it can be hardly optimized in a deep model with SGD on mini-batches. In this paper, for the first time we propose a deep model for PML to enhance the representation and discrimination ability. On one hand, we propose a novel curriculum based disambiguation strategy to progressively identify ground-truth labels by incorporating the varied difficulties of different classes. On the other hand, a consistency regularization is introduced for model retraining to balance fitting identified easy labels and exploiting potential relevant labels. Extensive experimental results on the commonly used benchmark datasets show the proposed method significantly outperforms the SOTA methods.
Authors: Jinhee Kim, Taesung Kim, Taewoo Kim, Jaegul Choo, Dong-Wook Kim, Byungduk Ahn, In-Seok Song, Yoon-Ji Kim
Abstract: Diagnosis based on medical images, such as X-ray images, often involves manual annotation of anatomical keypoints. However, this process involves significant human efforts and can thus be a bottleneck in the diagnostic process. To fully automate this procedure, deep-learning-based methods have been widely proposed and have achieved high performance in detecting keypoints in medical images. However, these methods still have clinical limitations: accuracy cannot be guaranteed for all cases, and it is necessary for doctors to double-check all predictions of models. In response, we propose a novel deep neural network that, given an X-ray image, automatically detects and refines the anatomical keypoints through a user-interactive system in which doctors can fix mispredicted keypoints with fewer clicks than needed during manual revision. Using our own collected data and the publicly available AASCE dataset, we demonstrate the effectiveness of the proposed method in reducing the annotation costs via extensive quantitative and qualitative results. A demo video of our approach is available on our project webpage.
Authors: Yunfeng Diao, He Wang, Tianjia Shao, Yong-Liang Yang, Kun Zhou, David Hogg, Meng Wang
Abstract: Human Activity Recognition (HAR) has been employed in a wide range of applications, e.g. self-driving cars, where safety and lives are at stake. Recently, the robustness of skeleton-based HAR methods have been questioned due to their vulnerability to adversarial attacks. However, the proposed attacks require the full-knowledge of the attacked classifier, which is overly restrictive. In this paper, we show such threats indeed exist, even when the attacker only has access to the input/output of the model. To this end, we propose the very first black-box adversarial attack approach in skeleton-based HAR called BASAR. BASAR explores the interplay between the classification boundary and the natural motion manifold. To our best knowledge, this is the first time data manifold is introduced in adversarial attacks on time series. Via BASAR, we find on-manifold adversarial samples are extremely deceitful and rather common in skeletal motions, in contrast to the common belief that adversarial samples only exist off-manifold. Through exhaustive evaluation, we show that BASAR can deliver successful attacks across classifiers, datasets, and attack modes. By attack, BASAR helps identify the potential causes of the model vulnerability and provides insights on possible improvements. Finally, to mitigate the newly identified threat, we propose a new adversarial training approach by leveraging the sophisticated distributions of on/off-manifold adversarial samples, called mixed manifold-based adversarial training (MMAT). MMAT can successfully help defend against adversarial attacks without compromising classification accuracy.
Authors: Zhenghao Xing, Tianyu Wang, Xiaowei Hu, Haoran Wu, Chi-Wing Fu, Pheng-Ann Heng
Abstract: Instance shadow detection, crucial for applications such as photo editing and light direction estimation, has undergone significant advancements in predicting shadow instances, object instances, and their associations. The extension of this task to videos presents challenges in annotating diverse video data and addressing complexities arising from occlusion and temporary disappearances within associations. In response to these challenges, we introduce ViShadow, a semi-supervised video instance shadow detection framework that leverages both labeled image data and unlabeled video data for training. ViShadow features a two-stage training pipeline: the first stage, utilizing labeled image data, identifies shadow and object instances through contrastive learning for cross-frame pairing. The second stage employs unlabeled videos, incorporating an associated cycle consistency loss to enhance tracking ability. A retrieval mechanism is introduced to manage temporary disappearances, ensuring tracking continuity. The SOBA-VID dataset, comprising unlabeled training videos and labeled testing videos, along with the SOAP-VID metric, is introduced for the quantitative evaluation of VISD solutions. The effectiveness of ViShadow is further demonstrated through various video-level applications such as video inpainting, instance cloning, shadow editing, and text-instructed shadow-object manipulation.
Authors: Yilun Wu, Federico Paredes-Vall\'es, Guido C. H. E. de Croon
Abstract: Inspired by frame-based methods, state-of-the-art event-based optical flow networks rely on the explicit construction of correlation volumes, which are expensive to compute and store, rendering them unsuitable for robotic applications with limited compute and energy budget. Moreover, correlation volumes scale poorly with resolution, prohibiting them from estimating high-resolution flow. We observe that the spatiotemporally continuous traces of events provide a natural search direction for seeking pixel correspondences, obviating the need to rely on gradients of explicit correlation volumes as such search directions. We introduce IDNet (Iterative Deblurring Network), a lightweight yet high-performing event-based optical flow network directly estimating flow from event traces without using correlation volumes. We further propose two iterative update schemes: "ID" which iterates over the same batch of events, and "TID" which iterates over time with streaming events in an online fashion. Our top-performing ID model sets a new state of the art on DSEC benchmark. Meanwhile, the base ID model is competitive with prior arts while using 80% fewer parameters, consuming 20x less memory footprint and running 40% faster on the NVidia Jetson Xavier NX. Furthermore, the TID model is even more efficient offering an additional 5x faster inference speed and 8 ms ultra-low latency at the cost of only a 9% performance drop, making it the only model among current literature capable of real-time operation while maintaining decent performance.
Authors: Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan
Abstract: Owing to the success of transformer models, recent works study their applicability in 3D medical segmentation tasks. Within the transformer models, the self-attention mechanism is one of the main building blocks that strives to capture long-range dependencies. However, the self-attention operation has quadratic complexity which proves to be a computational bottleneck, especially in volumetric medical imaging, where the inputs are 3D with numerous slices. In this paper, we propose a 3D medical image segmentation approach, named UNETR++, that offers both high-quality segmentation masks as well as efficiency in terms of parameters, compute cost, and inference speed. The core of our design is the introduction of a novel efficient paired attention (EPA) block that efficiently learns spatial and channel-wise discriminative features using a pair of inter-dependent branches based on spatial and channel attention. Our spatial attention formulation is efficient having linear complexity with respect to the input sequence length. To enable communication between spatial and channel-focused branches, we share the weights of query and key mapping functions that provide a complimentary benefit (paired attention), while also reducing the overall network parameters. Our extensive evaluations on five benchmarks, Synapse, BTCV, ACDC, BRaTs, and Decathlon-Lung, reveal the effectiveness of our contributions in terms of both efficiency and accuracy. On Synapse, our UNETR++ sets a new state-of-the-art with a Dice Score of 87.2%, while being significantly efficient with a reduction of over 71% in terms of both parameters and FLOPs, compared to the best method in the literature. Code: https://github.com/Amshaker/unetr_plus_plus.
Authors: Xiwei Xuan, Ziquan Deng, Hsuan-Tien Lin, Zhaodan Kong, Kwan-Liu Ma
Abstract: Researchers have proposed various methods for visually interpreting the Convolutional Neural Network (CNN) via saliency maps, which include Class-Activation-Map (CAM) based approaches as a leading family. However, in terms of the internal design logic, existing CAM-based approaches often overlook the causal perspective that answers the core "why" question to help humans understand the explanation. Additionally, current CNN explanations lack the consideration of both necessity and sufficiency, two complementary sides of a desirable explanation. This paper presents a causality-driven framework, SUNY, designed to rationalize the explanations toward better human understanding. Using the CNN model's input features or internal filters as hypothetical causes, SUNY generates explanations by bi-directional quantifications on both the necessary and sufficient perspectives. Extensive evaluations justify that SUNY not only produces more informative and convincing explanations from the angles of necessity and sufficiency, but also achieves performances competitive to other approaches across different CNN architectures over large-scale datasets, including ILSVRC2012 and CUB-200-2011.
Authors: Alexander Kolpakov, Michael Werman
Abstract: Robust Affine matching with Grassmannians (RAG) is a new algorithm to perform affine registration of point clouds. The algorithm is based on minimizing the Frobenius distance between two elements of the Grassmannian. For this purpose, an indefinite relaxation of the Quadratic Assignment Problem (QAP) is used, and several approaches to affine feature matching are studied and compared. Experiments demonstrate that RAG is more robust to noise and point discrepancy than previous methods.
Authors: Matthieu Delmas, Amine Kacete, Stephane Paquelet, Simon Leglaive, Renaud Seguier
Abstract: The classification of forged videos has been a challenge for the past few years. Deepfake classifiers can now reliably predict whether or not video frames have been tampered with. However, their performance is tied to both the dataset used for training and the analyst's computational power. We propose a deepfake detection method that operates in the latent space of a state-of-the-art generative adversarial network (GAN) trained on high-quality face images. The proposed method leverages the structure of the latent space of StyleGAN to learn a lightweight binary classification model. Experimental results on standard datasets reveal that the proposed approach outperforms other state-of-the-art deepfake classification methods, especially in contexts where the data available to train the models is rare, such as when a new manipulation method is introduced. To the best of our knowledge, this is the first study showing the interest of the latent space of StyleGAN for deepfake classification. Combined with other recent studies on the interpretation and manipulation of this latent space, we believe that the proposed approach can further help in developing frugal deepfake classification methods based on interpretable high-level properties of face images.
Authors: Jihan Yang, Runyu Ding, Weipeng Deng, Zhe Wang, Xiaojuan Qi
Abstract: We propose a lightweight and scalable Regional Point-Language Contrastive learning framework, namely \textbf{RegionPLC}, for open-world 3D scene understanding, aiming to identify and recognize open-set objects and categories. Specifically, based on our empirical studies, we introduce a 3D-aware SFusion strategy that fuses 3D vision-language pairs derived from multiple 2D foundation models, yielding high-quality, dense region-level language descriptions without human 3D annotations. Subsequently, we devise a region-aware point-discriminative contrastive learning objective to enable robust and effective 3D learning from dense regional language supervision. We carry out extensive experiments on ScanNet, ScanNet200, and nuScenes datasets, and our model outperforms prior 3D open-world scene understanding approaches by an average of 17.2\% and 9.1\% for semantic and instance segmentation, respectively, while maintaining greater scalability and lower resource demands. Furthermore, our method has the flexibility to be effortlessly integrated with language models to enable open-ended grounded 3D reasoning without extra task-specific training. Code is available at https://github.com/CVMI-Lab/PLA.
Authors: Sitian Shen, Zilin Zhu, Linqian Fan, Harry Zhang, Xinxiao Wu
Abstract: Large pre-trained models have had a significant impact on computer vision by enabling multi-modal learning, where the CLIP model has achieved impressive results in image classification, object detection, and semantic segmentation. However, the model's performance on 3D point cloud processing tasks is limited due to the domain gap between depth maps from 3D projection and training images of CLIP. This paper proposes DiffCLIP, a new pre-training framework that incorporates stable diffusion with ControlNet to minimize the domain gap in the visual branch. Additionally, a style-prompt generation module is introduced for few-shot tasks in the textual branch. Extensive experiments on the ModelNet10, ModelNet40, and ScanObjectNN datasets show that DiffCLIP has strong abilities for 3D understanding. By using stable diffusion and style-prompt generation, DiffCLIP achieves an accuracy of 43.2\% for zero-shot classification on OBJ\_BG of ScanObjectNN, which is state-of-the-art performance, and an accuracy of 80.6\% for zero-shot classification on ModelNet10, which is comparable to state-of-the-art performance.
Authors: Yukun Huang, Jianan Wang, Yukai Shi, Boshi Tang, Xianbiao Qi, Lei Zhang
Abstract: Text-to-image diffusion models pre-trained on billions of image-text pairs have recently enabled 3D content creation by optimizing a randomly initialized differentiable 3D representation with score distillation. However, the optimization process suffers slow convergence and the resultant 3D models often exhibit two limitations: (a) quality concerns such as missing attributes and distorted shape and texture; (b) extremely low diversity comparing to text-guided image synthesis. In this paper, we show that the conflict between the 3D optimization process and uniform timestep sampling in score distillation is the main reason for these limitations. To resolve this conflict, we propose to prioritize timestep sampling with monotonically non-increasing functions, which aligns the 3D optimization process with the sampling process of diffusion model. Extensive experiments show that our simple redesign significantly improves 3D content creation with faster convergence, better quality and diversity.
Authors: Sagnik Majumder, Ziad Al-Halah, Kristen Grauman
Abstract: We propose a self-supervised method for learning representations based on spatial audio-visual correspondences in egocentric videos. Our method uses a masked auto-encoding framework to synthesize masked binaural (multi-channel) audio through the synergy of audio and vision, thereby learning useful spatial relationships between the two modalities. We use our pretrained features to tackle two downstream video tasks requiring spatial understanding in social scenarios: active speaker detection and spatial audio denoising. Through extensive experiments, we show that our features are generic enough to improve over multiple state-of-the-art baselines on both tasks on two challenging egocentric video datasets that offer binaural audio, EgoCom and EasyCom. Project: http://vision.cs.utexas.edu/projects/ego_av_corr.
Authors: Soubhik Sanyal, Partha Ghosh, Jinlong Yang, Michael J. Black, Justus Thies, Timo Bolkart
Abstract: We present SCULPT, a novel 3D generative model for clothed and textured 3D meshes of humans. Specifically, we devise a deep neural network that learns to represent the geometry and appearance distribution of clothed human bodies. Training such a model is challenging, as datasets of textured 3D meshes for humans are limited in size and accessibility. Our key observation is that there exist medium-sized 3D scan datasets like CAPE, as well as large-scale 2D image datasets of clothed humans and multiple appearances can be mapped to a single geometry. To effectively learn from the two data modalities, we propose an unpaired learning procedure for pose-dependent clothed and textured human meshes. Specifically, we learn a pose-dependent geometry space from 3D scan data. We represent this as per vertex displacements w.r.t. the SMPL model. Next, we train a geometry conditioned texture generator in an unsupervised way using the 2D image data. We use intermediate activations of the learned geometry model to condition our texture generator. To alleviate entanglement between pose and clothing type, and pose and clothing appearance, we condition both the texture and geometry generators with attribute labels such as clothing types for the geometry, and clothing colors for the texture generator. We automatically generated these conditioning labels for the 2D images based on the visual question answering model BLIP and CLIP. We validate our method on the SCULPT dataset, and compare to state-of-the-art 3D generative models for clothed human bodies. Our code and data can be found at https://sculpt.is.tue.mpg.de.
