Authors: Md Manjurul Ahsan, Shivakumar Raman, Yingtao Liu, Zahed Siddique
Abstract: Diffusion Models are probabilistic models that create realistic samples by simulating the diffusion process, gradually adding and removing noise from data. These models have gained popularity in domains such as image processing, speech synthesis, and natural language processing due to their ability to produce high-quality samples. As Diffusion Models are being adopted in various domains, existing literature reviews that often focus on specific areas like computer vision or medical imaging may not serve a broader audience across multiple fields. Therefore, this review presents a comprehensive overview of Diffusion Models, covering their theoretical foundations and algorithmic innovations. We highlight their applications in diverse areas such as media quality, authenticity, synthesis, image transformation, healthcare, and more. By consolidating current knowledge and identifying emerging trends, this review aims to facilitate a deeper understanding and broader adoption of Diffusion Models and provide guidelines for future researchers and practitioners across diverse disciplines.
Authors: Ananya Pandey, Dinesh Kumar Vishwakarma
Abstract: Various linguistic and non-linguistic clues, such as excessive emphasis on a word, a shift in the tone of voice, or an awkward expression, frequently convey sarcasm. The computer vision problem of sarcasm recognition in conversation aims to identify hidden sarcastic, criticizing, and metaphorical information embedded in everyday dialogue. Prior, sarcasm recognition has focused mainly on text. Still, it is critical to consider all textual information, audio stream, facial expression, and body position for reliable sarcasm identification. Hence, we propose a novel approach that combines a lightweight depth attention module with a self-regulated ConvNet to concentrate on the most crucial features of visual data and an attentional tokenizer based strategy to extract the most critical context-specific information from the textual data. The following is a list of the key contributions that our experimentation has made in response to performing the task of Multi-modal Sarcasm Recognition: an attentional tokenizer branch to get beneficial features from the glossary content provided by the subtitles; a visual branch for acquiring the most prominent features from the video frames; an utterance-level feature extraction from acoustic content and a multi-headed attention based feature fusion branch to blend features obtained from multiple modalities. Extensive testing on one of the benchmark video datasets, MUSTaRD, yielded an accuracy of 79.86% for speaker dependent and 76.94% for speaker independent configuration demonstrating that our approach is superior to the existing methods. We have also conducted a cross-dataset analysis to test the adaptability of VyAnG-Net with unseen samples of another dataset MUStARD++.
Authors: Ananya Pandey, Dinesh Kumar Vishwakarma
Abstract: The natural language processing and multimedia field has seen a notable surge in interest in multimodal sentiment recognition. Hence, this study aims to employ Target-Dependent Multimodal Sentiment Analysis (TDMSA) to identify the level of sentiment associated with every target (aspect) stated within a multimodal post consisting of a visual-caption pair. Despite the recent advancements in multimodal sentiment recognition, there has been a lack of explicit incorporation of emotional clues from the visual modality, specifically those pertaining to facial expressions. The challenge at hand is to proficiently obtain visual and emotional clues and subsequently synchronise them with the textual content. In light of this fact, this study presents a novel approach called the Visual-to-Emotional-Caption Translation Network (VECTN) technique. The primary objective of this strategy is to effectively acquire visual sentiment clues by analysing facial expressions. Additionally, it effectively aligns and blends the obtained emotional clues with the target attribute of the caption mode. The experimental findings demonstrate that our methodology is capable of producing ground-breaking outcomes when applied to two publicly accessible multimodal Twitter datasets, namely, Twitter-2015 and Twitter-2017. The experimental results show that the suggested model achieves an accuracy of 81.23% and a macro-F1 of 80.61% on the Twitter-15 dataset, while 77.42% and 75.19% on the Twitter-17 dataset, respectively. The observed improvement in performance reveals that our model is better than others when it comes to collecting target-level sentiment in multimodal data using the expressions of the face.
Authors: Rishit Dagli, Atsuhiro Hibi, Rahul G. Krishnan, Pascal N. Tyrrell
Abstract: Current methods for performing 3D reconstruction and novel view synthesis (NVS) in ultrasound imaging data often face severe artifacts when training NeRF-based approaches. The artifacts produced by current approaches differ from NeRF floaters in general scenes because of the unique nature of ultrasound capture. Furthermore, existing models fail to produce reasonable 3D reconstructions when ultrasound data is captured or obtained casually in uncontrolled environments, which is common in clinical settings. Consequently, existing reconstruction and NVS methods struggle to handle ultrasound motion, fail to capture intricate details, and cannot model transparent and reflective surfaces. In this work, we introduced NeRF-US, which incorporates 3D-geometry guidance for border probability and scattering density into NeRF training, while also utilizing ultrasound-specific rendering over traditional volume rendering. These 3D priors are learned through a diffusion model. Through experiments conducted on our new "Ultrasound in the Wild" dataset, we observed accurate, clinically plausible, artifact-free reconstructions.
Authors: Federico Simonetta, Rishav Mondal, Luca Andrea Ludovico, Stavros Ntalampiras
Abstract: The Ricordi archive, a prestigious collection of significant musical manuscripts from renowned opera composers such as Donizetti, Verdi and Puccini, has been digitized. This process has allowed us to automatically extract samples that represent various musical elements depicted on the manuscripts, including notes, staves, clefs, erasures, and composer's annotations, among others. To distinguish between digitization noise and actual music elements, a subset of these images was meticulously grouped and labeled by multiple individuals into several classes. After assessing the consistency of the annotations, we trained multiple neural network-based classifiers to differentiate between the identified music elements. The primary objective of this study was to evaluate the reliability of these classifiers, with the ultimate goal of using them for the automatic categorization of the remaining unannotated data set. The dataset, complemented by manual annotations, models, and source code used in these experiments are publicly accessible for replication purposes.
Authors: Seunghwan Kim, Byunghwee Lee, Wonjae Lee
Abstract: The advent of computational and numerical methods in recent times has provided new avenues for analyzing art historiographical narratives and tracing the evolution of art styles therein. Here, we investigate an evolutionary process underpinning the emergence and stylization of contemporary user-generated visual art styles using the complexity-entropy (C-H) plane, which quantifies local structures in paintings. Informatizing 149,780 images curated in DeviantArt and Behance platforms from 2010 to 2020, we analyze the relationship between local information of the C-H space and multi-level image features generated by a deep neural network and a feature extraction algorithm. The results reveal significant statistical relationships between the C-H information of visual artistic styles and the dissimilarities of the multi-level image features over time within groups of artworks. By disclosing a particular C-H region where the diversity of image representations is noticeably manifested, our analyses reveal an empirical condition of emerging styles that are both novel in the C-H plane and characterized by greater stylistic diversity. Our research shows that visual art analyses combined with physics-inspired methodologies and machine learning, can provide macroscopic insights into quantitatively mapping relevant characteristics of an evolutionary process underpinning the creative stylization of uncharted visual arts of given groups and time.
Authors: Syed Rifat Raiyan, Zibran Zarif Amio, Sabbir Ahmed
Abstract: Hand shadow puppetry, also known as shadowgraphy or ombromanie, is a form of theatrical art and storytelling where hand shadows are projected onto flat surfaces to create illusions of living creatures. The skilled performers create these silhouettes by hand positioning, finger movements, and dexterous gestures to resemble shadows of animals and objects. Due to the lack of practitioners and a seismic shift in people's entertainment standards, this art form is on the verge of extinction. To facilitate its preservation and proliferate it to a wider audience, we introduce ${\rm H{\small A}SP{\small E}R}$, a novel dataset consisting of 8,340 images of hand shadow puppets across 11 classes extracted from both professional and amateur hand shadow puppeteer clips. We provide a detailed statistical analysis of the dataset and employ a range of pretrained image classification models to establish baselines. Our findings show a substantial performance superiority of traditional convolutional models over attention-based transformer architectures. We also find that lightweight models, such as MobileNetV2, suited for mobile applications and embedded devices, perform comparatively well. We surmise that such low-latency architectures can be useful in developing ombromanie teaching tools, and we create a prototype application to explore this surmission. Keeping the best-performing model InceptionV3 under the limelight, we conduct comprehensive feature-spatial, explainability, and error analyses to gain insights into its decision-making process. To the best of our knowledge, this is the first documented dataset and research endeavor to preserve this dying art for future generations, with computer vision approaches. Our code and data are publicly available.
Authors: Yue Zhang, Parisa Kordjamshidi
Abstract: The existing methods for Vision and Language Navigation in the Continuous Environment (VLN-CE) commonly incorporate a waypoint predictor to discretize the environment. This simplifies the navigation actions into a view selection task and improves navigation performance significantly compared to direct training using low-level actions. However, the VLN-CE agents are still far from the real robots since there are gaps between their visual perception and executed actions. First, VLN-CE agents that discretize the visual environment are primarily trained with high-level view selection, which causes them to ignore crucial spatial reasoning within the low-level action movements. Second, in these models, the existing waypoint predictors neglect object semantics and their attributes related to passibility, which can be informative in indicating the feasibility of actions. To address these two issues, we introduce a low-level action decoder jointly trained with high-level action prediction, enabling the current VLN agent to learn and ground the selected visual view to the low-level controls. Moreover, we enhance the current waypoint predictor by utilizing visual representations containing rich semantic information and explicitly masking obstacles based on humans' prior knowledge about the feasibility of actions. Empirically, our agent can improve navigation performance metrics compared to the strong baselines on both high-level and low-level actions.