Authors: Yuanhao Gong
Abstract: Inspired by Federated Learning, in this paper, we propose personal large models that are distilled from traditional large language models but more adaptive to local users' personal information such as education background and hobbies. We classify the large language models into three levels: the personal level, expert level and traditional level. The personal level models are adaptive to users' personal information. They encrypt the users' input and protect their privacy. The expert level models focus on merging specific knowledge such as finance, IT and art. The traditional models focus on the universal knowledge discovery and upgrading the expert models. In such classifications, the personal models directly interact with the user. For the whole system, the personal models have users' (encrypted) personal information. Moreover, such models must be small enough to be performed on personal computers or mobile devices. Finally, they also have to response in real-time for better user experience and produce high quality results. The proposed personal large models can be applied in a wide range of applications such as language and vision tasks.
Authors: Long Lian, Baifeng Shi, Adam Yala, Trevor Darrell, Boyi Li
Abstract: Text-conditioned diffusion models have emerged as a promising tool for neural video generation. However, current models still struggle with intricate spatiotemporal prompts and often generate restricted or incorrect motion. To address these limitations, we introduce LLM-grounded Video Diffusion (LVD). Instead of directly generating videos from the text inputs, LVD first leverages a large language model (LLM) to generate dynamic scene layouts based on the text inputs and subsequently uses the generated layouts to guide a diffusion model for video generation. We show that LLMs are able to understand complex spatiotemporal dynamics from text alone and generate layouts that align closely with both the prompts and the object motion patterns typically observed in the real world. We then propose to guide video diffusion models with these layouts by adjusting the attention maps. Our approach is training-free and can be integrated into any video diffusion model that admits classifier guidance. Our results demonstrate that LVD significantly outperforms its base video diffusion model and several strong baseline methods in faithfully generating videos with the desired attributes and motion patterns.
Authors: Muhammad Hamza Asad, Saeed Anwar, Abdul Bais
Abstract: Modern agriculture heavily relies on Site-Specific Farm Management practices, necessitating accurate detection, localization, and quantification of crops and weeds in the field, which can be achieved using deep learning techniques. In this regard, crop and weed-specific binary segmentation models have shown promise. However, uncontrolled field conditions limit their performance from one field to the other. To improve semantic model generalization, existing methods augment and synthesize agricultural data to account for uncontrolled field conditions. However, given highly varied field conditions, these methods have limitations. To overcome the challenges of model deterioration in such conditions, we propose utilizing data specific to other crops and weeds for our specific target problem. To achieve this, we propose a novel ensemble framework. Our approach involves utilizing different crop and weed models trained on diverse datasets and employing a teacher-student configuration. By using homogeneous stacking of base models and a trainable meta-architecture to combine their outputs, we achieve significant improvements for Canola crops and Kochia weeds on unseen test data, surpassing the performance of single semantic segmentation models. We identify the UNET meta-architecture as the most effective in this context. Finally, through ablation studies, we demonstrate and validate the effectiveness of our proposed model. We observe that including base models trained on other target crops and weeds can help generalize the model to capture varied field conditions. Lastly, we propose two novel datasets with varied conditions for comparisons.
Authors: Xianzhong Liu, Holger Caesar
Abstract: To reduce the expensive labor cost for manual labeling autonomous driving datasets, an alternative is to automatically label the datasets using an offline perception system. However, objects might be temporally occluded. Such occlusion scenarios in the datasets are common yet underexplored in offline auto labeling. In this work, we propose an offline tracking model that focuses on occluded object tracks. It leverages the concept of object permanence which means objects continue to exist even if they are not observed anymore. The model contains three parts: a standard online tracker, a re-identification (Re-ID) module that associates tracklets before and after occlusion, and a track completion module that completes the fragmented tracks. The Re-ID module and the track completion module use the vectorized map as one of the inputs to refine the tracking results with occlusion. The model can effectively recover the occluded object trajectories. It achieves state-of-the-art performance in 3D multi-object tracking by significantly improving the original online tracking result, showing its potential to be applied in offline auto labeling as a useful plugin to improve tracking by recovering occlusions.
Authors: Yingqian Cui, Jie Ren, Yuping Lin, Han Xu, Pengfei He, Yue Xing, Lingjuan Lyu, Wenqi Fan, Hui Liu, Jiliang Tang
Abstract: Text-to-image generative models, especially those based on latent diffusion models (LDMs), have demonstrated outstanding ability in generating high-quality and high-resolution images from textual prompts. With this advancement, various fine-tuning methods have been developed to personalize text-to-image models for specific applications such as artistic style adaptation and human face transfer. However, such advancements have raised copyright concerns, especially when the data are used for personalization without authorization. For example, a malicious user can employ fine-tuning techniques to replicate the style of an artist without consent. In light of this concern, we propose FT-Shield, a watermarking solution tailored for the fine-tuning of text-to-image diffusion models. FT-Shield addresses copyright protection challenges by designing new watermark generation and detection strategies. In particular, it introduces an innovative algorithm for watermark generation. It ensures the seamless transfer of watermarks from training images to generated outputs, facilitating the identification of copyrighted material use. To tackle the variability in fine-tuning methods and their impact on watermark detection, FT-Shield integrates a Mixture of Experts (MoE) approach for watermark detection. Comprehensive experiments validate the effectiveness of our proposed FT-Shield.
Authors: Ziqi Pang, Ziyang Xie, Yunze Man, Yu-Xiong Wang
Abstract: This paper reveals that large language models (LLMs), despite being trained solely on textual data, are surprisingly strong encoders for purely visual tasks in the absence of language. Even more intriguingly, this can be achieved by a simple yet previously overlooked strategy -- employing a frozen transformer block from pre-trained LLMs as a constituent encoder layer to directly process visual tokens. Our work pushes the boundaries of leveraging LLMs for computer vision tasks, significantly departing from conventional practices that typically necessitate a multi-modal vision-language setup with associated language prompts, inputs, or outputs. We demonstrate that our approach consistently enhances performance across a diverse range of tasks, encompassing pure 2D and 3D visual recognition tasks (e.g., image and point cloud classification), temporal modeling tasks (e.g., action recognition), non-semantic tasks (e.g., motion forecasting), and multi-modal tasks (e.g., 2D/3D visual question answering and image-text retrieval). Such improvements are a general phenomenon, applicable to various types of LLMs (e.g., LLaMA and OPT) and different LLM transformer blocks. We additionally propose the information filtering hypothesis to explain the effectiveness of pre-trained LLMs in visual encoding -- the pre-trained LLM transformer blocks discern informative visual tokens and further amplify their effect. This hypothesis is empirically supported by the observation that the feature activation, after training with LLM transformer blocks, exhibits a stronger focus on relevant regions. We hope that our work inspires new perspectives on utilizing LLMs and deepening our understanding of their underlying mechanisms. Code is available at https://github.com/ziqipang/LM4VisualEncoding.
Authors: Bar{\i}\c{s} B\"uy\"ukta\c{s}, Gencer Sumbul, Beg\"um Demir
Abstract: Federated learning (FL) enables the collaboration of multiple deep learning models to learn from decentralized data archives (i.e., clients) without accessing data on clients. Although FL offers ample opportunities in knowledge discovery from distributed image archives, it is seldom considered in remote sensing (RS). In this paper, as a first time in RS, we present a comparative study of state-of-the-art FL algorithms for RS image classification problems. To this end, we initially provide a systematic review of the FL algorithms presented in the computer vision and machine learning communities. Then, we select several state-of-the-art FL algorithms based on their effectiveness with respect to training data heterogeneity across clients (known as non-IID data). After presenting an extensive overview of the selected algorithms, a theoretical comparison of the algorithms is conducted based on their: 1) local training complexity; 2) aggregation complexity; 3) learning efficiency; 4) communication cost; and 5) scalability in terms of number of clients. After the theoretical comparison, experimental analyses are presented to compare them under different decentralization scenarios. For the experimental analyses, we focus our attention on multi-label image classification problems in RS. Based on our comprehensive analyses, we finally derive a guideline for selecting suitable FL algorithms in RS. The code of this work will be publicly available at https://git.tu-berlin.de/rsim/FL-RS.
Authors: Omri Avrahami, Amir Hertz, Yael Vinker, Moab Arar, Shlomi Fruchter, Ohad Fried, Daniel Cohen-Or, Dani Lischinski
Abstract: Recent advances in text-to-image generation models have unlocked vast potential for visual creativity. However, these models struggle with generation of consistent characters, a crucial aspect for numerous real-world applications such as story visualization, game development asset design, advertising, and more. Current methods typically rely on multiple pre-existing images of the target character or involve labor-intensive manual processes. In this work, we propose a fully automated solution for consistent character generation, with the sole input being a text prompt. We introduce an iterative procedure that, at each stage, identifies a coherent set of images sharing a similar identity and extracts a more consistent identity from this set. Our quantitative analysis demonstrates that our method strikes a better balance between prompt alignment and identity consistency compared to the baseline methods, and these findings are reinforced by a user study. To conclude, we showcase several practical applications of our approach. Project page is available at https://omriavrahami.com/the-chosen-one
Authors: Paul J. Claasen, J. P. de Villiers
Abstract: A novel Bayesian framework is proposed, which explicitly relates the homography of one video frame to the next through an affine transformation while explicitly modelling keypoint uncertainty. The literature has previously used differential homography between subsequent frames, but not in a Bayesian setting. In cases where Bayesian methods have been applied, camera motion is not adequately modelled, and keypoints are treated as deterministic. The proposed method, Bayesian Homography Inference from Tracked Keypoints (BHITK), employs a two-stage Kalman filter and significantly improves existing methods. Existing keypoint detection methods may be easily augmented with BHITK. It enables less sophisticated and less computationally expensive methods to outperform the state-of-the-art approaches in most homography evaluation metrics. Furthermore, the homography annotations of the WorldCup and TS-WorldCup datasets have been refined using a custom homography annotation tool that has been released for public use. The refined datasets are consolidated and released as the consolidated and refined WorldCup (CARWC) dataset.
Authors: Di Chang, Yichun Shi, Quankai Gao, Jessica Fu, Hongyi Xu, Guoxian Song, Qing Yan, Yizhe Zhu, Xiao Yang, Mohammad Soleymani
Abstract: In this work, we propose MagicPose, a diffusion-based model for 2D human pose and facial expression retargeting. Specifically, given a reference image, we aim to generate a person's new images by controlling the poses and facial expressions while keeping the identity unchanged. To this end, we propose a two-stage training strategy to disentangle human motions and appearance (e.g., facial expressions, skin tone and dressing), consisting of (1) the pre-training of an appearance-control block and (2) learning appearance-disentangled pose control. Our novel design enables robust appearance control over generated human images, including body, facial attributes, and even background. By leveraging the prior knowledge of image diffusion models, MagicPose generalizes well to unseen human identities and complex poses without the need for additional fine-tuning. Moreover, the proposed model is easy to use and can be considered as a plug-in module/extension to Stable Diffusion. The code is available at: https://github.com/Boese0601/MagicDance
Authors: Jun-En Ding, Chien-Chin Hsu, Feng Liu
Abstract: Parkinson's Disease (PD) affects millions globally, impacting movement. Prior research utilized deep learning for PD prediction, primarily focusing on medical images, neglecting the data's underlying manifold structure. This work proposes a multimodal approach encompassing both image and non-image features, leveraging contrastive cross-view graph fusion for PD classification. We introduce a novel multimodal co-attention module, integrating embeddings from separate graph views derived from low-dimensional representations of images and clinical features. This enables more robust and structured feature extraction for improved multi-view data analysis. Additionally, a simplified contrastive loss-based fusion method is devised to enhance cross-view fusion learning. Our graph-view multimodal approach achieves an accuracy of 91% and an area under the receiver operating characteristic curve (AUC) of 92.8% in five-fold cross-validation. It also demonstrates superior predictive capabilities on non-image data compared to solely machine learning-based methods.
Authors: Etai Sella, Gal Fiebelman, Noam Atia, Hadar Averbuch-Elor
Abstract: We are witnessing rapid progress in automatically generating and manipulating 3D assets due to the availability of pretrained text-image diffusion models. However, time-consuming optimization procedures are required for synthesizing each sample, hindering their potential for democratizing 3D content creation. Conversely, 3D diffusion models now train on million-scale 3D datasets, yielding high-quality text-conditional 3D samples within seconds. In this work, we present Spice-E - a neural network that adds structural guidance to 3D diffusion models, extending their usage beyond text-conditional generation. At its core, our framework introduces a cross-entity attention mechanism that allows for multiple entities (in particular, paired input and guidance 3D shapes) to interact via their internal representations within the denoising network. We utilize this mechanism for learning task-specific structural priors in 3D diffusion models from auxiliary guidance shapes. We show that our approach supports a variety of applications, including 3D stylization, semantic shape editing and text-conditional abstraction-to-3D, which transforms primitive-based abstractions into highly-expressive shapes. Extensive experiments demonstrate that Spice-E achieves SOTA performance over these tasks while often being considerably faster than alternative methods. Importantly, this is accomplished without tailoring our approach for any specific task.
Authors: Marco Cotogni, Jacopo Bonato, Luigi Sabetta, Francesco Pelosin, Alessandro Nicolosi
Abstract: Machine Unlearning is rising as a new field, driven by the pressing necessity of ensuring privacy in modern artificial intelligence models. This technique primarily aims to eradicate any residual influence of a specific subset of data from the knowledge acquired by a neural model during its training. This work introduces a novel unlearning algorithm, denoted as Distance-based Unlearning via Centroid Kinematics (DUCK), which employs metric learning to guide the removal of samples matching the nearest incorrect centroid in the embedding space. Evaluation of the algorithm's performance is conducted across various benchmark datasets in two distinct scenarios, class removal, and homogeneous sampling removal, obtaining state-of-the-art performance. We also introduce a novel metric, called Adaptive Unlearning Score (AUS), encompassing not only the efficacy of the unlearning process in forgetting target data but also quantifying the performance loss relative to the original model. Additionally, we conducted a thorough investigation of the unlearning mechanism in DUCK, examining its impact on the organization of the feature space and employing explainable AI techniques for deeper insights.
Authors: Zhi Zhang, Qizhe Zhang, Zijun Gao, Renrui Zhang, Ekaterina Shutova, Shiji Zhou, Shanghang Zhang
Abstract: With the growing size of pre-trained models, full fine-tuning and storing all the parameters for various downstream tasks is costly and infeasible. In this paper, we propose a new parameter-efficient fine-tuning method, Gradient-based Parameter Selection (GPS), demonstrating that only tuning a few selected parameters from the pre-trained model while keeping the remainder of the model frozen can generate similar or better performance compared with the full model fine-tuning method. Different from the existing popular and state-of-the-art parameter-efficient fine-tuning approaches, our method does not introduce any additional parameters and computational costs during both the training and inference stages. Another advantage is the model-agnostic and non-destructive property, which eliminates the need for any other design specific to a particular model. Compared with the full fine-tuning, GPS achieves 3.33% (91.78% vs. 88.45%, FGVC) and 9.61% (73.1% vs. 65.57%, VTAB) improvement of the accuracy with tuning only 0.36% parameters of the pre-trained model on average over 24 image classification tasks; it also demonstrates a significant improvement of 17% and 16.8% in mDice and mIoU, respectively, on medical image segmentation task. Moreover, GPS achieves state-of-the-art performance compared with existing PEFT methods.
Authors: Jiye Lee, Hanbyul Joo
Abstract: We present a lightweight and affordable motion capture method based on two smartwatches and a head-mounted camera. In contrast to the existing approaches that use six or more expert-level IMU devices, our approach is much more cost-effective and convenient. Our method can make wearable motion capture accessible to everyone everywhere, enabling 3D full-body motion capture in diverse environments. As a key idea to overcome the extreme sparsity and ambiguities of sensor inputs with different modalities, we integrate 6D head poses obtained from the head-mounted cameras for motion estimation. To enable capture in expansive indoor and outdoor scenes, we propose an algorithm to track and update floor level changes to define head poses, coupled with a multi-stage Transformer-based regression module. We also introduce novel strategies leveraging visual cues of egocentric images to further enhance the motion capture quality while reducing ambiguities. We demonstrate the performance of our method on various challenging scenarios, including complex outdoor environments and everyday motions including object interactions and social interactions among multiple individuals.
Authors: Kumar Ashutosh, Zihui Xue, Tushar Nagarajan, Kristen Grauman
Abstract: We introduce the video detours problem for navigating instructional videos. Given a source video and a natural language query asking to alter the how-to video's current path of execution in a certain way, the goal is to find a related ''detour video'' that satisfies the requested alteration. To address this challenge, we propose VidDetours, a novel video-language approach that learns to retrieve the targeted temporal segments from a large repository of how-to's using video-and-text conditioned queries. Furthermore, we devise a language-based pipeline that exploits how-to video narration text to create weakly supervised training data. We demonstrate our idea applied to the domain of how-to cooking videos, where a user can detour from their current recipe to find steps with alternate ingredients, tools, and techniques. Validating on a ground truth annotated dataset of 16K samples, we show our model's significant improvements over best available methods for video retrieval and question answering, with recall rates exceeding the state of the art by 35%.
Authors: Xingxing Zuo, Pouya Samangouei, Yunwen Zhou, Yan Di, Mingyang Li
Abstract: Precisely perceiving the geometric and semantic properties of real-world 3D objects is crucial for the continued evolution of augmented reality and robotic applications. To this end, we present Foundation Model Embedded Gaussian Splatting (FMGS), which incorporates vision-language embeddings of foundation models into 3D Gaussian Splatting (GS). The key contribution of this work is an efficient method to reconstruct and represent 3D vision-language models. This is achieved by distilling feature maps generated from image-based foundation models into those rendered from our 3D model. To ensure high-quality rendering and fast training, we introduce a novel scene representation by integrating strengths from both GS and multi-resolution hash encodings (MHE). Our effective training procedure also introduces a pixel alignment loss that makes the rendered feature distance of the same semantic entities close, following the pixel-level semantic boundaries. Our results demonstrate remarkable multi-view semantic consistency, facilitating diverse downstream tasks, beating state-of-the-art methods by 10.2 percent on open-vocabulary language-based object detection, despite that we are 851X faster for inference. This research explores the intersection of vision, language, and 3D scene representation, paving the way for enhanced scene understanding in uncontrolled real-world environments. We plan to release the code on the project page.
Authors: Genghao Zhang, Yuxi Wang, Chuanchen Luo, Shibiao Xu, Zhaoxiang Zhang, Man Zhang, Junran Peng
Abstract: Indoor scene generation has attracted significant attention recently as it is crucial for applications of gaming, virtual reality, and interior design. Current indoor scene generation methods can produce reasonable room layouts but often lack diversity and realism. This is primarily due to the limited coverage of existing datasets, including only large furniture without tiny furnishings in daily life. To address these challenges, we propose FurniScene, a large-scale 3D room dataset with intricate furnishing scenes from interior design professionals. Specifically, the FurniScene consists of 11,698 rooms and 39,691 unique furniture CAD models with 89 different types, covering things from large beds to small teacups on the coffee table. To better suit fine-grained indoor scene layout generation, we introduce a novel Two-Stage Diffusion Scene Model (TSDSM) and conduct an evaluation benchmark for various indoor scene generation based on FurniScene. Quantitative and qualitative evaluations demonstrate the capability of our method to generate highly realistic indoor scenes. Our dataset and code will be publicly available soon.
Authors: Xiwei Xuan, Jorge Piazentin Ono, Liang Gou, Kwan-Liu Ma, Liu Ren
Abstract: Data slice finding is an emerging technique for validating machine learning (ML) models by identifying and analyzing subgroups in a dataset that exhibit poor performance, often characterized by distinct feature sets or descriptive metadata. However, in the context of validating vision models involving unstructured image data, this approach faces significant challenges, including the laborious and costly requirement for additional metadata and the complex task of interpreting the root causes of underperformance. To address these challenges, we introduce AttributionScanner, an innovative human-in-the-loop Visual Analytics (VA) system, designed for metadata-free data slice finding. Our system identifies interpretable data slices that involve common model behaviors and visualizes these patterns through an Attribution Mosaic design. Our interactive interface provides straightforward guidance for users to detect, interpret, and annotate predominant model issues, such as spurious correlations (model biases) and mislabeled data, with minimal effort. Additionally, it employs a cutting-edge model regularization technique to mitigate the detected issues and enhance the model's performance. The efficacy of AttributionScanner is demonstrated through use cases involving two benchmark datasets, with qualitative and quantitative evaluations showcasing its substantial effectiveness in vision model validation, ultimately leading to more reliable and accurate models.
Authors: Zongxin Yang, Guikun Chen, Xiaodi Li, Wenguan Wang, Yi Yang
Abstract: Recent LLM-driven visual agents mainly focus on solving image-based tasks, which limits their ability to understand dynamic scenes, making it far from real-life applications like guiding students in laboratory experiments and identifying their mistakes. Hence, this paper explores DoraemonGPT, a comprehensive and conceptually elegant system driven by LLMs to understand dynamic scenes. Considering the video modality better reflects the ever-changing nature of real-world scenarios, we exemplify DoraemonGPT as a video agent. Given a video with a question/task, DoraemonGPT begins by converting the input video into a symbolic memory that stores task-related attributes. This structured representation allows for spatial-temporal querying and reasoning by well-designed sub-task tools, resulting in concise intermediate results. Recognizing that LLMs have limited internal knowledge when it comes to specialized domains (e.g., analyzing the scientific principles underlying experiments), we incorporate plug-and-play tools to assess external knowledge and address tasks across different domains. Moreover, a novel LLM-driven planner based on Monte Carlo Tree Search is introduced to explore the large planning space for scheduling various tools. The planner iteratively finds feasible solutions by backpropagating the result's reward, and multiple solutions can be summarized into an improved final answer. We extensively evaluate DoraemonGPT's effectiveness on three benchmarks and several in-the-wild scenarios. The code will be released at https://github.com/z-x-yang/DoraemonGPT.
Authors: Tianyi Zhao, Maoxun Yuan, Feng Jiang, Nan Wang, Xingxing Wei
Abstract: Object detection in visible (RGB) and infrared (IR) images has been widely applied in recent years. Leveraging the complementary characteristics of RGB and IR images, the object detector provides reliable and robust object localization from day to night. Most existing fusion strategies directly input RGB and IR images into deep neural networks, leading to inferior detection performance. However, the RGB and IR features have modality-specific noise, these strategies will exacerbate the fused features along with the propagation. Inspired by the mechanism of the human brain processing multimodal information, in this paper, we introduce a new coarse-to-fine perspective to purify and fuse two modality features. Specifically, following this perspective, we design a Redundant Spectrum Removal module to coarsely remove interfering information within each modality and a Dynamic Feature Selection module to finely select the desired features for feature fusion. To verify the effectiveness of the coarse-to-fine fusion strategy, we construct a new object detector called the Removal and Selection Detector (RSDet). Extensive experiments on three RGB-IR object detection datasets verify the superior performance of our method.
Authors: Ling Yang, Zhaochen Yu, Chenlin Meng, Minkai Xu, Stefano Ermon, Bin Cui
Abstract: Diffusion models have exhibit exceptional performance in text-to-image generation and editing. However, existing methods often face challenges when handling complex text prompts that involve multiple objects with multiple attributes and relationships. In this paper, we propose a brand new training-free text-to-image generation/editing framework, namely Recaption, Plan and Generate (RPG), harnessing the powerful chain-of-thought reasoning ability of multimodal LLMs to enhance the compositionality of text-to-image diffusion models. Our approach employs the MLLM as a global planner to decompose the process of generating complex images into multiple simpler generation tasks within subregions. We propose complementary regional diffusion to enable region-wise compositional generation. Furthermore, we integrate text-guided image generation and editing within the proposed RPG in a closed-loop fashion, thereby enhancing generalization ability. Extensive experiments demonstrate our RPG outperforms state-of-the-art text-to-image diffusion models, including DALL-E 3 and SDXL, particularly in multi-category object composition and text-image semantic alignment. Notably, our RPG framework exhibits wide compatibility with various MLLM architectures (e.g., MiniGPT-4) and diffusion backbones (e.g., ControlNet). Our code is available at: https://github.com/YangLing0818/RPG-DiffusionMaster
Authors: Juan Hu, Xin Liao, Difei Gao, Satoshi Tsutsui, Qian Wang, Zheng Qin, Mike Zheng Shou
Abstract: Deepfake videos are becoming increasingly realistic, showing few tampering traces on facial areasthat vary between frames. Consequently, existing Deepfake detection methods struggle to detect unknown domain Deepfake videos while accurately locating the tampered region. To address thislimitation, we propose Delocate, a novel Deepfake detection model that can both recognize andlocalize unknown domain Deepfake videos. Ourmethod consists of two stages named recoveringand localization. In the recovering stage, the modelrandomly masks regions of interest (ROIs) and reconstructs real faces without tampering traces, leading to a relatively good recovery effect for realfaces and a poor recovery effect for fake faces. Inthe localization stage, the output of the recoveryphase and the forgery ground truth mask serve assupervision to guide the forgery localization process. This process strategically emphasizes the recovery phase of fake faces with poor recovery, facilitating the localization of tampered regions. Ourextensive experiments on four widely used benchmark datasets demonstrate that Delocate not onlyexcels in localizing tampered areas but also enhances cross-domain detection performance.
Authors: Nachuan Ma, Rui Fan, Lihua Xie
Abstract: Over the past decade, automated methods have been developed to detect cracks more efficiently, accurately, and objectively, with the ultimate goal of replacing conventional manual visual inspection techniques. Among these methods, semantic segmentation algorithms have demonstrated promising results in pixel-wise crack detection tasks. However, training such networks requires a large amount of human-annotated datasets with pixel-level annotations, which is a highly labor-intensive and time-consuming process. Moreover, supervised learning-based methods often struggle with poor generalizability in unseen datasets. Therefore, we propose an unsupervised pixel-wise road crack detection network, known as UP-CrackNet. Our approach first generates multi-scale square masks and randomly selects them to corrupt undamaged road images by removing certain regions. Subsequently, a generative adversarial network is trained to restore the corrupted regions by leveraging the semantic context learned from surrounding uncorrupted regions. During the testing phase, an error map is generated by calculating the difference between the input and restored images, which allows for pixel-wise crack detection. Our comprehensive experimental results demonstrate that UP-CrackNet outperforms other general-purpose unsupervised anomaly detection algorithms, and exhibits satisfactory performance and superior generalizability when compared with state-of-the-art supervised crack segmentation algorithms. Our source code is publicly available at mias.group/UP-CrackNet.
Authors: Hanchao Liu, Wenyuan Xue, Yifei Chen, Dapeng Chen, Xiutian Zhao, Ke Wang, Liping Hou, Rongjun Li, Wei Peng
Abstract: Recent development of Large Vision-Language Models (LVLMs) has attracted growing attention within the AI landscape for its practical implementation potential. However, ``hallucination'', or more specifically, the misalignment between factual visual content and corresponding textual generation, poses a significant challenge of utilizing LVLMs. In this comprehensive survey, we dissect LVLM-related hallucinations in an attempt to establish an overview and facilitate future mitigation. Our scrutiny starts with a clarification of the concept of hallucinations in LVLMs, presenting a variety of hallucination symptoms and highlighting the unique challenges inherent in LVLM hallucinations. Subsequently, we outline the benchmarks and methodologies tailored specifically for evaluating hallucinations unique to LVLMs. Additionally, we delve into an investigation of the root causes of these hallucinations, encompassing insights from the training data and model components. We also critically review existing methods for mitigating hallucinations. The open questions and future directions pertaining to hallucinations within LVLMs are discussed to conclude this survey.
Authors: Shiyuan Yang, Liang Hou, Haibin Huang, Chongyang Ma, Pengfei Wan, Di Zhang, Xiaodong Chen, Jing Liao
Abstract: Recent text-to-video diffusion models have achieved impressive progress. In practice, users often desire the ability to control object motion and camera movement independently for customized video creation. However, current methods lack the focus on separately controlling object motion and camera movement in a decoupled manner, which limits the controllability and flexibility of text-to-video models. In this paper, we introduce Direct-a-Video, a system that allows users to independently specify motions for multiple objects as well as camera's pan and zoom movements, as if directing a video. We propose a simple yet effective strategy for the decoupled control of object motion and camera movement. Object motion is controlled through spatial cross-attention modulation using the model's inherent priors, requiring no additional optimization. For camera movement, we introduce new temporal cross-attention layers to interpret quantitative camera movement parameters. We further employ an augmentation-based approach to train these layers in a self-supervised manner on a small-scale dataset, eliminating the need for explicit motion annotation. Both components operate independently, allowing individual or combined control, and can generalize to open-domain scenarios. Extensive experiments demonstrate the superiority and effectiveness of our method. Project page and code are available at https://direct-a-video.github.io/.
Authors: Simon Ging, Mar\'ia A. Bravo, Thomas Brox
Abstract: The evaluation of text-generative vision-language models is a challenging yet crucial endeavor. By addressing the limitations of existing Visual Question Answering (VQA) benchmarks and proposing innovative evaluation methodologies, our research seeks to advance our understanding of these models' capabilities. We propose a novel VQA benchmark based on well-known visual classification datasets which allows a granular evaluation of text-generative vision-language models and their comparison with discriminative vision-language models. To improve the assessment of coarse answers on fine-grained classification tasks, we suggest using the semantic hierarchy of the label space to ask automatically generated follow-up questions about the ground-truth category. Finally, we compare traditional NLP and LLM-based metrics for the problem of evaluating model predictions given ground-truth answers. We perform a human evaluation study upon which we base our decision on the final metric. We apply our benchmark to a suite of vision-language models and show a detailed comparison of their abilities on object, action, and attribute classification. Our contributions aim to lay the foundation for more precise and meaningful assessments, facilitating targeted progress in the exciting field of vision-language modeling.