Authors: Khadija Iddrisu, Waseem Shariff, Noel E. OConnor, Joseph Lemley, Suzanne Little
Abstract: Event Cameras, also known as Neuromorphic sensors, capture changes in local light intensity at the pixel level, producing asynchronously generated data termed ``events''. This distinct data format mitigates common issues observed in conventional cameras, like under-sampling when capturing fast-moving objects, thereby preserving critical information that might otherwise be lost. However, leveraging this data often necessitates the development of specialized, handcrafted event representations that can integrate seamlessly with conventional Convolutional Neural Networks (CNNs), considering the unique attributes of event data. In this study, We evaluate event-based Face and Eye tracking. The core objective of our study is to showcase the viability of integrating conventional algorithms with event-based data, transformed into a frame format while preserving the unique benefits of event cameras. To validate our approach, we constructed a frame-based event dataset by simulating events between RGB frames derived from the publicly accessible Helen Dataset. We assess its utility for face and eye detection tasks through the application of GR-YOLO -- a pioneering technique derived from YOLOv3. This evaluation includes a comparative analysis with results derived from training the dataset with YOLOv8. Subsequently, the trained models were tested on real event streams from various iterations of Prophesee's event cameras and further evaluated on the Faces in Event Stream (FES) benchmark dataset. The models trained on our dataset shows a good prediction performance across all the datasets obtained for validation with the best results of a mean Average precision score of 0.91. Additionally, The models trained demonstrated robust performance on real event camera data under varying light conditions.
Authors: Vijul Shah, Brian B. Moser, Ko Watanabe, Andreas Dengel
Abstract: Capturing pupil diameter is essential for assessing psychological and physiological states such as stress levels and cognitive load. However, the low resolution of images in eye datasets often hampers precise measurement. This study evaluates the impact of various upscaling methods, ranging from bicubic interpolation to advanced super-resolution, on pupil diameter predictions. We compare several pre-trained methods, including CodeFormer, GFPGAN, Real-ESRGAN, HAT, and SRResNet. Our findings suggest that pupil diameter prediction models trained on upscaled datasets are highly sensitive to the selected upscaling method and scale. Our results demonstrate that upscaling methods consistently enhance the accuracy of pupil diameter prediction models, highlighting the importance of upscaling in pupilometry. Overall, our work provides valuable insights for selecting upscaling techniques, paving the way for more accurate assessments in psychological and physiological research.
Authors: Xiao Zhang, Zhanhong Huang, Garcia Gonzalez Antony, Witek Jachimczyk, Xinming Huang
Abstract: In this study, we introduce a novel parallel processing framework for real-time point cloud ground segmentation on FPGA platforms, aimed at adapting LiDAR algorithms to the evolving landscape from mechanical to solid-state LiDAR (SSL) technologies. Focusing on the ground segmentation task, we explore parallel processing techniques on existing approaches and adapt them to real-world SSL data handling. We validated frame-segmentation based parallel processing methods using point-based, voxel-based, and range-image-based ground segmentation approaches on the SemanticKITTI dataset based on mechanical LiDAR. The results revealed the superior performance and robustness of the range-image method, especially in its resilience to slicing. Further, utilizing a custom dataset from our self-built Camera-SSLSS equipment, we examined regular SSL data frames and validated the effectiveness of our parallel approach for SSL sensor. Additionally, our pioneering implementation of range-image ground segmentation on FPGA for SSL sensors demonstrated significant processing speed improvements and resource efficiency, achieving processing rates up to 50.3 times faster than conventional CPU setups. These findings underscore the potential of parallel processing strategies to significantly enhance LiDAR technologies for advanced perception tasks in autonomous systems. Post-publication, both the data and the code will be made available on GitHub.
Authors: Anna-Maria Nau, Phillip Ditto, Dawnie Wolfe Steadman, Audris Mockus
Abstract: Determining the stage of decomposition (SOD) is crucial for estimating the postmortem interval and identifying human remains. Currently, labor-intensive manual scoring methods are used for this purpose, but they are subjective and do not scale for the emerging large-scale archival collections of human decomposition photos. This study explores the feasibility of automating two common human decomposition scoring methods proposed by Megyesi and Gelderman using artificial intelligence (AI). We evaluated two popular deep learning models, Inception V3 and Xception, by training them on a large dataset of human decomposition images to classify the SOD for different anatomical regions, including the head, torso, and limbs. Additionally, an interrater study was conducted to assess the reliability of the AI models compared to human forensic examiners for SOD identification. The Xception model achieved the best classification performance, with macro-averaged F1 scores of .878, .881, and .702 for the head, torso, and limbs when predicting Megyesi's SODs, and .872, .875, and .76 for the head, torso, and limbs when predicting Gelderman's SODs. The interrater study results supported AI's ability to determine the SOD at a reliability level comparable to a human expert. This work demonstrates the potential of AI models trained on a large dataset of human decomposition images to automate SOD identification.
Authors: Yassine Ouali, Adrian Bulat, Brais Martinez, Georgios Tzimiropoulos
Abstract: Despite recent successes, LVLMs or Large Vision Language Models are prone to hallucinating details like objects and their properties or relations, limiting their real-world deployment. To address this and improve their robustness, we present CLIP-DPO, a preference optimization method that leverages contrastively pre-trained Vision-Language (VL) embedding models, such as CLIP, for DPO-based optimization of LVLMs. Unlike prior works tackling LVLM hallucinations, our method does not rely on paid-for APIs, and does not require additional training data or the deployment of other external LVLMs. Instead, starting from the initial pool of supervised fine-tuning data, we generate a diverse set of predictions, which are ranked based on their CLIP image-text similarities, and then filtered using a robust rule-based approach to obtain a set of positive and negative pairs for DPO-based training. We applied CLIP-DPO fine-tuning to the MobileVLM-v2 family of models and to LlaVA-1.5, in all cases observing significant improvements in terms of hallucination reduction over baseline models. We also observe better performance for zero-shot classification, suggesting improved grounding capabilities, and verify that the original performance on standard LVLM benchmarks is overall preserved.
Authors: Niyar R Barman, Krish Sharma, Ashhar Aziz, Shashwat Bajpai, Shwetangshu Biswas, Vasu Sharma, Vinija Jain, Aman Chadha, Amit Sheth, Amitava Das
Abstract: The rapid advancement of text-to-image generation systems, exemplified by models like Stable Diffusion, Midjourney, Imagen, and DALL-E, has heightened concerns about their potential misuse. In response, companies like Meta and Google have intensified their efforts to implement watermarking techniques on AI-generated images to curb the circulation of potentially misleading visuals. However, in this paper, we argue that current image watermarking methods are fragile and susceptible to being circumvented through visual paraphrase attacks. The proposed visual paraphraser operates in two steps. First, it generates a caption for the given image using KOSMOS-2, one of the latest state-of-the-art image captioning systems. Second, it passes both the original image and the generated caption to an image-to-image diffusion system. During the denoising step of the diffusion pipeline, the system generates a visually similar image that is guided by the text caption. The resulting image is a visual paraphrase and is free of any watermarks. Our empirical findings demonstrate that visual paraphrase attacks can effectively remove watermarks from images. This paper provides a critical assessment, empirically revealing the vulnerability of existing watermarking techniques to visual paraphrase attacks. While we do not propose solutions to this issue, this paper serves as a call to action for the scientific community to prioritize the development of more robust watermarking techniques. Our first-of-its-kind visual paraphrase dataset and accompanying code are publicly available.
Authors: Liu He, Yizhi Song, Hejun Huang, Daniel Aliaga, Xin Zhou
Abstract: Text-to-video generation has been dominated by end-to-end diffusion-based or autoregressive models. On one hand, those novel models provide plausible versatility, but they are criticized for physical correctness, shading and illumination, camera motion, and temporal consistency. On the other hand, film industry relies on manually-edited Computer-Generated Imagery (CGI) using 3D modeling software. Human-directed 3D synthetic videos and animations address the aforementioned shortcomings, but it is extremely tedious and requires tight collaboration between movie makers and 3D rendering experts. In this paper, we introduce an automatic synthetic video generation pipeline based on Vision Large Language Model (VLM) agent collaborations. Given a natural language description of a video, multiple VLM agents auto-direct various processes of the generation pipeline. They cooperate to create Blender scripts which render a video that best aligns with the given description. Based on film making inspiration and augmented with Blender-based movie making knowledge, the Director agent decomposes the input text-based video description into sub-processes. For each sub-process, the Programmer agent produces Python-based Blender scripts based on customized function composing and API calling. Then, the Reviewer agent, augmented with knowledge of video reviewing, character motion coordinates, and intermediate screenshots uses its compositional reasoning ability to provide feedback to the Programmer agent. The Programmer agent iteratively improves the scripts to yield the best overall video outcome. Our generated videos show better quality than commercial video generation models in 5 metrics on video quality and instruction-following performance. Moreover, our framework outperforms other approaches in a comprehensive user study on quality, consistency, and rationality.
Authors: Xinyu Liu, Jing Zhang, Kexin Zhang, Xu Liu, Lingling Li
Abstract: Video Object Segmentation (VOS) presents several challenges, including object occlusion and fragmentation, the dis-appearance and re-appearance of objects, and tracking specific objects within crowded scenes. In this work, we combine the strengths of the state-of-the-art (SOTA) models SAM2 and Cutie to address these challenges. Additionally, we explore the impact of various hyperparameters on video instance segmentation performance. Our approach achieves a J\&F score of 0.7952 in the testing phase of LSVOS challenge VOS track, ranking third overall.