Authors: Pierre Marza, Laetitia Matignon, Olivier Simonin, Christian Wolf
Abstract: Successfully addressing a wide variety of tasks is a core ability of autonomous agents, requiring flexibly adapting the underlying decision-making strategies and, as we argue in this work, also adapting the perception modules. An analogical argument would be the human visual system, which uses top-down signals to focus attention determined by the current task. Similarly, we adapt pre-trained large vision models conditioned on specific downstream tasks in the context of multi-task policy learning. We introduce task-conditioned adapters that do not require finetuning any pre-trained weights, combined with a single policy trained with behavior cloning and capable of addressing multiple tasks. We condition the visual adapters on task embeddings, which can be selected at inference if the task is known, or alternatively inferred from a set of example demonstrations. To this end, we propose a new optimization-based estimator. We evaluate the method on a wide variety of tasks from the CortexBench benchmark and show that, compared to existing work, it can be addressed with a single policy. In particular, we demonstrate that adapting visual features is a key design choice and that the method generalizes to unseen tasks given a few demonstrations.
Authors: Junzhe Zhang, Huixuan Zhang, Xunjian Yin, Xiaojun Wan
Abstract: News image captioning requires model to generate an informative caption rich in entities, with the news image and the associated news article. Though Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in addressing various vision-language tasks, our research finds that current MLLMs still bear limitations in handling entity information on news image captioning task. Besides, while MLLMs have the ability to process long inputs, generating high-quality news image captions still requires a trade-off between sufficiency and conciseness of textual input information. To explore the potential of MLLMs and address problems we discovered, we propose : an Entity-Aware Multimodal Alignment based approach for news image captioning. Our approach first aligns the MLLM through Balance Training Strategy with two extra alignment tasks: Entity-Aware Sentence Selection task and Entity Selection task, together with News Image Captioning task, to enhance its capability in handling multimodal entity information. The aligned MLLM will utilizes the additional entity-related information it explicitly extracts to supplement its textual input while generating news image captions. Our approach achieves better results than all previous models in CIDEr score on GoodNews dataset (72.33 -> 88.39) and NYTimes800k dataset (70.83 -> 85.61).
Authors: Niccol\`o Cavagnero, Gabriele Rosi, Claudia Cuttano, Francesca Pistilli, Marco Ciccone, Giuseppe Averta, Fabio Cermelli
Abstract: Recent transformer-based architectures have shown impressive results in the field of image segmentation. Thanks to their flexibility, they obtain outstanding performance in multiple segmentation tasks, such as semantic and panoptic, under a single unified framework. To achieve such impressive performance, these architectures employ intensive operations and require substantial computational resources, which are often not available, especially on edge devices. To fill this gap, we propose Prototype-based Efficient MaskFormer (PEM), an efficient transformer-based architecture that can operate in multiple segmentation tasks. PEM proposes a novel prototype-based cross-attention which leverages the redundancy of visual features to restrict the computation and improve the efficiency without harming the performance. In addition, PEM introduces an efficient multi-scale feature pyramid network, capable of extracting features that have high semantic content in an efficient way, thanks to the combination of deformable convolutions and context-based self-modulation. We benchmark the proposed PEM architecture on two tasks, semantic and panoptic segmentation, evaluated on two different datasets, Cityscapes and ADE20K. PEM demonstrates outstanding performance on every task and dataset, outperforming task-specific architectures while being comparable and even better than computationally-expensive baselines.
Authors: Tingke Shen, Surabhi S Nath, Aenne Brielmann, Peter Dayan
Abstract: The complexity of visual stimuli plays an important role in many cognitive phenomena, including attention, engagement, memorability, time perception and aesthetic evaluation. Despite its importance, complexity is poorly understood and ironically, previous models of image complexity have been quite complex. There have been many attempts to find handcrafted features that explain complexity, but these features are usually dataset specific, and hence fail to generalise. On the other hand, more recent work has employed deep neural networks to predict complexity, but these models remain difficult to interpret, and do not guide a theoretical understanding of the problem. Here we propose to model complexity using segment-based representations of images. We use state-of-the-art segmentation models, SAM and FC-CLIP, to quantify the number of segments at multiple granularities, and the number of classes in an image respectively. We find that complexity is well-explained by a simple linear model with these two features across six diverse image-sets of naturalistic scene and art images. This suggests that the complexity of images can be surprisingly simple.
Authors: Yao Jiang, Xinyu Yan, Ge-Peng Ji, Keren Fu, Meijun Sun, Huan Xiong, Deng-Ping Fan, Fahad Shahbaz Khan
Abstract: The advent of large vision-language models (LVLMs) represents a noteworthy advancement towards the pursuit of artificial general intelligence. However, the model efficacy across both specialized and general tasks warrants further investigation. This paper endeavors to evaluate the competency of popular LVLMs in specialized and general tasks, respectively, aiming to offer a comprehensive understanding of these novel models. To gauge their efficacy in specialized tasks, we employ six challenging tasks across three distinct application scenarios, namely natural, healthcare, and industrial ones. Such six tasks include salient/camouflaged/transparent object detection, as well as polyp detection, skin lesion detection, and industrial anomaly detection. We examine the performance of three recent open-source LVLMs, including MiniGPT-v2, LLaVA-1.5, and Shikra, on both visual recognition and localization under these tasks. Moreover, we conduct empirical investigations utilizing the aforementioned LVLMs together with GPT-4V, assessing their multi-modal understanding capabilities in general tasks including object counting, absurd question answering, affordance reasoning, attribute recognition, and spatial relation reasoning. Our investigations reveal that these LVLMs demonstrate limited proficiency not only in specialized tasks but also in general tasks. We delve deep into this inadequacy and uncover several potential factors, including limited cognition in specialized tasks, object hallucination, text-to-image interference, and decreased robustness in complex problems. We hope this study could provide useful insights for the future development of LVLMs, helping researchers improve LVLMs to cope with both general and specialized applications.
Authors: M Rakesh Reddy, Shubham Mandloi, Aman Kumar
Abstract: Moire pattern frequently appears in photographs captured with mobile devices and digital cameras, potentially degrading image quality. Despite recent advancements in computer vision, image demoire'ing remains a challenging task due to the dynamic textures and variations in colour, shape, and frequency of moire patterns. Most existing methods struggle to generalize to unseen datasets, limiting their effectiveness in removing moire patterns from real-world scenarios. In this paper, we propose a novel lightweight architecture, AADNet (Attention Aware Demoireing Network), for high-resolution image demoire'ing that effectively works across different frequency bands and generalizes well to unseen datasets. Extensive experiments conducted on the UHDM dataset validate the effectiveness of our approach, resulting in high-fidelity images.
Authors: Raza Yunus, Jan Eric Lenssen, Michael Niemeyer, Yiyi Liao, Christian Rupprecht, Christian Theobalt, Gerard Pons-Moll, Jia-Bin Huang, Vladislav Golyanik, Eddy Ilg
Abstract: Reconstructing models of the real world, including 3D geometry, appearance, and motion of real scenes, is essential for computer graphics and computer vision. It enables the synthesizing of photorealistic novel views, useful for the movie industry and AR/VR applications. It also facilitates the content creation necessary in computer games and AR/VR by avoiding laborious manual design processes. Further, such models are fundamental for intelligent computing systems that need to interpret real-world scenes and actions to act and interact safely with the human world. Notably, the world surrounding us is dynamic, and reconstructing models of dynamic, non-rigidly moving scenes is a severely underconstrained and challenging problem. This state-of-the-art report (STAR) offers the reader a comprehensive summary of state-of-the-art techniques with monocular and multi-view inputs such as data from RGB and RGB-D sensors, among others, conveying an understanding of different approaches, their potential applications, and promising further research directions. The report covers 3D reconstruction of general non-rigid scenes and further addresses the techniques for scene decomposition, editing and controlling, and generalizable and generative modeling. More specifically, we first review the common and fundamental concepts necessary to understand and navigate the field and then discuss the state-of-the-art techniques by reviewing recent approaches that use traditional and machine-learning-based neural representations, including a discussion on the newly enabled applications. The STAR is concluded with a discussion of the remaining limitations and open challenges.
Authors: Minchan Kim, Minyeong Kim, Junik Bae, Suhwan Choi, Sungkyung Kim, Buru Chang
Abstract: Hallucinations in vision-language models pose a significant challenge to their reliability, particularly in the generation of long captions. Current methods fall short of accurately identifying and mitigating these hallucinations. To address this issue, we introduce ESREAL, a novel unsupervised learning framework designed to suppress the generation of hallucinations through accurate localization and penalization of hallucinated tokens. Initially, ESREAL creates a reconstructed image based on the generated caption and aligns its corresponding regions with those of the original image. This semantic reconstruction aids in identifying both the presence and type of token-level hallucinations within the generated caption. Subsequently, ESREAL computes token-level hallucination scores by assessing the semantic similarity of aligned regions based on the type of hallucination. Finally, ESREAL employs a proximal policy optimization algorithm, where it selectively penalizes hallucinated tokens according to their token-level hallucination scores. Our framework notably reduces hallucinations in LLaVA, InstructBLIP, and mPLUG-Owl2 by 32.81%, 27.08%, and 7.46% on the CHAIR metric. This improvement is achieved solely through signals derived from the image itself, without the need for any image-text pairs.
Authors: Huei-Chung Hu, Xuyang Wu, Yuan Wang, Yi Fang, Hsin-Tai Wu
Abstract: Numerous works concerning head pose estimation (HPE) offer algorithms or proposed neural network-based approaches for extracting Euler angles from either facial key points or directly from images of the head region. However, many works failed to provide clear definitions of the coordinate systems and Euler or Tait-Bryan angles orders in use. It is a well-known fact that rotation matrices depend on coordinate systems, and yaw, roll, and pitch angles are sensitive to their application order. Without precise definitions, it becomes challenging to validate the correctness of the output head pose and drawing routines employed in prior works. In this paper, we thoroughly examined the Euler angles defined in the 300W-LP dataset, head pose estimation such as 3DDFA-v2, 6D-RepNet, WHENet, etc, and the validity of their drawing routines of the Euler angles. When necessary, we infer their coordinate system and sequence of yaw, roll, pitch from provided code. This paper presents (1) code and algorithms for inferring coordinate system from provided source code, code for Euler angle application order and extracting precise rotation matrices and the Euler angles, (2) code and algorithms for converting poses from one rotation system to another, (3) novel formulae for 2D augmentations of the rotation matrices, and (4) derivations and code for the correct drawing routines for rotation matrices and poses. This paper also addresses the feasibility of defining rotations with right-handed coordinate system in Wikipedia and SciPy, which makes the Euler angle extraction much easier for full-range head pose research.
Authors: Jianfeng Cai, Yue Ma, Zhixi Feng, Shuyuan Yang
Abstract: Polarimetric synthetic aperture radar (PolSAR) image interpretation is widely used in various fields. Recently, deep learning has made significant progress in PolSAR image classification. Supervised learning (SL) requires a large amount of labeled PolSAR data with high quality to achieve better performance, however, manually labeled data is insufficient. This causes the SL to fail into overfitting and degrades its generalization performance. Furthermore, the scattering confusion problem is also a significant challenge that attracts more attention. To solve these problems, this article proposes a Heterogeneous Network based Contrastive Learning method(HCLNet). It aims to learn high-level representation from unlabeled PolSAR data for few-shot classification according to multi-features and superpixels. Beyond the conventional CL, HCLNet introduces the heterogeneous architecture for the first time to utilize heterogeneous PolSAR features better. And it develops two easy-to-use plugins to narrow the domain gap between optics and PolSAR, including feature filter and superpixel-based instance discrimination, which the former is used to enhance the complementarity of multi-features, and the latter is used to increase the diversity of negative samples. Experiments demonstrate the superiority of HCLNet on three widely used PolSAR benchmark datasets compared with state-of-the-art methods. Ablation studies also verify the importance of each component. Besides, this work has implications for how to efficiently utilize the multi-features of PolSAR data to learn better high-level representation in CL and how to construct networks suitable for PolSAR data better.
Authors: Bo Wang, Jian Li, Yang Yu, Li Liu, Zhenping Sun, Dewen Hu
Abstract: Considering the complementarity of scene flow estimation in the spatial domain's focusing capability and 3D object tracking in the temporal domain's coherence, this study aims to address a comprehensive new task that can simultaneously capture fine-grained and long-term 3D motion in an online manner: long-term scene flow estimation (LSFE). We introduce SceneTracker, a novel learning-based LSFE network that adopts an iterative approach to approximate the optimal trajectory. Besides, it dynamically indexes and constructs appearance and depth correlation features simultaneously and employs the Transformer to explore and utilize long-range connections within and between trajectories. With detailed experiments, SceneTracker shows superior capabilities in handling 3D spatial occlusion and depth noise interference, highly tailored to the LSFE task's needs. Finally, we build the first real-world evaluation dataset, LSFDriving, further substantiating SceneTracker's commendable generalization capacity. The code and data for SceneTracker is available at https://github.com/wwsource/SceneTracker.
Authors: Dehao Yuan, Cornelia Ferm\"uller, Tahseen Rabbani, Furong Huang, Yiannis Aloimonos
Abstract: We propose VecKM, a local point cloud geometry encoder that is descriptive and efficient to compute. VecKM leverages a unique approach by vectorizing a kernel mixture to represent the local point cloud. Such representation's descriptiveness is supported by two theorems that validate its ability to reconstruct and preserve the similarity of the local shape. Unlike existing encoders downsampling the local point cloud, VecKM constructs the local geometry encoding using all neighboring points, producing a more descriptive encoding. Moreover, VecKM is efficient to compute and scalable to large point cloud inputs: VecKM reduces the memory cost from $(n^2+nKd)$ to $(nd+np)$; and reduces the major runtime cost from computing $nK$ MLPs to $n$ MLPs, where $n$ is the size of the point cloud, $K$ is the neighborhood size, $d$ is the encoding dimension, and $p$ is a marginal factor. The efficiency is due to VecKM's unique factorizable property that eliminates the need of explicitly grouping points into neighbors. In the normal estimation task, VecKM demonstrates not only 100x faster inference speed but also highest accuracy and strongest robustness. In classification and segmentation tasks, integrating VecKM as a preprocessing module achieves consistently better performance than the PointNet, PointNet++, and point transformer baselines, and runs consistently faster by up to 10 times.
Authors: Beibei Wang, Shuang Meng, Lu Zhang, Chenjie Wang, Jingjing Huang, Yao Li, Haojie Ren, Yuxuan Xiao, Yuru Peng, Jianmin Ji, Yu Zhang, Yanyong Zhang
Abstract: Numerous roadside perception datasets have been introduced to propel advancements in autonomous driving and intelligent transportation systems research and development. However, it has been observed that the majority of their concentrates is on urban arterial roads, inadvertently overlooking residential areas such as parks and campuses that exhibit entirely distinct characteristics. In light of this gap, we propose CORP, which stands as the first public benchmark dataset tailored for multi-modal roadside perception tasks under campus scenarios. Collected in a university campus, CORP consists of over 205k images plus 102k point clouds captured from 18 cameras and 9 LiDAR sensors. These sensors with different configurations are mounted on roadside utility poles to provide diverse viewpoints within the campus region. The annotations of CORP encompass multi-dimensional information beyond 2D and 3D bounding boxes, providing extra support for 3D seamless tracking and instance segmentation with unique IDs and pixel masks for identifying targets, to enhance the understanding of objects and their behaviors distributed across the campus premises. Unlike other roadside datasets about urban traffic, CORP extends the spectrum to highlight the challenges for multi-modal perception in campuses and other residential areas.
Authors: Jiong Wu, Shuang Zhou, Li Lin, Xin Wang, Wenxue Tan
Abstract: Diffeomorphic image registration is a fundamental step in medical image analysis, owing to its capability to ensure the invertibility of transformations and preservation of topology. Currently, unsupervised learning-based registration techniques primarily extract features at the image level, potentially limiting their efficacy. This paper proposes a novel unsupervised learning-based fully convolutional network (FCN) framework for fast diffeomorphic image registration, emphasizing feature acquisition at the image patch level. Furthermore, a novel differential operator is introduced and integrated into the FCN architecture for parameter learning. Experiments are conducted on three distinct T1-weighted magnetic resonance imaging (T1w MRI) datasets. Comparative analyses with three state-of-the-art diffeomorphic image registration approaches including a typical conventional registration algorithm and two representative unsupervised learning-based methods, reveal that the proposed method exhibits superior performance in both registration accuracy and topology preservation.
Authors: Reuben Tan, Ximeng Sun, Ping Hu, Jui-hsien Wang, Hanieh Deilamsalehy, Bryan A. Plummer, Bryan Russell, Kate Saenko
Abstract: Long video question answering is a challenging task that involves recognizing short-term activities and reasoning about their fine-grained relationships. State-of-the-art video Large Language Models (vLLMs) hold promise as a viable solution due to their demonstrated emergent capabilities on new tasks. However, despite being trained on millions of short seconds-long videos, vLLMs are unable to understand minutes-long videos and accurately answer questions about them. To address this limitation, we propose a lightweight and self-supervised approach, Key frame-conditioned long video-LLM (Koala), that introduces learnable spatiotemporal queries to adapt pretrained vLLMs for generalizing to longer videos. Our approach introduces two new tokenizers that condition on visual tokens computed from sparse video key frames for understanding short and long video moments. We train our proposed approach on HowTo100M and demonstrate its effectiveness on zero-shot long video understanding benchmarks, where it outperforms state-of-the-art large models by 3 - 6% in absolute accuracy across all tasks. Surprisingly, we also empirically show that our approach not only helps a pretrained vLLM to understand long videos but also improves its accuracy on short-term action recognition.
Authors: Ziang Guo, Stepan Perminov, Mikhail Konenkov, Dzmitry Tsetserukou
Abstract: Many established vision perception systems for autonomous driving scenarios ignore the influence of light conditions, one of the key elements for driving safety. To address this problem, we present HawkDrive, a novel perception system with hardware and software solutions. Hardware that utilizes stereo vision perception, which has been demonstrated to be a more reliable way of estimating depth information than monocular vision, is partnered with the edge computing device Nvidia Jetson Xavier AGX. Our software for low light enhancement, depth estimation, and semantic segmentation tasks, is a transformer-based neural network. Our software stack, which enables fast inference and noise reduction, is packaged into system modules in Robot Operating System 2 (ROS2). Our experimental results have shown that the proposed end-to-end system is effective in improving the depth estimation and semantic segmentation performance. Our dataset and codes will be released at https://github.com/ZionGo6/HawkDrive.
Authors: Zijin Yang, Kai Zeng, Kejiang Chen, Han Fang, Weiming Zhang, Nenghai Yu
Abstract: Ethical concerns surrounding copyright protection and inappropriate content generation pose challenges for the practical implementation of diffusion models. One effective solution involves watermarking the generated images. However, existing methods often compromise the model performance or require additional training, which is undesirable for operators and users. To address this issue, we propose Gaussian Shading, a diffusion model watermarking technique that is both performance-lossless and training-free, while serving the dual purpose of copyright protection and tracing of offending content. Our watermark embedding is free of model parameter modifications and thus is plug-and-play. We map the watermark to latent representations following a standard Gaussian distribution, which is indistinguishable from latent representations obtained from the non-watermarked diffusion model. Therefore we can achieve watermark embedding with lossless performance, for which we also provide theoretical proof. Furthermore, since the watermark is intricately linked with image semantics, it exhibits resilience to lossy processing and erasure attempts. The watermark can be extracted by Denoising Diffusion Implicit Models (DDIM) inversion and inverse sampling. We evaluate Gaussian Shading on multiple versions of Stable Diffusion, and the results demonstrate that Gaussian Shading not only is performance-lossless but also outperforms existing methods in terms of robustness.
Authors: Jing Gu, Yilin Wang, Nanxuan Zhao, Wei Xiong, Qing Liu, Zhifei Zhang, He Zhang, Jianming Zhang, HyunJoon Jung, Xin Eric Wang
Abstract: Effective editing of personal content holds a pivotal role in enabling individuals to express their creativity, weaving captivating narratives within their visual stories, and elevate the overall quality and impact of their visual content. Therefore, in this work, we introduce SwapAnything, a novel framework that can swap any objects in an image with personalized concepts given by the reference, while keeping the context unchanged. Compared with existing methods for personalized subject swapping, SwapAnything has three unique advantages: (1) precise control of arbitrary objects and parts rather than the main subject, (2) more faithful preservation of context pixels, (3) better adaptation of the personalized concept to the image. First, we propose targeted variable swapping to apply region control over latent feature maps and swap masked variables for faithful context preservation and initial semantic concept swapping. Then, we introduce appearance adaptation, to seamlessly adapt the semantic concept into the original image in terms of target location, shape, style, and content during the image generation process. Extensive results on both human and automatic evaluation demonstrate significant improvements of our approach over baseline methods on personalized swapping. Furthermore, SwapAnything shows its precise and faithful swapping abilities across single object, multiple objects, partial object, and cross-domain swapping tasks. SwapAnything also achieves great performance on text-based swapping and tasks beyond swapping such as object insertion.
Authors: Hongli Wen, Yang Xu, Lin Li, Xudong Ru, Xingce Wang, Zhongke Wu
Abstract: Traditional sign language teaching methods face challenges such as limited feedback and diverse learning scenarios. Although 2D resources lack real-time feedback, classroom teaching is constrained by a scarcity of teacher. Methods based on VR and AR have relatively primitive interaction feedback mechanisms. This study proposes an innovative teaching model that uses real-time monocular vision and mixed reality technology. First, we introduce an improved hand-posture reconstruction method to achieve sign language semantic retention and real-time feedback. Second, a ternary system evaluation algorithm is proposed for a comprehensive assessment, maintaining good consistency with experts in sign language. Furthermore, we use mixed reality technology to construct a scenario-based 3D sign language classroom and explore the user experience of scenario teaching. Overall, this paper presents a novel teaching method that provides an immersive learning experience, advanced posture reconstruction, and precise feedback, achieving positive feedback on user experience and learning effectiveness.
Authors: Ying Zhang, Yuezun Li, Bo Peng, Jiaran Zhou, Huiyu Zhou, Junyu Dong
Abstract: The task of video inpainting detection is to expose the pixel-level inpainted regions within a video sequence. Existing methods usually focus on leveraging spatial and temporal inconsistencies. However, these methods typically employ fixed operations to combine spatial and temporal clues, limiting their applicability in different scenarios. In this paper, we introduce a novel Multilateral Temporal-view Pyramid Transformer ({\em MumPy}) that collaborates spatial-temporal clues flexibly. Our method utilizes a newly designed multilateral temporal-view encoder to extract various collaborations of spatial-temporal clues and introduces a deformable window-based temporal-view interaction module to enhance the diversity of these collaborations. Subsequently, we develop a multi-pyramid decoder to aggregate the various types of features and generate detection maps. By adjusting the contribution strength of spatial and temporal clues, our method can effectively identify inpainted regions. We validate our method on existing datasets and also introduce a new challenging and large-scale Video Inpainting dataset based on the YouTube-VOS dataset, which employs several more recent inpainting methods. The results demonstrate the superiority of our method in both in-domain and cross-domain evaluation scenarios.
Authors: Kuan-Chieh Wang, Daniil Ostashev, Yuwei Fang, Sergey Tulyakov, Kfir Aberman
Abstract: We introduce a new architecture for personalization of text-to-image diffusion models, coined Mixture-of-Attention (MoA). Inspired by the Mixture-of-Experts mechanism utilized in large language models (LLMs), MoA distributes the generation workload between two attention pathways: a personalized branch and a non-personalized prior branch. MoA is designed to retain the original model's prior by fixing its attention layers in the prior branch, while minimally intervening in the generation process with the personalized branch that learns to embed subjects in the layout and context generated by the prior branch. A novel routing mechanism manages the distribution of pixels in each layer across these branches to optimize the blend of personalized and generic content creation. Once trained, MoA facilitates the creation of high-quality, personalized images featuring multiple subjects with compositions and interactions as diverse as those generated by the original model. Crucially, MoA enhances the distinction between the model's pre-existing capability and the newly augmented personalized intervention, thereby offering a more disentangled subject-context control that was previously unattainable. Project page: https://snap-research.github.io/mixture-of-attention
Authors: Xingyu Fu, Yushi Hu, Bangzheng Li, Yu Feng, Haoyu Wang, Xudong Lin, Dan Roth, Noah A. Smith, Wei-Chiu Ma, Ranjay Krishna
Abstract: We introduce Blink, a new benchmark for multimodal language models (LLMs) that focuses on core visual perception abilities not found in other evaluations. Most of the Blink tasks can be solved by humans "within a blink" (e.g., relative depth estimation, visual correspondence, forensics detection, and multi-view reasoning). However, we find these perception-demanding tasks cast significant challenges for current multimodal LLMs because they resist mediation through natural language. Blink reformats 14 classic computer vision tasks into 3,807 multiple-choice questions, paired with single or multiple images and visual prompting. While humans get 95.70% accuracy on average, Blink is surprisingly challenging for existing multimodal LLMs: even the best-performing GPT-4V and Gemini achieve accuracies of 51.26% and 45.72%, only 13.17% and 7.63% higher than random guessing, indicating that such perception abilities have not "emerged" yet in recent multimodal LLMs. Our analysis also highlights that specialist CV models could solve these problems much better, suggesting potential pathways for future improvements. We believe Blink will stimulate the community to help multimodal LLMs catch up with human-level visual perception.
Authors: Jihao Dong, Renjie Pan, Hua Yang
Abstract: Human-Object Interaction (HOI) detection aims to localize human-object pairs and comprehend their interactions. Recently, two-stage transformer-based methods have demonstrated competitive performance. However, these methods frequently focus on object appearance features and ignore global contextual information. Besides, vision-language model CLIP which effectively aligns visual and text embeddings has shown great potential in zero-shot HOI detection. Based on the former facts, We introduce a novel HOI detector named ISA-HOI, which extensively leverages knowledge from CLIP, aligning interactive semantics between visual and textual features. We first extract global context of image and local features of object to Improve interaction Features in images (IF). On the other hand, we propose a Verb Semantic Improvement (VSI) module to enhance textual features of verb labels via cross-modal fusion. Ultimately, our method achieves competitive results on the HICO-DET and V-COCO benchmarks with much fewer training epochs, and outperforms the state-of-the-art under zero-shot settings.
Authors: Hanzhe Li, Yuezun Li, Jiaran Zhou, Bin Li, Junyu Dong
Abstract: Generating synthetic fake faces, known as pseudo-fake faces, is an effective way to improve the generalization of DeepFake detection. Existing methods typically generate these faces by blending real or fake faces in color space. While these methods have shown promise, they overlook the simulation of frequency distribution in pseudo-fake faces, limiting the learning of generic forgery traces in-depth. To address this, this paper introduces {\em FreqBlender}, a new method that can generate pseudo-fake faces by blending frequency knowledge. Specifically, we investigate the major frequency components and propose a Frequency Parsing Network to adaptively partition frequency components related to forgery traces. Then we blend this frequency knowledge from fake faces into real faces to generate pseudo-fake faces. Since there is no ground truth for frequency components, we describe a dedicated training strategy by leveraging the inner correlations among different frequency knowledge to instruct the learning process. Experimental results demonstrate the effectiveness of our method in enhancing DeepFake detection, making it a potential plug-and-play strategy for other methods.
Authors: Yunfei Li, Yuezun Li, Xin Wang, Jiaran Zhou, Junyu Dong
Abstract: Sequential DeepFake detection is an emerging task that aims to predict the manipulation sequence in order. Existing methods typically formulate it as an image-to-sequence problem, employing conventional Transformer architectures for detection. However, these methods lack dedicated design and consequently result in limited performance. In this paper, we propose a novel Texture-aware and Shape-guided Transformer to enhance detection performance. Our method features four major improvements. Firstly, we describe a texture-aware branch that effectively captures subtle manipulation traces with the Diversiform Pixel Difference Attention module. Then we introduce a Bidirectional Interaction Cross-attention module that seeks deep correlations among spatial and sequential features, enabling effective modeling of complex manipulation traces. To further enhance the cross-attention, we describe a Shape-guided Gaussian mapping strategy, providing initial priors of the manipulation shape. Finally, observing that the latter manipulation in a sequence may influence traces left in the earlier one, we intriguingly invert the prediction order from forward to backward, leading to notable gains as expected. Extensive experimental results demonstrate that our method outperforms others by a large margin, highlighting the superiority of our method.
Authors: Xuzheng Yu, Chen Jiang, Xingning Dong, Tian Gan, Ming Yang, Qingpei Guo
Abstract: The user base of short video apps has experienced unprecedented growth in recent years, resulting in a significant demand for video content analysis. In particular, text-video retrieval, which aims to find the top matching videos given text descriptions from a vast video corpus, is an essential function, the primary challenge of which is to bridge the modality gap. Nevertheless, most existing approaches treat texts merely as discrete tokens and neglect their syntax structures. Moreover, the abundant spatial and temporal clues in videos are often underutilized due to the lack of interaction with text. To address these issues, we argue that using texts as guidance to focus on relevant temporal frames and spatial regions within videos is beneficial. In this paper, we propose a novel Syntax-Hierarchy-Enhanced text-video retrieval method (SHE-Net) that exploits the inherent semantic and syntax hierarchy of texts to bridge the modality gap from two perspectives. First, to facilitate a more fine-grained integration of visual content, we employ the text syntax hierarchy, which reveals the grammatical structure of text descriptions, to guide the visual representations. Second, to further enhance the multi-modal interaction and alignment, we also utilize the syntax hierarchy to guide the similarity calculation. We evaluated our method on four public text-video retrieval datasets of MSR-VTT, MSVD, DiDeMo, and ActivityNet. The experimental results and ablation studies confirm the advantages of our proposed method.