Authors: Xiao Wang, Chao wang, Shiao Wang, Xixi Wang, Zhicheng Zhao, Lin Zhu, Bo Jiang
Abstract: Event camera-based visual tracking has drawn more and more attention in recent years due to the unique imaging principle and advantages of low energy consumption, high dynamic range, and dense temporal resolution. Current event-based tracking algorithms are gradually hitting their performance bottlenecks, due to the utilization of vision Transformer and the static template for target object localization. In this paper, we propose a novel Mamba-based visual tracking framework that adopts the state space model with linear complexity as a backbone network. The search regions and target template are fed into the vision Mamba network for simultaneous feature extraction and interaction. The output tokens of search regions will be fed into the tracking head for target localization. More importantly, we consider introducing a dynamic template update strategy into the tracking framework using the Memory Mamba network. By considering the diversity of samples in the target template library and making appropriate adjustments to the template memory module, a more effective dynamic template can be integrated. The effective combination of dynamic and static templates allows our Mamba-based tracking algorithm to achieve a good balance between accuracy and computational cost on multiple large-scale datasets, including EventVOT, VisEvent, and FE240hz. The source code will be released on https://github.com/Event-AHU/MambaEVT
Authors: Xiao Wang, Yao Rong, Fuling Wang, Jianing Li, Lin Zhu, Bo Jiang, Yaowei Wang
Abstract: Sign Language Translation (SLT) is a core task in the field of AI-assisted disability. Unlike traditional SLT based on visible light videos, which is easily affected by factors such as lighting, rapid hand movements, and privacy breaches, this paper proposes the use of high-definition Event streams for SLT, effectively mitigating the aforementioned issues. This is primarily because Event streams have a high dynamic range and dense temporal signals, which can withstand low illumination and motion blur well. Additionally, due to their sparsity in space, they effectively protect the privacy of the target person. More specifically, we propose a new high-resolution Event stream sign language dataset, termed Event-CSL, which effectively fills the data gap in this area of research. It contains 14,827 videos, 14,821 glosses, and 2,544 Chinese words in the text vocabulary. These samples are collected in a variety of indoor and outdoor scenes, encompassing multiple angles, light intensities, and camera movements. We have benchmarked existing mainstream SLT works to enable fair comparison for future efforts. Based on this dataset and several other large-scale datasets, we propose a novel baseline method that fully leverages the Mamba model's ability to integrate temporal information of CNN features, resulting in improved sign language translation outcomes. Both the benchmark dataset and source code will be released on https://github.com/Event-AHU/OpenESL
Authors: Yu Lu, YuYu Chen, Gang Zhou, Zhenghua Lan
Abstract: Textile pilling assessment is critical for textile quality control. We collect thousands of 3D point cloud images in the actual test environment of textiles and organize and label them as TextileNet8 dataset. To the best of our knowledge, it is the first publicly available eight-categories 3D point cloud dataset in the field of textile pilling assessment. Based on PointGPT, the GPT-like big model of point cloud analysis, we incorporate the global features of the input point cloud extracted from the non-parametric network into it, thus proposing the PointGPT+NN model. Using TextileNet8 as a benchmark, the experimental results show that the proposed PointGPT+NN model achieves an overall accuracy (OA) of 91.8% and a mean per-class accuracy (mAcc) of 92.2%. Test results on other publicly available datasets also validate the competitive performance of the proposed PointGPT+NN model. The proposed TextileNet8 dataset will be publicly available.
Authors: Thanh Thi Nguyen, Campbell Wilson, Janis Dalins
Abstract: Assessing the forensic value of hand images involves the use of unique features and patterns present in an individual's hand. The human hand has distinct characteristics, such as the pattern of veins, fingerprints, and the geometry of the hand itself. This paper investigates the use of vision transformers (ViTs) for classification of hand images. We use explainability tools to explore the internal representations of ViTs and assess their impact on the model outputs. Utilizing the internal understanding of ViTs, we introduce distillation methods that allow a student model to adaptively extract knowledge from a teacher model while learning on data of a different domain to prevent catastrophic forgetting. Two publicly available hand image datasets are used to conduct a series of experiments to evaluate performance of the ViTs and our proposed adaptive distillation methods. The experimental results demonstrate that ViT models significantly outperform traditional machine learning methods and the internal states of ViTs are useful for explaining the model outputs in the classification task. By averting catastrophic forgetting, our distillation methods achieve excellent performance on data from both source and target domains, particularly when these two domains exhibit significant dissimilarity. The proposed approaches therefore can be developed and implemented effectively for real-world applications such as access control, identity verification, and authentication systems.
Authors: Md Hadiuzzaman, Mohammed Sowket Ali, Tamanna Sultana, Abdur Raj Shafi, Abu Saleh Musa Miah, Jungpil Shin
Abstract: People commonly communicate in English, Arabic, and Bengali spoken languages through various mediums. However, deaf and hard-of-hearing individuals primarily use body language and sign language to express their needs and achieve independence. Sign language research is burgeoning to enhance communication with the deaf community. While many researchers have made strides in recognizing sign languages such as French, British, Arabic, Turkish, and American, there has been limited research on Bangla sign language (BdSL) with less-than-satisfactory results. One significant barrier has been the lack of a comprehensive Bangla sign language dataset. In our work, we introduced a new BdSL dataset comprising alphabets totaling 18,000 images, with each image being 224x224 pixels in size. Our dataset encompasses 36 Bengali symbols, of which 30 are consonants and the remaining six are vowels. Despite our dataset contribution, many existing systems continue to grapple with achieving high-performance accuracy for BdSL. To address this, we devised a hybrid Convolutional Neural Network (CNN) model, integrating multiple convolutional layers, activation functions, dropout techniques, and LSTM layers. Upon evaluating our hybrid-CNN model with the newly created BdSL dataset, we achieved an accuracy rate of 97.92\%. We are confident that both our BdSL dataset and hybrid CNN model will be recognized as significant milestones in BdSL research.
Authors: Jinghuai Jie, Yan Guo, Guixing Wu, Junmin Wu, Baojian Hua
Abstract: Transformers, renowned for their powerful feature extraction capabilities, have played an increasingly prominent role in various vision tasks. Especially, recent advancements present transformer with hierarchical structures such as Dilated Neighborhood Attention Transformer (DiNAT), demonstrating outstanding ability to efficiently capture both global and local features. However, transformers' application in edge detection has not been fully exploited. In this paper, we propose EdgeNAT, a one-stage transformer-based edge detector with DiNAT as the encoder, capable of extracting object boundaries and meaningful edges both accurately and efficiently. On the one hand, EdgeNAT captures global contextual information and detailed local cues with DiNAT, on the other hand, it enhances feature representation with a novel SCAF-MLA decoder by utilizing both inter-spatial and inter-channel relationships of feature maps. Extensive experiments on multiple datasets show that our method achieves state-of-the-art performance on both RGB and depth images. Notably, on the widely used BSDS500 dataset, our L model achieves impressive performances, with ODS F-measure and OIS F-measure of 86.0%, 87.6% for multi-scale input,and 84.9%, and 86.3% for single-scale input, surpassing the current state-of-the-art EDTER by 1.2%, 1.1%, 1.7%, and 1.6%, respectively. Moreover, as for throughput, our approach runs at 20.87 FPS on RTX 4090 GPU with single-scale input. The code for our method will be released soon.
Authors: Michelle Han, Junyao Chen
Abstract: With diet and nutrition apps reaching 1.4 billion users in 2022 [1], it's no surprise that health apps like MyFitnessPal, Noom, and Calorie Counter, are surging in popularity. However, one major setback [2] of nearly all nutrition applications is that users must enter food data manually, which is time-consuming and tedious. Thus, there has been an increasing demand for applications that can accurately identify food items, analyze their nutritional content, and offer dietary recommendations in real-time. This paper introduces a comprehensive system that combines advanced computer vision techniques with nutrition analysis, implemented in a versatile mobile and web application. The system is divided into three key components: 1) food detection using the YOLOv8 model, 2) nutrient analysis via the Edamam Nutrition Analysis API, and 3) personalized meal recommendations using the Edamam Meal Planning and Recipe Search APIs. Designed for both mobile and web platforms, the application ensures fast processing times with an intuitive user interface, with features such as data visualizations using Chart.js, a login system, and personalized settings for dietary preferences, allergies, and cuisine choices. Preliminary results showcase the system's effectiveness, making it a valuable tool for users to make informed dietary decisions.
Authors: Yuexing Han, Liheng Ruan, Bing Wang
Abstract: The goal of image style transfer is to render an image guided by a style reference while maintaining the original content. Existing image-guided methods rely on specific style reference images, restricting their wider application and potentially compromising result quality. As a flexible alternative, text-guided methods allow users to describe the desired style using text prompts. Despite their versatility, these methods often struggle with maintaining style consistency, reflecting the described style accurately, and preserving the content of the target image. To address these challenges, we introduce FAGStyle, a zero-shot text-guided diffusion image style transfer method. Our approach enhances inter-patch information interaction by incorporating the Sliding Window Crop technique and Feature Augmentation on Geodesic Surface into our style control loss. Furthermore, we integrate a Pre-Shape self-correlation consistency loss to ensure content consistency. FAGStyle demonstrates superior performance over existing methods, consistently achieving stylization that retains the semantic content of the source image. Experimental results confirms the efficacy of FAGStyle across a diverse range of source contents and styles, both imagined and common.