Authors: Chenhui Wang, Tao Chen, Zhihao Chen, Zhizhong Huang, Taoran Jiang, Qi Wang, Hongming Shan
Abstract: Despite their impressive generative performance, latent diffusion model-based virtual try-on (VTON) methods lack faithfulness to crucial details of the clothes, such as style, pattern, and text. To alleviate these issues caused by the diffusion stochastic nature and latent supervision, we propose a novel Faithful Latent Diffusion Model for VTON, termed FLDM-VTON. FLDM-VTON improves the conventional latent diffusion process in three major aspects. First, we propose incorporating warped clothes as both the starting point and local condition, supplying the model with faithful clothes priors. Second, we introduce a novel clothes flattening network to constrain generated try-on images, providing clothes-consistent faithful supervision. Third, we devise a clothes-posterior sampling for faithful inference, further enhancing the model performance over conventional clothes-agnostic Gaussian sampling. Extensive experimental results on the benchmark VITON-HD and Dress Code datasets demonstrate that our FLDM-VTON outperforms state-of-the-art baselines and is able to generate photo-realistic try-on images with faithful clothing details.
Authors: Panagiotis Sapoutzoglou, Georgios Giapitzakis Tzintanos, George Terzakis, Maria Pateraki
Abstract: We present a generic algorithm for scoring pose estimation methods that rely on single image semantic analysis. The algorithm employs a lightweight putative shape representation using a combination of multiple Gaussian Processes. Each Gaussian Process (GP) yields distance normal distributions from multiple reference points in the object's coordinate system to its surface, thus providing a geometric evaluation framework for scoring predicted poses. Our confidence measure comprises the average mixture probability of pixel back-projections onto the shape template. In the reported experiments, we compare the accuracy of our GP based representation of objects versus the actual geometric models and demonstrate the ability of our method to capture the influence of outliers as opposed to the corresponding intrinsic measures that ship with the segmentation and pose estimation methods.
Authors: Tengjun Huang
Abstract: With the rise of Visual and Language Pretraining (VLP), an increasing number of downstream tasks are adopting the paradigm of pretraining followed by fine-tuning. Although this paradigm has demonstrated potential in various multimodal downstream tasks, its implementation in the remote sensing domain encounters some obstacles. Specifically, the tendency for same-modality embeddings to cluster together impedes efficient transfer learning. To tackle this issue, we review the aim of multimodal transfer learning for downstream tasks from a unified perspective, and rethink the optimization process based on three distinct objectives. We propose "Harmonized Transfer Learning and Modality Alignment (HarMA)", a method that simultaneously satisfies task constraints, modality alignment, and single-modality uniform alignment, while minimizing training overhead through parameter-efficient fine-tuning. Remarkably, without the need for external data for training, HarMA achieves state-of-the-art performance in two popular multimodal retrieval tasks in the field of remote sensing. Our experiments reveal that HarMA achieves competitive and even superior performance to fully fine-tuned models with only minimal adjustable parameters. Due to its simplicity, HarMA can be integrated into almost all existing multimodal pretraining models. We hope this method can facilitate the efficient application of large models to a wide range of downstream tasks while significantly reducing the resource consumption. Code is available at https://github.com/seekerhuang/HarMA.
Authors: Paulina Tomaszewska, Przemys{\l}aw Biecek
Abstract: Does the stethoscope in the picture make the adjacent person a doctor or a patient? This, of course, depends on the contextual relationship of the two objects. If it is obvious, why don not explanation methods for vision models use contextual information? In this paper, we (1) review the most popular methods of explaining computer vision models by pointing out that they do not take into account context information, (2) provide examples of real-world use cases where spatial context plays a significant role, (3) propose new research directions that may lead to better use of context information in explaining computer vision models, (4) argue that a change in approach to explanations is needed from 'where' to 'how'.
Authors: Yeon-Ji Song, Suhyung Choi, Jaein Kim, Jin-Hwa Kim, Byoung-Tak Zhang
Abstract: Human perception involves discerning complex multi-object scenes into time-static object appearance (ie, size, shape, color) and time-varying object motion (ie, location, velocity, acceleration). This innate ability to unconsciously understand the environment is the motivation behind the success of dynamics modeling. Object-centric representations have emerged as a promising tool for dynamics prediction, yet they primarily focus on the objects' appearance, often overlooking other crucial attributes. In this paper, we propose Object-Centric Kinematics (OCK), a framework for dynamics prediction leveraging object-centric representations. Our model utilizes a novel component named object kinematics, which comprises low-level structured states of objects' position, velocity, and acceleration. The object kinematics are obtained via either implicit or explicit approaches, enabling comprehensive spatiotemporal object reasoning, and integrated through various transformer mechanisms, facilitating effective object-centric dynamics modeling. Our model demonstrates superior performance when handling objects and backgrounds in complex scenes characterized by a wide range of object attributes and dynamic movements. Moreover, our model demonstrates generalization capabilities across diverse synthetic environments, highlighting its potential for broad applicability in vision-related tasks.
Authors: Ziyan Chen, Jingwen He, Xinqi Lin, Yu Qiao, Chao Dong
Abstract: Blind face restoration (BFR) on images has significantly progressed over the last several years, while real-world video face restoration (VFR), which is more challenging for more complex face motions such as moving gaze directions and facial orientations involved, remains unsolved. Typical BFR methods are evaluated on privately synthesized datasets or self-collected real-world low-quality face images, which are limited in their coverage of real-world video frames. In this work, we introduced new real-world datasets named FOS with a taxonomy of "Full, Occluded, and Side" faces from mainly video frames to study the applicability of current methods on videos. Compared with existing test datasets, FOS datasets cover more diverse degradations and involve face samples from more complex scenarios, which helps to revisit current face restoration approaches more comprehensively. Given the established datasets, we benchmarked both the state-of-the-art BFR methods and the video super resolution (VSR) methods to comprehensively study current approaches, identifying their potential and limitations in VFR tasks. In addition, we studied the effectiveness of the commonly used image quality assessment (IQA) metrics and face IQA (FIQA) metrics by leveraging a subjective user study. With extensive experimental results and detailed analysis provided, we gained insights from the successes and failures of both current BFR and VSR methods. These results also pose challenges to current face restoration approaches, which we hope stimulate future advances in VFR research.
Authors: Yuliang Liu, Mingxin Huang, Hao Yan, Linger Deng, Weijia Wu, Hao Lu, Chunhua Shen, Lianwen Jin, Xiang Bai
Abstract: Text spotting, a task involving the extraction of textual information from image or video sequences, faces challenges in cross-domain adaption, such as image-to-image and image-to-video generalization. In this paper, we introduce a new method, termed VimTS, which enhances the generalization ability of the model by achieving better synergy among different tasks. Typically, we propose a Prompt Queries Generation Module and a Tasks-aware Adapter to effectively convert the original single-task model into a multi-task model suitable for both image and video scenarios with minimal additional parameters. The Prompt Queries Generation Module facilitates explicit interaction between different tasks, while the Tasks-aware Adapter helps the model dynamically learn suitable features for each task. Additionally, to further enable the model to learn temporal information at a lower cost, we propose a synthetic video text dataset (VTD-368k) by leveraging the Content Deformation Fields (CoDeF) algorithm. Notably, our method outperforms the state-of-the-art method by an average of 2.6% in six cross-domain benchmarks such as TT-to-IC15, CTW1500-to-TT, and TT-to-CTW1500. For video-level cross-domain adaption, our method even surpasses the previous end-to-end video spotting method in ICDAR2015 video and DSText v2 by an average of 5.5% on the MOTA metric, using only image-level data. We further demonstrate that existing Large Multimodal Models exhibit limitations in generating cross-domain scene text spotting, in contrast to our VimTS model which requires significantly fewer parameters and data. The code and datasets will be made available at the https://VimTextSpotter.github.io.
Authors: Hang Du, Sicheng Zhang, Binzhu Xie, Guoshun Nan, Jiayang Zhang, Junrui Xu, Hangyu Liu, Sicong Leng, Jiangming Liu, Hehe Fan, Dajiu Huang, Jing Feng, Linli Chen, Can Zhang, Xuhuan Li, Hao Zhang, Jianhang Chen, Qimei Cui, Xiaofeng Tao
Abstract: Video anomaly understanding (VAU) aims to automatically comprehend unusual occurrences in videos, thereby enabling various applications such as traffic surveillance and industrial manufacturing. While existing VAU benchmarks primarily concentrate on anomaly detection and localization, our focus is on more practicality, prompting us to raise the following crucial questions: "what anomaly occurred?", "why did it happen?", and "how severe is this abnormal event?". In pursuit of these answers, we present a comprehensive benchmark for Causation Understanding of Video Anomaly (CUVA). Specifically, each instance of the proposed benchmark involves three sets of human annotations to indicate the "what", "why" and "how" of an anomaly, including 1) anomaly type, start and end times, and event descriptions, 2) natural language explanations for the cause of an anomaly, and 3) free text reflecting the effect of the abnormality. In addition, we also introduce MMEval, a novel evaluation metric designed to better align with human preferences for CUVA, facilitating the measurement of existing LLMs in comprehending the underlying cause and corresponding effect of video anomalies. Finally, we propose a novel prompt-based method that can serve as a baseline approach for the challenging CUVA. We conduct extensive experiments to show the superiority of our evaluation metric and the prompt-based approach. Our code and dataset are available at https://github.com/fesvhtr/CUVA.
Authors: Samuel Lavoie, Polina Kirichenko, Mark Ibrahim, Mahmoud Assran, Andrew Gordon Wildon, Aaron Courville, Nicolas Ballas
Abstract: There are a thousand ways to caption an image. Contrastive Language Pretraining (CLIP) on the other hand, works by mapping an image and its caption to a single vector -- limiting how well CLIP-like models can represent the diverse ways to describe an image. In this work, we introduce Llip, Latent Language Image Pretraining, which models the diversity of captions that could match an image. Llip's vision encoder outputs a set of visual features that are mixed into a final representation by conditioning on information derived from the text. We show that Llip outperforms non-contextualized baselines like CLIP and SigLIP on a variety of tasks even with large-scale encoders. Llip improves zero-shot classification by an average of 2.9% zero-shot classification benchmarks with a ViT-G/14 encoder. Specifically, Llip attains a zero-shot top-1 accuracy of 83.5% on ImageNet outperforming a similarly sized CLIP by 1.4%. We also demonstrate improvement on zero-shot retrieval on MS-COCO by 6.0%. We provide a comprehensive analysis of the components introduced by the method and demonstrate that Llip leads to richer visual representations.
Authors: Shanlin Sun, Bingbing Zhuang, Ziyu Jiang, Buyu Liu, Xiaohui Xie, Manmohan Chandraker
Abstract: Photorealistic simulation plays a crucial role in applications such as autonomous driving, where advances in neural radiance fields (NeRFs) may allow better scalability through the automatic creation of digital 3D assets. However, reconstruction quality suffers on street scenes due to largely collinear camera motions and sparser samplings at higher speeds. On the other hand, the application often demands rendering from camera views that deviate from the inputs to accurately simulate behaviors like lane changes. In this paper, we propose several insights that allow a better utilization of Lidar data to improve NeRF quality on street scenes. First, our framework learns a geometric scene representation from Lidar, which is fused with the implicit grid-based representation for radiance decoding, thereby supplying stronger geometric information offered by explicit point cloud. Second, we put forth a robust occlusion-aware depth supervision scheme, which allows utilizing densified Lidar points by accumulation. Third, we generate augmented training views from Lidar points for further improvement. Our insights translate to largely improved novel view synthesis under real driving scenes.
Authors: Zixun Jiao, Xihan Wang, Quanli Gao
Abstract: Reconstructing a hand mesh from a single RGB image is a challenging task because hands are often occluded by objects. Most previous works attempted to introduce more additional information and adopt attention mechanisms to improve 3D reconstruction results, but it would increased computational complexity. This observation prompts us to propose a new and concise architecture while improving computational efficiency. In this work, we propose a simple and effective 3D hand mesh reconstruction network HandSSCA, which is the first to incorporate state space modeling into the field of hand pose estimation. In the network, we have designed a novel state space channel attention module that extends the effective sensory field, extracts hand features in the spatial dimension, and enhances hand regional features in the channel dimension. This design helps to reconstruct a complete and detailed hand mesh. Extensive experiments conducted on well-known datasets featuring challenging hand-object occlusions (such as FREIHAND, DEXYCB, and HO3D) demonstrate that our proposed HandSSCA achieves state-of-the-art performance while maintaining a minimal parameter count.
Authors: Yicheng Zhan, Liang Shi, Wojciech Matusik, Qi Sun, Kaan Ak\c{s}it
Abstract: In the pursuit of advancing holographic display technology, we face a unique yet persistent roadblock: the inflexibility of learned holography in adapting to various hardware configurations. This is due to the variances in the complex optical components and system settings in existing holographic displays. Although the emerging learned approaches have enabled rapid and high-quality hologram generation, any alteration in display hardware still requires a retraining of the model. Our work introduces a configurable learned model that interactively computes 3D holograms from RGB-only 2D images for a variety of holographic displays. The model can be conditioned to predefined hardware parameters of existing holographic displays such as working wavelengths, pixel pitch, propagation distance, and peak brightness without having to retrain. In addition, our model accommodates various hologram types, including conventional single-color and emerging multi-color holograms that simultaneously use multiple color primaries in holographic displays. Notably, we enabled our hologram computations to rely on identifying the correlation between depth estimation and 3D hologram synthesis tasks within the learning domain for the first time in the literature. We employ knowledge distillation via a student-teacher learning strategy to streamline our model for interactive performance. Achieving up to a 2x speed improvement compared to state-of-the-art models while consistently generating high-quality 3D holograms with different hardware configurations.
Authors: Tushar Verma, Jyotsna Singh, Yash Bhartari, Rishi Jarwal, Suraj Singh, Shubhkarman Singh
Abstract: Small object detection in aerial imagery presents significant challenges in computer vision due to the minimal data inherent in small-sized objects and their propensity to be obscured by larger objects and background noise. Traditional methods using transformer-based models often face limitations stemming from the lack of specialized databases, which adversely affect their performance with objects of varying orientations and scales. This underscores the need for more adaptable, lightweight models. In response, this paper introduces two innovative approaches that significantly enhance detection and segmentation capabilities for small aerial objects. Firstly, we explore the use of the SAHI framework on the newly introduced lightweight YOLO v9 architecture, which utilizes Programmable Gradient Information (PGI) to reduce the substantial information loss typically encountered in sequential feature extraction processes. The paper employs the Vision Mamba model, which incorporates position embeddings to facilitate precise location-aware visual understanding, combined with a novel bidirectional State Space Model (SSM) for effective visual context modeling. This State Space Model adeptly harnesses the linear complexity of CNNs and the global receptive field of Transformers, making it particularly effective in remote sensing image classification. Our experimental results demonstrate substantial improvements in detection accuracy and processing efficiency, validating the applicability of these approaches for real-time small object detection across diverse aerial scenarios. This paper also discusses how these methodologies could serve as foundational models for future advancements in aerial object recognition technologies. The source code will be made accessible here.