Authors: Jiawei Han, Kaiqi Liu, Wei Li, Guangzhi Chen
Abstract: Point cloud semantic segmentation can significantly enhance the perception of an intelligent agent. Nevertheless, the discriminative capability of the segmentation network is influenced by the quantity of samples available for different categories. To mitigate the cognitive bias induced by class imbalance, this paper introduces a novel method, namely subspace prototype guidance (\textbf{SPG}), to guide the training of segmentation network. Specifically, the point cloud is initially separated into independent point sets by category to provide initial conditions for the generation of feature subspaces. The auxiliary branch which consists of an encoder and a projection head maps these point sets into separate feature subspaces. Subsequently, the feature prototypes which are extracted from the current separate subspaces and then combined with prototypes of historical subspaces guide the feature space of main branch to enhance the discriminability of features of minority categories. The prototypes derived from the feature space of main branch are also employed to guide the training of the auxiliary branch, forming a supervisory loop to maintain consistent convergence of the entire network. The experiments conducted on the large public benchmarks (i.e. S3DIS, ScanNet v2, ScanNet200, Toronto-3D) and collected real-world data illustrate that the proposed method significantly improves the segmentation performance and surpasses the state-of-the-art method. The code is available at \url{https://github.com/Javion11/PointLiBR.git}.
Authors: Diandian Guo, Weixin Si, Zhixi Li, Jialun Pei, Pheng-Ann Heng
Abstract: Pringle maneuver (PM) in laparoscopic liver resection aims to reduce blood loss and provide a clear surgical view by intermittently blocking blood inflow of the liver, whereas prolonged PM may cause ischemic injury. To comprehensively monitor this surgical procedure and provide timely warnings of ineffective and prolonged blocking, we suggest two complementary AI-assisted surgical monitoring tasks: workflow recognition and blocking effectiveness detection in liver resections. The former presents challenges in real-time capturing of short-term PM, while the latter involves the intraoperative discrimination of long-term liver ischemia states. To address these challenges, we meticulously collect a novel dataset, called PmLR50, consisting of 25,037 video frames covering various surgical phases from 50 laparoscopic liver resection procedures. Additionally, we develop an online baseline for PmLR50, termed PmNet. This model embraces Masked Temporal Encoding (MTE) and Compressed Sequence Modeling (CSM) for efficient short-term and long-term temporal information modeling, and embeds Contrastive Prototype Separation (CPS) to enhance action discrimination between similar intraoperative operations. Experimental results demonstrate that PmNet outperforms existing state-of-the-art surgical workflow recognition methods on the PmLR50 benchmark. Our research offers potential clinical applications for the laparoscopic liver surgery community. Source code and data will be publicly available.
Authors: Wenze Liu, Zixuan Ye, Hao Lu, Zhiguo Cao, Xiangyu Yue
Abstract: The labelling difficulty has been a longstanding problem in deep image matting. To escape from fine labels, this work explores using rough annotations such as trimaps coarsely indicating the foreground/background as supervision. We present that the cooperation between learned semantics from indicated known regions and proper assumed matting rules can help infer alpha values at transition areas. Inspired by the nonlocal principle in traditional image matting, we build a directional distance consistency loss (DDC loss) at each pixel neighborhood to constrain the alpha values conditioned on the input image. DDC loss forces the distance of similar pairs on the alpha matte and on its corresponding image to be consistent. In this way, the alpha values can be propagated from learned known regions to unknown transition areas. With only images and trimaps, a matting model can be trained under the supervision of a known loss and the proposed DDC loss. Experiments on AM-2K and P3M-10K dataset show that our paradigm achieves comparable performance with the fine-label-supervised baseline, while sometimes offers even more satisfying results than human-labelled ground truth. Code is available at \url{https://github.com/poppuppy/alpha-free-matting}.
Authors: Bin Cao, Yisi Zhang, Hanyi Wang, Xingjian He, Jing Liu
Abstract: Referring Video Object Segmentation is an emerging multi-modal task that aims to segment objects in the video given a natural language expression. In this work, we build two instance-centric models and fuse predicted results from frame-level and instance-level. First, we introduce instance mask into the DETR-based model for query initialization to achieve temporal enhancement and employ SAM for spatial refinement. Secondly, we build an instance retrieval model conducting binary instance mask classification whether the instance is referred. Finally, we fuse predicted results and our method achieved a score of 52.67 J&F in the validation phase and 60.36 J&F in the test phase, securing the final ranking of 3rd place in the 6-th LSVOS Challenge RVOS Track.
Authors: Kai Liu, Kang You, Pan Gao
Abstract: Stable diffusion networks have emerged as a groundbreaking development for their ability to produce realistic and detailed visual content. This characteristic renders them ideal decoders, capable of producing high-quality and aesthetically pleasing reconstructions. In this paper, we introduce the first diffusion-based point cloud compression method, dubbed Diff-PCC, to leverage the expressive power of the diffusion model for generative and aesthetically superior decoding. Different from the conventional autoencoder fashion, a dual-space latent representation is devised in this paper, in which a compressor composed of two independent encoding backbones is considered to extract expressive shape latents from distinct latent spaces. At the decoding side, a diffusion-based generator is devised to produce high-quality reconstructions by considering the shape latents as guidance to stochastically denoise the noisy point clouds. Experiments demonstrate that the proposed Diff-PCC achieves state-of-the-art compression performance (e.g., 7.711 dB BD-PSNR gains against the latest G-PCC standard at ultra-low bitrate) while attaining superior subjective quality. Source code will be made publicly available.
Authors: Cong Wan, Yuhang He, Xiang Song, Yihong Gong
Abstract: Diffusion models have revolutionized customized text-to-image generation, allowing for efficient synthesis of photos from personal data with textual descriptions. However, these advancements bring forth risks including privacy breaches and unauthorized replication of artworks. Previous researches primarily center around using prompt-specific methods to generate adversarial examples to protect personal images, yet the effectiveness of existing methods is hindered by constrained adaptability to different prompts. In this paper, we introduce a Prompt-Agnostic Adversarial Perturbation (PAP) method for customized diffusion models. PAP first models the prompt distribution using a Laplace Approximation, and then produces prompt-agnostic perturbations by maximizing a disturbance expectation based on the modeled distribution. This approach effectively tackles the prompt-agnostic attacks, leading to improved defense stability. Extensive experiments in face privacy and artistic style protection, demonstrate the superior generalization of our method in comparison to existing techniques.
Authors: Haoran Tang, Meng Cao, Jinfa Huang, Ruyang Liu, Peng Jin, Ge Li, Xiaodan Liang
Abstract: Text-Video Retrieval (TVR) aims to align and associate relevant video content with corresponding natural language queries. Most existing TVR methods are based on large-scale pre-trained vision-language models (e.g., CLIP). However, due to the inherent plain structure of CLIP, few TVR methods explore the multi-scale representations which offer richer contextual information for a more thorough understanding. To this end, we propose MUSE, a multi-scale mamba with linear computational complexity for efficient cross-resolution modeling. Specifically, the multi-scale representations are generated by applying a feature pyramid on the last single-scale feature map. Then, we employ the Mamba structure as an efficient multi-scale learner to jointly learn scale-wise representations. Furthermore, we conduct comprehensive studies to investigate different model structures and designs. Extensive results on three popular benchmarks have validated the superiority of MUSE.
Authors: Lixin Yang, Licheng Zhong, Pengxiang Zhu, Xinyu Zhan, Junxiao Kong, Jian Xu, Cewu Lu
Abstract: This work introduces a novel and generalizable multi-view Hand Mesh Reconstruction (HMR) model, named POEM, designed for practical use in real-world hand motion capture scenarios. The advances of the POEM model consist of two main aspects. First, concerning the modeling of the problem, we propose embedding a static basis point within the multi-view stereo space. A point represents a natural form of 3D information and serves as an ideal medium for fusing features across different views, given its varied projections across these views. Consequently, our method harnesses a simple yet effective idea: a complex 3D hand mesh can be represented by a set of 3D basis points that 1) are embedded in the multi-view stereo, 2) carry features from the multi-view images, and 3) encompass the hand in it. The second advance lies in the training strategy. We utilize a combination of five large-scale multi-view datasets and employ randomization in the number, order, and poses of the cameras. By processing such a vast amount of data and a diverse array of camera configurations, our model demonstrates notable generalizability in the real-world applications. As a result, POEM presents a highly practical, plug-and-play solution that enables user-friendly, cost-effective multi-view motion capture for both left and right hands. The model and source codes are available at https://github.com/JubSteven/POEM-v2.
Authors: Zhijing Shao, Duotun Wang, Qing-Yao Tian, Yao-Dong Yang, Hengyu Meng, Zeyu Cai, Bo Dong, Yu Zhang, Kang Zhang, Zeyu Wang
Abstract: Although neural rendering has made significant advancements in creating lifelike, animatable full-body and head avatars, incorporating detailed expressions into full-body avatars remains largely unexplored. We present DEGAS, the first 3D Gaussian Splatting (3DGS)-based modeling method for full-body avatars with rich facial expressions. Trained on multiview videos of a given subject, our method learns a conditional variational autoencoder that takes both the body motion and facial expression as driving signals to generate Gaussian maps in the UV layout. To drive the facial expressions, instead of the commonly used 3D Morphable Models (3DMMs) in 3D head avatars, we propose to adopt the expression latent space trained solely on 2D portrait images, bridging the gap between 2D talking faces and 3D avatars. Leveraging the rendering capability of 3DGS and the rich expressiveness of the expression latent space, the learned avatars can be reenacted to reproduce photorealistic rendering images with subtle and accurate facial expressions. Experiments on an existing dataset and our newly proposed dataset of full-body talking avatars demonstrate the efficacy of our method. We also propose an audio-driven extension of our method with the help of 2D talking faces, opening new possibilities to interactive AI agents.
Authors: Yunxin Tang, Siyuan Tang, Jian Zhang, Hao Chen
Abstract: Background: Breast ultrasound is prominently used in diagnosing breast tumors. At present, many automatic systems based on deep learning have been developed to help radiologists in diagnosis. However, training such systems remains challenging because they are usually data-hungry and demand amounts of labeled data, which need professional knowledge and are expensive. Methods: We adopted a triplet network and a self-supervised contrastive learning technique to learn representations from unlabeled breast ultrasound video clips. We further designed a new hard triplet loss to to learn representations that particularly discriminate positive and negative image pairs that are hard to recognize. We also constructed a pretraining dataset from breast ultrasound videos (1,360 videos from 200 patients), which includes an anchor sample dataset with 11,805 images, a positive sample dataset with 188,880 images, and a negative sample dataset dynamically generated from video clips. Further, we constructed a finetuning dataset, including 400 images from 66 patients. We transferred the pretrained network to a downstream benign/malignant classification task and compared the performance with other state-of-the-art models, including three models pretrained on ImageNet and a previous contrastive learning model retrained on our datasets. Results and conclusion: Experiments revealed that our model achieved an area under the receiver operating characteristic curve (AUC) of 0.952, which is significantly higher than the others. Further, we assessed the dependence of our pretrained model on the number of labeled data and revealed that <100 samples were required to achieve an AUC of 0.901. The proposed framework greatly reduces the demand for labeled data and holds potential for use in automatic breast ultrasound image diagnosis.
Authors: Jintao Cheng, Xingming Chen, Jinxin Liang, Xiaoyu Tang, Xieyuanli Chen, Dachuan Li
Abstract: Effectively summarizing dense 3D point cloud data and extracting motion information of moving objects (moving object segmentation, MOS) is crucial to autonomous driving and robotics applications. How to effectively utilize motion and semantic features and avoid information loss during 3D-to-2D projection is still a key challenge. In this paper, we propose a novel multi-view MOS model (MV-MOS) by fusing motion-semantic features from different 2D representations of point clouds. To effectively exploit complementary information, the motion branches of the proposed model combines motion features from both bird's eye view (BEV) and range view (RV) representations. In addition, a semantic branch is introduced to provide supplementary semantic features of moving objects. Finally, a Mamba module is utilized to fuse the semantic features with motion features and provide effective guidance for the motion branches. We validated the effectiveness of the proposed multi-branch fusion MOS framework via comprehensive experiments, and our proposed model outperforms existing state-of-the-art models on the SemanticKITTI benchmark.
Authors: Yanbo Ding, Shaobin Zhuang, Kunchang Li, Zhengrong Yue, Yu Qiao, Yali Wang
Abstract: Despite recent advancements in text-to-image generation, most existing methods struggle to create images with multiple objects and complex spatial relationships in 3D world. To tackle this limitation, we introduce a generic AI system, namely MUSES, for 3D-controllable image generation from user queries. Specifically, our MUSES addresses this challenging task by developing a progressive workflow with three key components, including (1) Layout Manager for 2D-to-3D layout lifting, (2) Model Engineer for 3D object acquisition and calibration, (3) Image Artist for 3D-to-2D image rendering. By mimicking the collaboration of human professionals, this multi-modal agent pipeline facilitates the effective and automatic creation of images with 3D-controllable objects, through an explainable integration of top-down planning and bottom-up generation. Additionally, we find that existing benchmarks lack detailed descriptions of complex 3D spatial relationships of multiple objects. To fill this gap, we further construct a new benchmark of T2I-3DisBench (3D image scene), which describes diverse 3D image scenes with 50 detailed prompts. Extensive experiments show the state-of-the-art performance of MUSES on both T2I-CompBench and T2I-3DisBench, outperforming recent strong competitors such as DALL-E 3 and Stable Diffusion 3. These results demonstrate a significant step of MUSES forward in bridging natural language, 2D image generation, and 3D world.
Authors: Yuhang Zhang, Xiuqi Zheng, Chenyi Liang, Jiani Hu, Weihong Deng
Abstract: SOTA facial expression recognition (FER) methods fail on test sets that have domain gaps with the train set. Recent domain adaptation FER methods need to acquire labeled or unlabeled samples of target domains to fine-tune the FER model, which might be infeasible in real-world deployment. In this paper, we aim to improve the zero-shot generalization ability of FER methods on different unseen test sets using only one train set. Inspired by how humans first detect faces and then select expression features, we propose a novel FER pipeline to extract expression-related features from any given face images. Our method is based on the generalizable face features extracted by large models like CLIP. However, it is non-trivial to adapt the general features of CLIP for specific tasks like FER. To preserve the generalization ability of CLIP and the high precision of the FER model, we design a novel approach that learns sigmoid masks based on the fixed CLIP face features to extract expression features. To further improve the generalization ability on unseen test sets, we separate the channels of the learned masked features according to the expression classes to directly generate logits and avoid using the FC layer to reduce overfitting. We also introduce a channel-diverse loss to make the learned masks separated. Extensive experiments on five different FER datasets verify that our method outperforms SOTA FER methods by large margins. Code is available in https://github.com/zyh-uaiaaaa/Generalizable-FER.
Authors: Christoph Redies, Ralf Bartho, Lisa Ko{\ss}mann, Branka Spehar, Ronald H\"ubner, Johan Wagemans, Gregor U. Hayn-Leichsenring
Abstract: Over the past two decades, researchers in the field of visual aesthetics have studied numerous quantitative (objective) image properties and how they relate to visual aesthetic appreciation. However, results are difficult to compare between research groups. One reason is that researchers use different sets of image properties in their studies. But even if the same properties are used, the image pre-processing techniques may differ and often researchers use their own customized scripts to calculate the image properties. To provide greater accessibility and comparability of research results in visual experimental aesthetics, we developed an open-access and easy-to-use toolbox (called the 'Aesthetics Toolbox'). The Toolbox allows users to calculate a well-defined set of quantitative image properties popular in contemporary research. The properties include lightness and color statistics, Fourier spectral properties, fractality, self-similarity, symmetry, as well as different entropy measures and CNN-based variances. Compatible with most devices, the Toolbox provides an intuitive click-and-drop web interface. In the Toolbox, we integrated the original scripts of four different research groups and translated them into Python 3. To ensure that results were consistent across analyses, we took care that results from the Python versions of the scripts were the same as those from the original scripts. The toolbox, detailed documentation, and a link to the cloud version are available via Github: https://github.com/RBartho/Aesthetics-Toolbox. In summary, we developed a toolbox that helps to standardize and simplify the calculation of quantitative image properties for visual aesthetics research.
Authors: Andrew Kiruluta, Eric Lundy, Andreas Lemos
Abstract: Change detection is a crucial task in remote sensing, enabling the monitoring of environmental changes, urban growth, and disaster impact. Conventional change detection techniques, such as image differencing and ratioing, often struggle with noise and fail to capture complex variations in imagery. Recent advancements in machine learning, particularly generative models like diffusion models, offer new opportunities for enhancing change detection accuracy. In this paper, we propose a novel change detection framework that combines the strengths of Stable Diffusion models with the Structural Similarity Index (SSIM) to create robust and interpretable change maps. Our approach, named Diffusion Based Change Detector, is evaluated on both synthetic and real-world remote sensing datasets and compared with state-of-the-art methods. The results demonstrate that our method significantly outperforms traditional differencing techniques and recent deep learning-based methods, particularly in scenarios with complex changes and noise.
Authors: Tong Wang, Xiaochao Qu, Ting Liu
Abstract: Scene text editing aims to modify texts on images while maintaining the style of newly generated text similar to the original. Given an image, a target area, and target text, the task produces an output image with the target text in the selected area, replacing the original. This task has been studied extensively, with initial success using Generative Adversarial Networks (GANs) to balance text fidelity and style similarity. However, GAN-based methods struggled with complex backgrounds or text styles. Recent works leverage diffusion models, showing improved results, yet still face challenges, especially with non-Latin languages like CJK characters (Chinese, Japanese, Korean) that have complex glyphs, often producing inaccurate or unrecognizable characters. To address these issues, we present \emph{TextMastero} - a carefully designed multilingual scene text editing architecture based on latent diffusion models (LDMs). TextMastero introduces two key modules: a glyph conditioning module for fine-grained content control in generating accurate texts, and a latent guidance module for providing comprehensive style information to ensure similarity before and after editing. Both qualitative and quantitative experiments demonstrate that our method surpasses all known existing works in text fidelity and style similarity.
Authors: Yonggan Wu, Ling-Chao Meng, Yuan Zichao, Sixian Chan, Hong-Qiang Wang
Abstract: For the visible-infrared person re-identification (VI-ReID) task, one of the primary challenges lies in significant cross-modality discrepancy. Existing methods struggle to conduct modality-invariant information mining. They often focus solely on mining singular dimensions like spatial or channel, and overlook the extraction of specific-modality multi-dimension information. To fully mine modality-invariant information across a wide range, we introduce the Wide-Ranging Information Mining Network (WRIM-Net), which mainly comprises a Multi-dimension Interactive Information Mining (MIIM) module and an Auxiliary-Information-based Contrastive Learning (AICL) approach. Empowered by the proposed Global Region Interaction (GRI), MIIM comprehensively mines non-local spatial and channel information through intra-dimension interaction. Moreover, Thanks to the low computational complexity design, separate MIIM can be positioned in shallow layers, enabling the network to better mine specific-modality multi-dimension information. AICL, by introducing the novel Cross-Modality Key-Instance Contrastive (CMKIC) loss, effectively guides the network in extracting modality-invariant information. We conduct extensive experiments not only on the well-known SYSU-MM01 and RegDB datasets but also on the latest large-scale cross-modality LLCM dataset. The results demonstrate WRIM-Net's superiority over state-of-the-art methods.