Authors: Yadong Li, Dongheng Zhang, Ruixu Geng, Jincheng Wu, Yang Hu, Qibin Sun, Yan Chen
Abstract: Recent advancements have showcased the potential of handheld millimeter-wave (mmWave) imaging, which applies synthetic aperture radar (SAR) principles in portable settings. However, existing studies addressing handheld motion errors either rely on costly tracking devices or employ simplified imaging models, leading to impractical deployment or limited performance. In this paper, we present IFNet, a novel deep unfolding network that combines the strengths of signal processing models and deep neural networks to achieve robust imaging and focusing for handheld mmWave systems. We first formulate the handheld imaging model by integrating multiple priors about mmWave images and handheld phase errors. Furthermore, we transform the optimization processes into an iterative network structure for improved and efficient imaging performance. Extensive experiments demonstrate that IFNet effectively compensates for handheld phase errors and recovers high-fidelity images from severely distorted signals. In comparison with existing methods, IFNet can achieve at least 11.89 dB improvement in average peak signal-to-noise ratio (PSNR) and 64.91% improvement in average structural similarity index measure (SSIM) on a real-world dataset.
Authors: Ming-Wei Li, Qing-Yuan Jiang, Wu-Jun Li
Abstract: Due to its low storage cost and fast query speed, hashing has been widely used in large-scale image retrieval tasks. Hash bucket search returns data points within a given Hamming radius to each query, which can enable search at a constant or sub-linear time cost. However, existing hashing methods cannot achieve satisfactory retrieval performance for hash bucket search in complex scenarios, since they learn only one hash code for each image. More specifically, by using one hash code to represent one image, existing methods might fail to put similar image pairs to the buckets with a small Hamming distance to the query when the semantic information of images is complex. As a result, a large number of hash buckets need to be visited for retrieving similar images, based on the learned codes. This will deteriorate the efficiency of hash bucket search. In this paper, we propose a novel hashing framework, called multiple code hashing (MCH), to improve the performance of hash bucket search. The main idea of MCH is to learn multiple hash codes for each image, with each code representing a different region of the image. Furthermore, we propose a deep reinforcement learning algorithm to learn the parameters in MCH. To the best of our knowledge, this is the first work that proposes to learn multiple hash codes for each image in image retrieval. Experiments demonstrate that MCH can achieve a significant improvement in hash bucket search, compared with existing methods that learn only one hash code for each image.
Authors: Mazhi Qiang, Fengming Zhou
Abstract: Synthetic aperture radar has the ability to work 24/7 and 24/7, and has high application value. Propose a new SAR image matching algorithm based on multi class features, mainly using two different types of features: straight lines and regions to enhance the robustness of the matching algorithm; On the basis of using prior knowledge of images, combined with LSD (Line Segment Detector) line detection and template matching algorithm, by analyzing the attribute correlation between line and surface features in SAR images, selecting line and region features in SAR images to match the images, the matching accuracy between SAR images and visible light images is improved, and the probability of matching errors is reduced. The experimental results have verified that this algorithm can obtain high-precision matching results, achieve precise target positioning, and has good robustness to changes in perspective and lighting. The results are accurate and false positives are controllable.
Authors: Alexia Jolicoeur-Martineau, Emy Gervais, Kilian Fatras, Yan Zhang, Simon Lacoste-Julien
Abstract: Ensemble methods combine the predictions of multiple models to improve performance, but they require significantly higher computation costs at inference time. To avoid these costs, multiple neural networks can be combined into one by averaging their weights. However, this usually performs significantly worse than ensembling. Weight averaging is only beneficial when different enough to benefit from combining them, but similar enough to average well. Based on this idea, we propose PopulAtion Parameter Averaging (PAPA): a method that combines the generality of ensembling with the efficiency of weight averaging. PAPA leverages a population of diverse models (trained on different data orders, augmentations, and regularizations) while slowly pushing the weights of the networks toward the population average of the weights. We also propose PAPA variants (PAPA-all, and PAPA-2) that average weights rarely rather than continuously; all methods increase generalization, but PAPA tends to perform best. PAPA reduces the performance gap between averaging and ensembling, increasing the average accuracy of a population of models by up to 0.8% on CIFAR-10, 1.9% on CIFAR-100, and 1.6% on ImageNet when compared to training independent (non-averaged) models.
Authors: Md. Alamin Talukder, Md. Manowarul Islam, Md Ashraf Uddin
Abstract: Brain tumors present a grave risk to human life, demanding precise and timely diagnosis for effective treatment. Inaccurate identification of brain tumors can significantly diminish life expectancy, underscoring the critical need for precise diagnostic methods. Manual identification of brain tumors within vast Magnetic Resonance Imaging (MRI) image datasets is arduous and time-consuming. Thus, the development of a reliable deep learning (DL) model is essential to enhance diagnostic accuracy and ultimately save lives. This study introduces an innovative optimization-based deep ensemble approach employing transfer learning (TL) to efficiently classify brain tumors. Our methodology includes meticulous preprocessing, reconstruction of TL architectures, fine-tuning, and ensemble DL models utilizing weighted optimization techniques such as Genetic Algorithm-based Weight Optimization (GAWO) and Grid Search-based Weight Optimization (GSWO). Experimentation is conducted on the Figshare Contrast-Enhanced MRI (CE-MRI) brain tumor dataset, comprising 3064 images. Our approach achieves notable accuracy scores, with Xception, ResNet50V2, ResNet152V2, InceptionResNetV2, GAWO, and GSWO attaining 99.42%, 98.37%, 98.22%, 98.26%, 99.71%, and 99.76% accuracy, respectively. Notably, GSWO demonstrates superior accuracy, averaging 99.76\% accuracy across five folds on the Figshare CE-MRI brain tumor dataset. The comparative analysis highlights the significant performance enhancement of our proposed model over existing counterparts. In conclusion, our optimized deep ensemble model exhibits exceptional accuracy in swiftly classifying brain tumors. Furthermore, it has the potential to assist neurologists and clinicians in making accurate and immediate diagnostic decisions.
Authors: Pangoth Santhosh Kumar, Kundrapu Supriya, Mallikharjuna Rao K, Taraka Satya Krishna Teja Malisetti
Abstract: Lung disease is a common health problem in many parts of the world. It is a significant risk to people health and quality of life all across the globe since it is responsible for five of the top thirty leading causes of death. Among them are COVID 19, pneumonia, and tuberculosis, to name just a few. It is critical to diagnose lung diseases in their early stages. Several different models including machine learning and image processing have been developed for this purpose. The earlier a condition is diagnosed, the better the patient chances of making a full recovery and surviving into the long term. Thanks to deep learning algorithms, there is significant promise for the autonomous, rapid, and accurate identification of lung diseases based on medical imaging. Several different deep learning strategies, including convolutional neural networks (CNN), vanilla neural networks, visual geometry group based networks (VGG), and capsule networks , are used for the goal of making lung disease forecasts. The standard CNN has a poor performance when dealing with rotated, tilted, or other aberrant picture orientations. As a result of this, within the scope of this study, we have suggested a vision transformer based approach end to end framework for the diagnosis of lung disorders. In the architecture, data augmentation, training of the suggested models, and evaluation of the models are all included. For the purpose of detecting lung diseases such as pneumonia, Covid 19, lung opacity, and others, a specialised Compact Convolution Transformers (CCT) model have been tested and evaluated on datasets such as the Covid 19 Radiography Database. The model has achieved a better accuracy for both its training and validation purposes on the Covid 19 Radiography Database.
Authors: Haechan Mark Bong, Rongge Zhang, Ricardo de Azambuja, Giovanni Beltrame
Abstract: This work targets what we consider to be the foundational step for urban airborne robots, a safe landing. Our attention is directed toward what we deem the most crucial aspect of the safe landing perception stack: segmentation. We present a streamlined reactive UAV system that employs visual servoing by harnessing the capabilities of open vocabulary image segmentation. This approach can adapt to various scenarios with minimal adjustments, bypassing the necessity for extensive data accumulation for refining internal models, thanks to its open vocabulary methodology. Given the limitations imposed by local authorities, our primary focus centers on operations originating from altitudes of 100 meters. This choice is deliberate, as numerous preceding works have dealt with altitudes up to 30 meters, aligning with the capabilities of small stereo cameras. Consequently, we leave the remaining 20m to be navigated using conventional 3D path planning methods. Utilizing monocular cameras and image segmentation, our findings demonstrate the system's capability to successfully execute landing maneuvers at altitudes as low as 20 meters. However, this approach is vulnerable to intermittent and occasionally abrupt fluctuations in the segmentation between frames in a video stream. To address this challenge, we enhance the image segmentation output by introducing what we call a dynamic focus: a masking mechanism that self adjusts according to the current landing stage. This dynamic focus guides the control system to avoid regions beyond the drone's safety radius projected onto the ground, thus mitigating the problems with fluctuations. Through the implementation of this supplementary layer, our experiments have reached improvements in the landing success rate of almost tenfold when compared to global segmentation. All the source code is open source and available online (github.com/MISTLab/DOVESEI).
Authors: Ang Li, Yifei Wang, Yiwen Guo, Yisen Wang
Abstract: The existence of adversarial examples has been a mystery for years and attracted much interest. A well-known theory by \citet{ilyas2019adversarial} explains adversarial vulnerability from a data perspective by showing that one can extract non-robust features from adversarial examples and these features alone are useful for classification. However, the explanation remains quite counter-intuitive since non-robust features are mostly noise features to humans. In this paper, we re-examine the theory from a larger context by incorporating multiple learning paradigms. Notably, we find that contrary to their good usefulness under supervised learning, non-robust features attain poor usefulness when transferred to other self-supervised learning paradigms, such as contrastive learning, masked image modeling, and diffusion models. It reveals that non-robust features are not really as useful as robust or natural features that enjoy good transferability between these paradigms. Meanwhile, for robustness, we also show that naturally trained encoders from robust features are largely non-robust under AutoAttack. Our cross-paradigm examination suggests that the non-robust features are not really useful but more like paradigm-wise shortcuts, and robust features alone might be insufficient to attain reliable model robustness. Code is available at \url{https://github.com/PKU-ML/AdvNotRealFeatures}.
Authors: Naoki Wake, Atsushi Kanehira, Kazuhiro Sasabuchi, Jun Takamatsu, Katsushi Ikeuchi
Abstract: We introduce a pipeline that enhances a general-purpose Vision Language Model, GPT-4V(ision), to facilitate one-shot visual teaching for robotic manipulation. This system analyzes videos of humans performing tasks and outputs executable robot programs that incorporate insights into affordances. The process begins with GPT-4V analyzing the videos to obtain textual explanations of environmental and action details. A GPT-4-based task planner then encodes these details into a symbolic task plan. Subsequently, vision systems spatially and temporally ground the task plan in the videos. Object are identified using an open-vocabulary object detector, and hand-object interactions are analyzed to pinpoint moments of grasping and releasing. This spatiotemporal grounding allows for the gathering of affordance information (e.g., grasp types, waypoints, and body postures) critical for robot execution. Experiments across various scenarios demonstrate the method's efficacy in achieving real robots' operations from human demonstrations in a one-shot manner. Meanwhile, quantitative tests have revealed instances of hallucination in GPT-4V, highlighting the importance of incorporating human supervision within the pipeline. The prompts of GPT-4V/GPT-4 are available at this project page:
Authors: Raghavv Goel, Cecilia Morales, Manpreet Singh, Artur Dubrawski, John Galeotti, Howie Choset
Abstract: Segmenting a moving needle in ultrasound images is challenging due to the presence of artifacts, noise, and needle occlusion. This task becomes even more demanding in scenarios where data availability is limited. In this paper, we present a novel approach for needle segmentation for 2D ultrasound that combines classical Kalman Filter (KF) techniques with data-driven learning, incorporating both needle features and needle motion. Our method offers three key contributions. First, we propose a compatible framework that seamlessly integrates into commonly used encoder-decoder style architectures. Second, we demonstrate superior performance compared to recent state-of-the-art needle segmentation models using our novel convolutional neural network (CNN) based KF-inspired block, achieving a 15\% reduction in pixel-wise needle tip error and an 8\% reduction in length error. Third, to our knowledge we are the first to implement a learnable filter to incorporate non-linear needle motion for improving needle segmentation.
Authors: Ying Jiang, Chang Yu, Tianyi Xie, Xuan Li, Yutao Feng, Huamin Wang, Minchen Li, Henry Lau, Feng Gao, Yin Yang, Chenfanfu Jiang
Abstract: As consumer Virtual Reality (VR) and Mixed Reality (MR) technologies gain momentum, there's a growing focus on the development of engagements with 3D virtual content. Unfortunately, traditional techniques for content creation, editing, and interaction within these virtual spaces are fraught with difficulties. They tend to be not only engineering-intensive but also require extensive expertise, which adds to the frustration and inefficiency in virtual object manipulation. Our proposed VR-GS system represents a leap forward in human-centered 3D content interaction, offering a seamless and intuitive user experience. By developing a physical dynamics-aware interactive Gaussian Splatting in a Virtual Reality setting, and constructing a highly efficient two-level embedding strategy alongside deformable body simulations, VR-GS ensures real-time execution with highly realistic dynamic responses. The components of our Virtual Reality system are designed for high efficiency and effectiveness, starting from detailed scene reconstruction and object segmentation, advancing through multi-view image in-painting, and extending to interactive physics-based editing. The system also incorporates real-time deformation embedding and dynamic shadow casting, ensuring a comprehensive and engaging virtual experience.Our project page is available at: https://yingjiang96.github.io/VR-GS/.
Authors: Xiaoyu Qiao, Weisheng Li, Guofen Wang, Yuping Huang
Abstract: Deep learning (DL)-based methods offer a promising solution to reduce the prolonged scanning time in magnetic resonance imaging (MRI). While model-driven DL methods have demonstrated convincing results by incorporating prior knowledge into deep networks, further exploration is needed to optimize the integration of diverse priors.. Existing model-driven networks typically utilize linearly stacked unrolled cascades to mimic iterative solution steps in optimization algorithms. However, this approach needs to find a balance between different prior-based regularizers during training, resulting in slower convergence and suboptimal reconstructions. To overcome the limitations, we propose a collaborative model-driven network to maximally exploit the complementarity of different regularizers. We design attention modules to learn both the relative confidence (RC) and overall confidence (OC) for the intermediate reconstructions (IRs) generated by different prior-based subnetworks. RC assigns more weight to the areas of expertise of the subnetworks, enabling precise element-wise collaboration. We design correction modules to tackle bottleneck scenarios where both subnetworks exhibit low accuracy, and they further optimize the IRs based on OC maps. IRs across various stages are concatenated and fed to the attention modules to build robust and accurate confidence maps. Experimental results on multiple datasets showed significant improvements in the final results without additional computational costs. Moreover, the proposed model-driven network design strategy can be conveniently applied to various model-driven methods to improve their performance.
Authors: Yameng Peng, Andy Song, Haytham M. Fayek, Vic Ciesielski, Xiaojun Chang
Abstract: Training-free metrics (a.k.a. zero-cost proxies) are widely used to avoid resource-intensive neural network training, especially in Neural Architecture Search (NAS). Recent studies show that existing training-free metrics have several limitations, such as limited correlation and poor generalisation across different search spaces and tasks. Hence, we propose Sample-Wise Activation Patterns and its derivative, SWAP-Score, a novel high-performance training-free metric. It measures the expressivity of networks over a batch of input samples. The SWAP-Score is strongly correlated with ground-truth performance across various search spaces and tasks, outperforming 15 existing training-free metrics on NAS-Bench-101/201/301 and TransNAS-Bench-101. The SWAP-Score can be further enhanced by regularisation, which leads to even higher correlations in cell-based search space and enables model size control during the search. For example, Spearman's rank correlation coefficient between regularised SWAP-Score and CIFAR-100 validation accuracies on NAS-Bench-201 networks is 0.90, significantly higher than 0.80 from the second-best metric, NWOT. When integrated with an evolutionary algorithm for NAS, our SWAP-NAS achieves competitive performance on CIFAR-10 and ImageNet in approximately 6 minutes and 9 minutes of GPU time respectively.
Authors: Juan Manuel Zambrano Chaves, Shih-Cheng Huang, Yanbo Xu, Hanwen Xu, Naoto Usuyama, Sheng Zhang, Fei Wang, Yujia Xie, Mahmoud Khademi, Ziyi Yang, Hany Awadalla, Julia Gong, Houdong Hu, Jianwei Yang, Chunyuan Li, Jianfeng Gao, Yu Gu, Cliff Wong, Mu Wei, Tristan Naumann, Muhao Chen, Matthew P. Lungren, Serena Yeung-Levy, Curtis P. Langlotz, Sheng Wang, Hoifung Poon
Abstract: The scaling laws and extraordinary performance of large foundation models motivate the development and utilization of such models in biomedicine. However, despite early promising results on some biomedical benchmarks, there are still major challenges that need to be addressed before these models can be used in real-world clinics. Frontier general-domain models such as GPT-4V still have significant performance gaps in multimodal biomedical applications. More importantly, less-acknowledged pragmatic issues, including accessibility, model cost, and tedious manual evaluation make it hard for clinicians to use state-of-the-art large models directly on private patient data. Here, we explore training open-source small multimodal models (SMMs) to bridge competency gaps for unmet clinical needs in radiology. To maximize data efficiency, we adopt a modular approach by incorporating state-of-the-art pre-trained models for image and text modalities, and focusing on training a lightweight adapter to ground each modality to the text embedding space, as exemplified by LLaVA-Med. For training, we assemble a large dataset of over 697 thousand radiology image-text pairs. For evaluation, we propose CheXprompt, a GPT-4-based metric for factuality evaluation, and demonstrate its parity with expert evaluation. For best practice, we conduct a systematic ablation study on various choices in data engineering and multimodal training. The resulting LlaVA-Rad (7B) model attains state-of-the-art results on standard radiology tasks such as report generation and cross-modal retrieval, even outperforming much larger models such as GPT-4V and Med-PaLM M (84B). The inference of LlaVA-Rad is fast and can be performed on a single V100 GPU in private settings, offering a promising state-of-the-art tool for real-world clinical applications.
Authors: Jay Shah, Md Mahfuzur Rahman Siddiquee, Yi Su, Teresa Wu, Baoxin Li
Abstract: Age is one of the major known risk factors for Alzheimer's Disease (AD). Detecting AD early is crucial for effective treatment and preventing irreversible brain damage. Brain age, a measure derived from brain imaging reflecting structural changes due to aging, may have the potential to identify AD onset, assess disease risk, and plan targeted interventions. Deep learning-based regression techniques to predict brain age from magnetic resonance imaging (MRI) scans have shown great accuracy recently. However, these methods are subject to an inherent regression to the mean effect, which causes a systematic bias resulting in an overestimation of brain age in young subjects and underestimation in old subjects. This weakens the reliability of predicted brain age as a valid biomarker for downstream clinical applications. Here, we reformulate the brain age prediction task from regression to classification to address the issue of systematic bias. Recognizing the importance of preserving ordinal information from ages to understand aging trajectory and monitor aging longitudinally, we propose a novel ORdinal Distance Encoded Regularization (ORDER) loss that incorporates the order of age labels, enhancing the model's ability to capture age-related patterns. Extensive experiments and ablation studies demonstrate that this framework reduces systematic bias, outperforms state-of-art methods by statistically significant margins, and can better capture subtle differences between clinical groups in an independent AD dataset. Our implementation is publicly available at https://github.com/jaygshah/Robust-Brain-Age-Prediction.
URLs: https://github.com/jaygshah/Robust-Brain-Age-Prediction.
Authors: Hugo Caselles-Dupr\'e, Charles Mellerio, Paul H\'erent, Aliz\'ee Lopez-Persem, Benoit B\'eranger, Mathieu Soularue, Pierre Fautrel, Gauthier Vernier, Matthieu Cord
Abstract: The reconstruction of images observed by subjects from fMRI data collected during visual stimuli has made strong progress in the past decade, thanks to the availability of extensive fMRI datasets and advancements in generative models for image generation. However, the application of visual reconstruction has remained limited. Reconstructing visual imagination presents a greater challenge, with potentially revolutionary applications ranging from aiding individuals with disabilities to verifying witness accounts in court. The primary hurdles in this field are the absence of data collection protocols for visual imagery and the lack of datasets on the subject. Traditionally, fMRI-to-image relies on data collected from subjects exposed to visual stimuli, which poses issues for generating visual imagery based on the difference of brain activity between visual stimulation and visual imagery. For the first time, we have compiled a substantial dataset (around 6h of scans) on visual imagery along with a proposed data collection protocol. We then train a modified version of an fMRI-to-image model and demonstrate the feasibility of reconstructing images from two modes of imagination: from memory and from pure imagination. The resulting pipeline we call Mind-to-Image marks a step towards creating a technology that allow direct reconstruction of visual imagery.
Authors: Zhaoxi Mu, Xinyu Yang
Abstract: The integration of visual cues has revitalized the performance of the target speech extraction task, elevating it to the forefront of the field. Nevertheless, this multi-modal learning paradigm often encounters the challenge of modality imbalance. In audio-visual target speech extraction tasks, the audio modality tends to dominate, potentially overshadowing the importance of visual guidance. To tackle this issue, we propose AVSepChain, drawing inspiration from the speech chain concept. Our approach partitions the audio-visual target speech extraction task into two stages: speech perception and speech production. In the speech perception stage, audio serves as the dominant modality, while visual information acts as the conditional modality. Conversely, in the speech production stage, the roles are reversed. This transformation of modality status aims to alleviate the problem of modality imbalance. Additionally, we introduce a contrastive semantic matching loss to ensure that the semantic information conveyed by the generated speech aligns with the semantic information conveyed by lip movements during the speech production stage. Through extensive experiments conducted on multiple benchmark datasets for audio-visual target speech extraction, we showcase the superior performance achieved by our proposed method.
Authors: Zirui Zang, Ahmad Amine, Rahul Mangharam
Abstract: Estimating ego-pose from cameras is an important problem in robotics with applications ranging from mobile robotics to augmented reality. While SOTA models are becoming increasingly accurate, they can still be unwieldy due to high computational costs. In this paper, we propose to solve the problem by using invertible neural networks (INN) to find the mapping between the latent space of images and poses for a given scene. Our model achieves similar performance to the SOTA while being faster to train and only requiring offline rendering of low-resolution synthetic data. By using normalizing flows, the proposed method also provides uncertainty estimation for the output. We also demonstrated the efficiency of this method by deploying the model on a mobile robot.
Authors: Shitong Shao, Zikai Zhou, Huanran Chen, Zhiqiang Shen
Abstract: Dataset condensation, a concept within data-centric learning, efficiently transfers critical attributes from an original dataset to a synthetic version, maintaining both diversity and realism. This approach significantly improves model training efficiency and is adaptable across multiple application areas. Previous methods in dataset condensation have faced challenges: some incur high computational costs which limit scalability to larger datasets (e.g., MTT, DREAM, and TESLA), while others are restricted to less optimal design spaces, which could hinder potential improvements, especially in smaller datasets (e.g., SRe2L, G-VBSM, and RDED). To address these limitations, we propose a comprehensive design framework that includes specific, effective strategies like implementing soft category-aware matching and adjusting the learning rate schedule. These strategies are grounded in empirical evidence and theoretical backing. Our resulting approach, Elucidate Dataset Condensation (EDC), establishes a benchmark for both small and large-scale dataset condensation. In our testing, EDC achieves state-of-the-art accuracy, reaching 48.6% on ImageNet-1k with a ResNet-18 model at an IPC of 10, which corresponds to a compression ratio of 0.78%. This performance exceeds those of SRe2L, G-VBSM, and RDED by margins of 27.3%, 17.2%, and 6.6%, respectively.
Authors: Shihao Zhang, kenji kawaguchi, Angela Yao
Abstract: Most works studying representation learning focus only on classification and neglect regression. Yet, the learning objectives and therefore the representation topologies of the two tasks are fundamentally different: classification targets class separation, leading to disconnected representations, whereas regression requires ordinality with respect to the target, leading to continuous representations. We thus wonder how the effectiveness of a regression representation is influenced by its topology, with evaluation based on the Information Bottleneck (IB) principle. The IB principle is an important framework that provides principles for learning effectiveness representations. We establish two connections between it and the topology of regression representations. The first connection reveals that a lower intrinsic dimension of the feature space implies a reduced complexity of the representation Z. This complexity can be quantified as the conditional entropy of Z on the target space Y and serves as an upper bound on the generalization error. The second connection suggests learning a feature space that is topologically similar to the target space will better align with the IB principle. Based on these two connections, we introduce PH-Reg, a regularizer specific to regression that matches the intrinsic dimension and topology of the feature space with the target space. Experiments on synthetic and real-world regression tasks demonstrate the benefits of PH-Reg.
Authors: Yi Yu, Yufei Wang, Song Xia, Wenhan Yang, Shijian Lu, Yap-Peng Tan, Alex C. Kot
Abstract: Unlearnable examples (UEs) seek to maximize testing error by making subtle modifications to training examples that are correctly labeled. Defenses against these poisoning attacks can be categorized based on whether specific interventions are adopted during training. The first approach is training-time defense, such as adversarial training, which can mitigate poisoning effects but is computationally intensive. The other approach is pre-training purification, e.g., image short squeezing, which consists of several simple compressions but often encounters challenges in dealing with various UEs. Our work provides a novel disentanglement mechanism to build an efficient pre-training purification method. Firstly, we uncover rate-constrained variational autoencoders (VAEs), demonstrating a clear tendency to suppress the perturbations in UEs. We subsequently conduct a theoretical analysis for this phenomenon. Building upon these insights, we introduce a disentangle variational autoencoder (D-VAE), capable of disentangling the perturbations with learnable class-wise embeddings. Based on this network, a two-stage purification approach is naturally developed. The first stage focuses on roughly eliminating perturbations, while the second stage produces refined, poison-free results, ensuring effectiveness and robustness across various scenarios. Extensive experiments demonstrate the remarkable performance of our method across CIFAR-10, CIFAR-100, and a 100-class ImageNet-subset. Code is available at https://github.com/yuyi-sd/D-VAE.
Authors: Deegan Atha, R. Michael Swan, Abhishek Cauligi, Anne Bettens, Edwin Goh, Dima Kogan, Larry Matthies, Masahiro Ono
Abstract: The ability to determine the pose of a rover in an inertial frame autonomously is a crucial capability necessary for the next generation of surface rover missions on other planetary bodies. Currently, most on-going rover missions utilize ground-in-the-loop interventions to manually correct for drift in the pose estimate and this human supervision bottlenecks the distance over which rovers can operate autonomously and carry out scientific measurements. In this paper, we present ShadowNav, an autonomous approach for global localization on the Moon with an emphasis on driving in darkness and at nighttime. Our approach uses the leading edge of Lunar craters as landmarks and a particle filtering approach is used to associate detected craters with known ones on an offboard map. We discuss the key design decisions in developing the ShadowNav framework for use with a Lunar rover concept equipped with a stereo camera and an external illumination source. Finally, we demonstrate the efficacy of our proposed approach in both a Lunar simulation environment and on data collected during a field test at Cinder Lakes, Arizona.
Authors: Guanyiman Fu, Fengchao Xiong, Jianfeng Lu, Jun Zhou, Yuntao Qian
Abstract: Denoising hyperspectral images (HSIs) is a crucial preprocessing procedure due to the noise originating from intra-imaging mechanisms and environmental factors. Utilizing domain-specific knowledge of HSIs, such as spectral correlation, spatial self-similarity, and spatial-spectral correlation, is essential for deep learning-based denoising. Existing methods are often constrained by running time, space complexity, and computational complexity, employing strategies that explore these priors separately. While these strategies can avoid some redundant information, they inevitably overlook broader and more underlying long-range spatial-spectral information that positively impacts image restoration. This paper proposes a Spatial-Spectral Selective State Space Model-based U-shaped network, termed Spatial-Spectral U-Mamba (SSUMamba), for hyperspectral image denoising. We can obtain complete global spatial-spectral correlation within a module thanks to the linear space complexity in State Space Model (SSM) computations. We introduce a Spatial-Spectral Alternating Scan (SSAS) strategy for HSIs, which helps model the information flow in multiple directions in 3-D HSIs. Experimental results demonstrate that our method outperforms compared methods. The source code will be available at https://github.com/lronkitty/SSUMamba.
Authors: Alessandro Pianese, Davide Cozzolino, Giovanni Poggi, Luisa Verdoliva
Abstract: Generalization is a main issue for current audio deepfake detectors, which struggle to provide reliable results on out-of-distribution data. Given the speed at which more and more accurate synthesis methods are developed, it is very important to design techniques that work well also on data they were not trained for. In this paper we study the potential of large-scale pre-trained models for audio deepfake detection, with special focus on generalization ability. To this end, the detection problem is reformulated in a speaker verification framework and fake audios are exposed by the mismatch between the voice sample under test and the voice of the claimed identity. With this paradigm, no fake speech sample is necessary in training, cutting off any link with the generation method at the root, and ensuring full generalization ability. Features are extracted by general-purpose large pre-trained models, with no need for training or fine-tuning on specific fake detection or speaker verification datasets. At detection time only a limited set of voice fragments of the identity under test is required. Experiments on several datasets widespread in the community show that detectors based on pre-trained models achieve excellent performance and show strong generalization ability, rivaling supervised methods on in-distribution data and largely overcoming them on out-of-distribution data.