Authors: Chen Liang, Qiang Guo, Xiaochao Qu, Luoqi Liu, Ting Liu
Abstract: Video segmentation aims at partitioning video sequences into meaningful segments based on objects or regions of interest within frames. Current video segmentation models are often derived from image segmentation techniques, which struggle to cope with small-scale or class-imbalanced video datasets. This leads to inconsistent segmentation results across frames. To address these issues, we propose a training strategy Masked Video Consistency, which enhances spatial and temporal feature aggregation. MVC introduces a training strategy that randomly masks image patches, compelling the network to predict the entire semantic segmentation, thus improving contextual information integration. Additionally, we introduce Object Masked Attention (OMA) to optimize the cross-attention mechanism by reducing the impact of irrelevant queries, thereby enhancing temporal modeling capabilities. Our approach, integrated into the latest decoupled universal video segmentation framework, achieves state-of-the-art performance across five datasets for three video segmentation tasks, demonstrating significant improvements over previous methods without increasing model parameters.
Authors: Yuxiao Wang, Qiwei Xiong, Yu Lei, Weiying Xue, Qi Liu, Zhenao Wei
Abstract: Human-object interaction (HOI) detection plays a key role in high-level visual understanding, facilitating a deep comprehension of human activities. Specifically, HOI detection aims to locate the humans and objects involved in interactions within images or videos and classify the specific interactions between them. The success of this task is influenced by several key factors, including the accurate localization of human and object instances, as well as the correct classification of object categories and interaction relationships. This paper systematically summarizes and discusses the recent work in image-based HOI detection. First, the mainstream datasets involved in HOI relationship detection are introduced. Furthermore, starting with two-stage methods and end-to-end one-stage detection approaches, this paper comprehensively discusses the current developments in image-based HOI detection, analyzing the strengths and weaknesses of these two methods. Additionally, the advancements of zero-shot learning, weakly supervised learning, and the application of large-scale language models in HOI detection are discussed. Finally, the current challenges in HOI detection are outlined, and potential research directions and future trends are explored.
Authors: Guofeng Mei, Luigi Riz, Yiming Wang, Fabio Poiesi
Abstract: Most recent 3D instance segmentation methods are open vocabulary, offering a greater flexibility than closed-vocabulary methods. Yet, they are limited to reasoning within a specific set of concepts, \ie the vocabulary, prompted by the user at test time. In essence, these models cannot reason in an open-ended fashion, i.e., answering ``List the objects in the scene.''. We introduce the first method to address 3D instance segmentation in a setting that is void of any vocabulary prior, namely a vocabulary-free setting. We leverage a large vision-language assistant and an open-vocabulary 2D instance segmenter to discover and ground semantic categories on the posed images. To form 3D instance mask, we first partition the input point cloud into dense superpoints, which are then merged into 3D instance masks. We propose a novel superpoint merging strategy via spectral clustering, accounting for both mask coherence and semantic coherence that are estimated from the 2D object instance masks. We evaluate our method using ScanNet200 and Replica, outperforming existing methods in both vocabulary-free and open-vocabulary settings. Code will be made available.
Authors: Yingtie Lei, Jia Yu, Yihang Dong, Changwei Gong, Ziyang Zhou, Chi-Man Pun
Abstract: Underwater image enhancement (UIE) plays a crucial role in various marine applications, but it remains challenging due to the complex underwater environment. Current learning-based approaches frequently lack explicit incorporation of prior knowledge about the physical processes involved in underwater image formation, resulting in limited optimization despite their impressive enhancement results. This paper proposes a novel deep unfolding network (DUN) for UIE that integrates color priors and inter-stage feature transformation to improve enhancement performance. The proposed DUN model combines the iterative optimization and reliability of model-based methods with the flexibility and representational power of deep learning, offering a more explainable and stable solution compared to existing learning-based UIE approaches. The proposed model consists of three key components: a Color Prior Guidance Block (CPGB) that establishes a mapping between color channels of degraded and original images, a Nonlinear Activation Gradient Descent Module (NAGDM) that simulates the underwater image degradation process, and an Inter Stage Feature Transformer (ISF-Former) that facilitates feature exchange between different network stages. By explicitly incorporating color priors and modeling the physical characteristics of underwater image formation, the proposed DUN model achieves more accurate and reliable enhancement results. Extensive experiments on multiple underwater image datasets demonstrate the superiority of the proposed model over state-of-the-art methods in both quantitative and qualitative evaluations. The proposed DUN-based approach offers a promising solution for UIE, enabling more accurate and reliable scientific analysis in marine research. The code is available at https://github.com/CXH-Research/UIE-UnFold.
Authors: Deyu Li, Longfei Xiao, Handi Wei, Yan Li, Binghua Zhang
Abstract: The accurate measurement of the wave field and its spatiotemporal evolution is essential in many hydrodynamic experiments and engineering applications. The binocular stereo imaging technique has been widely used to measure waves. However, the optical properties of indoor water surfaces, including transparency, specular reflection, and texture absence, pose challenges for image processing and stereo reconstruction. This study proposed a novel technique that combined thermal stereography and deep learning to achieve fully noncontact wave measurements. The optical imaging properties of water in the long-wave infrared spectrum were found to be suitable for stereo matching, effectively avoiding the issues in the visible-light spectrum. After capturing wave images using thermal stereo cameras, a reconstruction strategy involving deep learning techniques was proposed to improve stereo matching performance. A generative approach was employed to synthesize a dataset with ground-truth disparity from unannotated infrared images. This dataset was then fed to a pretrained stereo neural network for fine-tuning to achieve domain adaptation. Wave flume experiments were conducted to validate the feasibility and accuracy of the proposed technique. The final reconstruction results indicated great agreement and high accuracy with a mean bias of less than 2.1% compared with the measurements obtained using wave probes, suggesting that the novel technique effectively measures the spatiotemporal distribution of wave surface in hydrodynamic experiments.
Authors: Shuning Xu, Xina Liu, Binbin Song, Xiangyu Chen, Qiubo Chen, Jiantao Zhou
Abstract: Moire patterns arise when two similar repetitive patterns interfere, a phenomenon frequently observed during the capture of images or videos on screens. The color, shape, and location of moire patterns may differ across video frames, posing a challenge in learning information from adjacent frames and preserving temporal consistency. Previous video demoireing methods heavily rely on well-designed alignment modules, resulting in substantial computational burdens. Recently, Mamba, an improved version of the State Space Model (SSM), has demonstrated significant potential for modeling long-range dependencies with linear complexity, enabling efficient temporal modeling in video demoireing without requiring a specific alignment module. In this paper, we propose a novel alignment-free Raw video demoireing network with frequency-assisted spatio-temporal Mamba (DemMamba). The Spatial Mamba Block (SMB) and Temporal Mamba Block (TMB) are sequentially arranged to facilitate effective intra- and inter-relationship modeling in Raw videos with moire patterns. Within SMB, an Adaptive Frequency Block (AFB) is introduced to aid demoireing in the frequency domain. For TMB, a Channel Attention Block (CAB) is embedded to further enhance temporal information interactions by exploiting the inter-channel relationships among features. Extensive experiments demonstrate that our proposed DemMamba surpasses state-of-the-art approaches by 1.3 dB and delivers a superior visual experience.
Authors: Bin Wang, Wenqian Wang
Abstract: Recently, large-scale pre-trained vision-language models (e.g., CLIP), have garnered significant attention thanks to their powerful representative capabilities. This inspires researchers in transferring the knowledge from these large pre-trained models to other task-specific models, e.g., Video Action Recognition (VAR) models, via particularly leveraging side networks to enhance the efficiency of parameter-efficient fine-tuning (PEFT). However, current transferring approaches in VAR tend to directly transfer the frozen knowledge from large pre-trained models to action recognition networks with minimal cost, instead of exploiting the temporal modeling capabilities of the action recognition models themselves. Therefore, in this paper, we propose a memory-efficient Temporal Difference Side Network (TDS-CLIP) to balance knowledge transferring and temporal modeling, avoiding backpropagation in frozen parameter models. Specifically, we introduce a Temporal Difference Adapter (TD-Adapter), which can effectively capture local temporal differences in motion features to strengthen the model's global temporal modeling capabilities. Furthermore, we designed a Side Motion Enhancement Adapter (SME-Adapter) to guide the proposed side network in efficiently learning the rich motion information in videos, thereby improving the side network's ability to capture and learn motion information. Extensive experiments are conducted on three benchmark datasets, including Something-Something V1\&V2, and Kinetics-400. Experimental results demonstrate that our approach achieves competitive performance.
Authors: Huafeng Qin, Yuming Fu, Huiyan Zhang, Mounim A. El-Yacoubi, Xinbo Gao, Qun Song, Jun Wang
Abstract: Deep neural networks have recently achieved promising performance in the vein recognition task and have shown an increasing application trend, however, they are prone to adversarial perturbation attacks by adding imperceptible perturbations to the input, resulting in making incorrect recognition. To address this issue, we propose a novel defense model named MsMemoryGAN, which aims to filter the perturbations from adversarial samples before recognition. First, we design a multi-scale autoencoder to achieve high-quality reconstruction and two memory modules to learn the detailed patterns of normal samples at different scales. Second, we investigate a learnable metric in the memory module to retrieve the most relevant memory items to reconstruct the input image. Finally, the perceptional loss is combined with the pixel loss to further enhance the quality of the reconstructed image. During the training phase, the MsMemoryGAN learns to reconstruct the input by merely using fewer prototypical elements of the normal patterns recorded in the memory. At the testing stage, given an adversarial sample, the MsMemoryGAN retrieves its most relevant normal patterns in memory for the reconstruction. Perturbations in the adversarial sample are usually not reconstructed well, resulting in purifying the input from adversarial perturbations. We have conducted extensive experiments on two public vein datasets under different adversarial attack methods to evaluate the performance of the proposed approach. The experimental results show that our approach removes a wide variety of adversarial perturbations, allowing vein classifiers to achieve the highest recognition accuracy.
Authors: Mingrui Ma, Weijie Wang, Jie Ning, Jianfeng He, Nicu Sebe, Bruno Lepri
Abstract: The challenge of Multimodal Deformable Image Registration (MDIR) lies in the conversion and alignment of features between images of different modalities. Generative models (GMs) cannot retain the necessary information enough from the source modality to the target one, while non-GMs struggle to align features across these two modalities. In this paper, we propose a novel coarse-to-fine MDIR framework,LLM-Morph, which is applicable to various pre-trained Large Language Models (LLMs) to solve these concerns by aligning the deep features from different modal medical images. Specifically, we first utilize a CNN encoder to extract deep visual features from cross-modal image pairs, then we use the first adapter to adjust these tokens, and use LoRA in pre-trained LLMs to fine-tune their weights, both aimed at eliminating the domain gap between the pre-trained LLMs and the MDIR task. Third, for the alignment of tokens, we utilize other four adapters to transform the LLM-encoded tokens into multi-scale visual features, generating multi-scale deformation fields and facilitating the coarse-to-fine MDIR task. Extensive experiments in MR-CT Abdomen and SR-Reg Brain datasets demonstrate the effectiveness of our framework and the potential of pre-trained LLMs for MDIR task. Our code is availabel at: https://github.com/ninjannn/LLM-Morph.
Authors: Pengkun Wei, Shuo Cheng, Dayou Li, Ran Song, Yipeng Zhang, Wei Zhang
Abstract: Efficiently detecting target weld seams while ensuring sub-millimeter accuracy has always been an important challenge in autonomous welding, which has significant application in industrial practice. Previous works mostly focused on recognizing and localizing welding seams one by one, leading to inferior efficiency in modeling the workpiece. This paper proposes a novel framework capable of multiple weld seams extraction using both RGB images and 3D point clouds. The RGB image is used to obtain the region of interest by approximately localizing the weld seams, and the point cloud is used to achieve the fine-edge extraction of the weld seams within the region of interest using region growth. Our method is further accelerated by using a pre-trained deep learning model to ensure both efficiency and generalization ability. The performance of the proposed method has been comprehensively tested on various workpieces featuring both linear and curved weld seams and in physical experiment systems. The results showcase considerable potential for real-world industrial applications, emphasizing the method's efficiency and effectiveness. Videos of the real-world experiments can be found at https://youtu.be/pq162HSP2D4.
Authors: Jinjie Mai, Wenxuan Zhu, Sara Rojas, Jesus Zarzar, Abdullah Hamdi, Guocheng Qian, Bing Li, Silvio Giancola, Bernard Ghanem
Abstract: Neural radiance fields (NeRFs) generally require many images with accurate poses for accurate novel view synthesis, which does not reflect realistic setups where views can be sparse and poses can be noisy. Previous solutions for learning NeRFs with sparse views and noisy poses only consider local geometry consistency with pairs of views. Closely following \textit{bundle adjustment} in Structure-from-Motion (SfM), we introduce TrackNeRF for more globally consistent geometry reconstruction and more accurate pose optimization. TrackNeRF introduces \textit{feature tracks}, \ie connected pixel trajectories across \textit{all} visible views that correspond to the \textit{same} 3D points. By enforcing reprojection consistency among feature tracks, TrackNeRF encourages holistic 3D consistency explicitly. Through extensive experiments, TrackNeRF sets a new benchmark in noisy and sparse view reconstruction. In particular, TrackNeRF shows significant improvements over the state-of-the-art BARF and SPARF by $\sim8$ and $\sim1$ in terms of PSNR on DTU under various sparse and noisy view setups. The code is available at \href{https://tracknerf.github.io/}.
Authors: Huafeng Chen, Pengxu Wei, Guangqian Guo, Shan Gao
Abstract: Most Camouflaged Object Detection (COD) methods heavily rely on mask annotations, which are time-consuming and labor-intensive to acquire. Existing weakly-supervised COD approaches exhibit significantly inferior performance compared to fully-supervised methods and struggle to simultaneously support all the existing types of camouflaged object labels, including scribbles, bounding boxes, and points. Even for Segment Anything Model (SAM), it is still problematic to handle the weakly-supervised COD and it typically encounters challenges of prompt compatibility of the scribble labels, extreme response, semantically erroneous response, and unstable feature representations, producing unsatisfactory results in camouflaged scenes. To mitigate these issues, we propose a unified COD framework in this paper, termed SAM-COD, which is capable of supporting arbitrary weakly-supervised labels. Our SAM-COD employs a prompt adapter to handle scribbles as prompts based on SAM. Meanwhile, we introduce response filter and semantic matcher modules to improve the quality of the masks obtained by SAM under COD prompts. To alleviate the negative impacts of inaccurate mask predictions, a new strategy of prompt-adaptive knowledge distillation is utilized to ensure a reliable feature representation. To validate the effectiveness of our approach, we have conducted extensive empirical experiments on three mainstream COD benchmarks. The results demonstrate the superiority of our method against state-of-the-art weakly-supervised and even fully-supervised methods.
Authors: Antoine P. Sanner, Nils F. Grauhan, Marc A. Brockmann, Ahmed E. Othman, Anirban Mukhopadhyay
Abstract: Whole-body CT is used for multi-trauma patients in the search of any and all injuries. Since an initial assessment needs to be rapid and the search for lesions is done for the whole body, very little time can be allocated for the inspection of a specific anatomy. In particular, intracranial hemorrhages are still missed, especially by clinical students. In this work, we present a Deep Learning approach for highlighting such lesions to improve the diagnostic accuracy. While most works on intracranial hemorrhages perform segmentation, detection only requires bounding boxes for the localization of the bleeding. In this paper, we propose a novel Voxel-Complete IoU (VC-IoU) loss that encourages the network to learn the 3D aspect ratios of bounding boxes and leads to more precise detections. We extensively experiment on brain bleeding detection using a publicly available dataset, and validate it on a private cohort, where we achieve 0.877 AR30, 0.728 AP30, and 0.653 AR30, 0.514 AP30 respectively. These results constitute a relative +5% improvement in Average Recall for both datasets compared to other loss functions. Finally, as there is little data currently publicly available for 3D object detection and as annotation resources are limited in the clinical setting, we evaluate the cost of different annotation methods, as well as the impact of imprecise bounding boxes in the training data on the detection performance.
Authors: Hans Aoyang Zhou, Dominik Wolfschl\"ager, Constantinos Florides, Jonas Werheid, Hannes Behnen, Jan-Henrick Woltersmann, Tiago C. Pinto, Marco Kemmerling, Anas Abdelrazeq, Robert H. Schmitt
Abstract: Machine vision enhances automation, quality control, and operational efficiency in industrial applications by enabling machines to interpret and act on visual data. While traditional computer vision algorithms and approaches remain widely utilized, machine learning has become pivotal in current research activities. In particular, generative AI demonstrates promising potential by improving pattern recognition capabilities, through data augmentation, increasing image resolution, and identifying anomalies for quality control. However, the application of generative AI in machine vision is still in its early stages due to challenges in data diversity, computational requirements, and the necessity for robust validation methods. A comprehensive literature review is essential to understand the current state of generative AI in industrial machine vision, focusing on recent advancements, applications, and research trends. Thus, a literature review based on the PRISMA guidelines was conducted, analyzing over 1,200 papers on generative AI in industrial machine vision. Our findings reveal various patterns in current research, with the primary use of generative AI being data augmentation, for machine vision tasks such as classification and object detection. Furthermore, we gather a collection of application challenges together with data requirements to enable a successful application of generative AI in industrial machine vision. This overview aims to provide researchers with insights into the different areas and applications within current research, highlighting significant advancements and identifying opportunities for future work.
Authors: Huafeng Chen, Dian Shao, Guangqian Guo, Shan Gao
Abstract: Camouflaged Object Detection (COD) demands models to expeditiously and accurately distinguish objects which conceal themselves seamlessly in the environment. Owing to the subtle differences and ambiguous boundaries, COD is not only a remarkably challenging task for models but also for human annotators, requiring huge efforts to provide pixel-wise annotations. To alleviate the heavy annotation burden, we propose to fulfill this task with the help of only one point supervision. Specifically, by swiftly clicking on each object, we first adaptively expand the original point-based annotation to a reasonable hint area. Then, to avoid partial localization around discriminative parts, we propose an attention regulator to scatter model attention to the whole object through partially masking labeled regions. Moreover, to solve the unstable feature representation of camouflaged objects under only point-based annotation, we perform unsupervised contrastive learning based on differently augmented image pairs (e.g. changing color or doing translation). On three mainstream COD benchmarks, experimental results show that our model outperforms several weakly-supervised methods by a large margin across various metrics.
Authors: Binta Sow, Bilal Faye, Hanane Azzag, Mustapha Lebbah
Abstract: Object detection in computer vision traditionally involves identifying objects in images. By integrating textual descriptions, we enhance this process, providing better context and accuracy. The MDETR model significantly advances this by combining image and text data for more versatile object detection and classification. However, MDETR's complexity and high computational demands hinder its practical use. In this paper, we introduce Lightweight MDETR (LightMDETR), an optimized MDETR variant designed for improved computational efficiency while maintaining robust multimodal capabilities. Our approach involves freezing the MDETR backbone and training a sole component, the Deep Fusion Encoder (DFE), to represent image and text modalities. A learnable context vector enables the DFE to switch between these modalities. Evaluation on datasets like RefCOCO, RefCOCO+, and RefCOCOg demonstrates that LightMDETR achieves superior precision and accuracy.
Authors: Zhirui Gao, Renjiao Yi, Yuhang Huang, Wei Chen, Chenyang Zhu, Kai Xu
Abstract: Low-level 3D representations, such as point clouds, meshes, NeRFs, and 3D Gaussians, are commonly used to represent 3D objects or scenes. However, humans usually perceive 3D objects or scenes at a higher level as a composition of parts or structures rather than points or voxels. Representing 3D as semantic parts can benefit further understanding and applications. We aim to solve part-aware 3D reconstruction, which parses objects or scenes into semantic parts. In this paper, we introduce a hybrid representation of superquadrics and 2D Gaussians, trying to dig 3D structural clues from multi-view image inputs. Accurate structured geometry reconstruction and high-quality rendering are achieved at the same time. We incorporate parametric superquadrics in mesh forms into 2D Gaussians by attaching Gaussian centers to faces in meshes. During the training, superquadrics parameters are iteratively optimized, and Gaussians are deformed accordingly, resulting in an efficient hybrid representation. On the one hand, this hybrid representation inherits the advantage of superquadrics to represent different shape primitives, supporting flexible part decomposition of scenes. On the other hand, 2D Gaussians are incorporated to model the complex texture and geometry details, ensuring high-quality rendering and geometry reconstruction. The reconstruction is fully unsupervised. We conduct extensive experiments on data from DTU and ShapeNet datasets, in which the method decomposes scenes into reasonable parts, outperforming existing state-of-the-art approaches.
Authors: Malsha Ashani Mahawatta Dona, Beatriz Cabrero-Daniel, Yinan Yu, Christian Berger
Abstract: Today's advanced automotive systems are turning into intelligent Cyber-Physical Systems (CPS), bringing computational intelligence to their cyber-physical context. Such systems power advanced driver assistance systems (ADAS) that observe a vehicle's surroundings for their functionality. However, such ADAS have clear limitations in scenarios when the direct line-of-sight to surrounding objects is occluded, like in urban areas. Imagine now automated driving (AD) systems that ideally could benefit from other vehicles' field-of-view in such occluded situations to increase traffic safety if, for example, locations about pedestrians can be shared across vehicles. Current literature suggests vehicle-to-infrastructure (V2I) via roadside units (RSUs) or vehicle-to-vehicle (V2V) communication to address such issues that stream sensor or object data between vehicles. When considering the ongoing revolution in vehicle system architectures towards powerful, centralized processing units with hardware accelerators, foreseeing the onboard presence of large language models (LLMs) to improve the passengers' comfort when using voice assistants becomes a reality. We are suggesting and evaluating a concept to complement the ego vehicle's field-of-view (FOV) with another vehicle's FOV by tapping into their onboard LLM to let the machines have a dialogue about what the other vehicle ``sees''. Our results show that very recent versions of LLMs, such as GPT-4V and GPT-4o, understand a traffic situation to an impressive level of detail, and hence, they can be used even to spot traffic participants. However, better prompts are needed to improve the detection quality and future work is needed towards a standardised message interchange format between vehicles.
Authors: Seyed Abolfazl Ghasemzadeh, Alexandre Alahi, Christophe De Vleeschouwer
Abstract: Estimating 3D human poses from 2D images is challenging due to occlusions and projective acquisition. Learning-based approaches have been largely studied to address this challenge, both in single and multi-view setups. These solutions however fail to generalize to real-world cases due to the lack of (multi-view) 'in-the-wild' images paired with 3D poses for training. For this reason, we propose combining 2D pose estimation, for which large and rich training datasets exist, and 2D-to-3D pose lifting, using a transformer-based network that can be trained from synthetic 2D-3D pose pairs. Our experiments demonstrate decreases up to 45% in MPJPE errors compared to the 3D pose obtained by triangulating the 2D poses. The framework's source code is available at https://github.com/aghasemzadeh/OpenMPL .
Authors: Yutian Han, Baoxiang Huang, He Gao
Abstract: Lakes provide a wide range of valuable ecosystem services, such as water supply, biodiversity habitats, and carbon sequestration. However, lakes are increasingly threatened by climate change and human activities. Therefore, continuous global monitoring of lake dynamics is crucial, but remains challenging on a large scale. The recently developed Global Lakes Area Database (GLAKES) has mapped over 3.4 million lakes worldwide, but it only provides data at decadal intervals, which may be insufficient to capture rapid or short-term changes.This paper introduces an expanded lake database, GLAKES-Additional, which offers biennial delineations and area measurements for 152,567 lakes globally from 1990 to 2021. We employed the Swin-Unet model, replacing traditional convolution operations, to effectively address the challenges posed by the receptive field requirements of high spatial resolution satellite imagery. The increased biennial time resolution helps to quantitatively attribute lake area changes to climatic and hydrological drivers, such as precipitation and temperature changes.For predicting lake area changes, we used a Long Short-Term Memory (LSTM) neural network and an extended time series dataset for preliminary modeling. Under climate and land use scenarios, our model achieved an RMSE of 0.317 km^2 in predicting future lake area changes.
Authors: Sandra Bergmann, Denise Moussa, Christian Riess
Abstract: Image-based biometrics can aid law enforcement in various aspects, for example in iris, fingerprint and soft-biometric recognition. A critical precondition for recognition is the availability of sufficient biometric information in images. It is visually apparent that strong JPEG compression removes such details. However, latest AI-based image compression seemingly preserves many image details even for very strong compression factors. Yet, these perceived details are not necessarily grounded in measurements, which raises the question whether these images can still be used for biometric recognition. In this work, we investigate how AI compression impacts iris, fingerprint and soft-biometric (fabrics and tattoo) images. We also investigate the recognition performance for iris and fingerprint images after AI compression. It turns out that iris recognition can be strongly affected, while fingerprint recognition is quite robust. The loss of detail is qualitatively best seen in fabrics and tattoos images. Overall, our results show that AI-compression still permits many biometric tasks, but attention to strong compression factors in sensitive tasks is advisable.
Authors: Elia Bonetto, Aamir Ahmad
Abstract: Synthetic data is increasingly being used to address the lack of labeled images in uncommon domains for deep learning tasks. A prominent example is 2D pose estimation of animals, particularly wild species like zebras, for which collecting real-world data is complex and impractical. However, many approaches still require real images, consistency and style constraints, sophisticated animal models, and/or powerful pre-trained networks to bridge the syn-to-real gap. Moreover, they often assume that the animal can be reliably detected in images or videos, a hypothesis that often does not hold, e.g. in wildlife scenarios or aerial images. To solve this, we use synthetic data generated with a 3D photorealistic simulator to obtain the first synthetic dataset that can be used for both detection and 2D pose estimation of zebras without applying any of the aforementioned bridging strategies. Unlike previous works, we extensively train and benchmark our detection and 2D pose estimation models on multiple real-world and synthetic datasets using both pre-trained and non-pre-trained backbones. These experiments show how the models trained from scratch and only with synthetic data can consistently generalize to real-world images of zebras in both tasks. Moreover, we show it is possible to easily generalize those same models to 2D pose estimation of horses with a minimal amount of real-world images to account for the domain transfer. Code, results, trained models; and the synthetic, training, and validation data, including 104K manually labeled frames, are provided as open-source at https://zebrapose.is.tue.mpg.de/
Authors: Julius Pesonen, Teemu Hakala, V\"ain\"o Karjalainen, Niko Koivum\"aki, Lauri Markelin, Anna-Maria Raita-Hakola, Juha Suomalainen, Ilkka P\"ol\"onen, Eija Honkavaara
Abstract: Early detection of wildfires is essential to prevent large-scale fires resulting in extensive environmental, structural, and societal damage. Uncrewed aerial vehicles (UAVs) can cover large remote areas effectively with quick deployment requiring minimal infrastructure and equipping them with small cameras and computers enables autonomous real-time detection. In remote areas, however, the UAVs are limited to on-board computing for detection due to the lack of high-bandwidth mobile networks. This limits the detection to methods which are light enough for the on-board computer alone. For accurate camera-based localisation, segmentation of the detected smoke is essential but training data for deep learning-based wildfire smoke segmentation is limited. This study shows how small specialised segmentation models can be trained using only bounding box labels, leveraging zero-shot foundation model supervision. The method offers the advantages of needing only fairly easily obtainable bounding box labels and requiring training solely for the smaller student network. The proposed method achieved 63.3% mIoU on a manually annotated and diverse wildfire dataset. The used model can perform in real-time at ~11 fps with a UAV-carried NVIDIA Jetson Orin NX computer while reliably recognising smoke, demonstrated at real-world forest burning events. Code is available at https://gitlab.com/fgi_nls/public/wildfire-real-time-segmentation
URLs: https://gitlab.com/fgi_nls/public/wildfire-real-time-segmentation
Authors: Ombretta Strafforello, Osman S. Kayhan, Oana Inel, Klamer Schutte, Jan van Gemert
Abstract: Previous work shows that humans tend to prefer large bounding boxes over small bounding boxes with the same IoU. However, we show here that commonly used object detectors predict large and small boxes equally often. In this work, we investigate how to align automatically detected object boxes with human preference and study whether this improves human quality perception. We evaluate the performance of three commonly used object detectors through a user study (N = 123). We find that humans prefer object detections that are upscaled with factors of 1.5 or 2, even if the corresponding AP is close to 0. Motivated by this result, we propose an asymmetric bounding box regression loss that encourages large over small predicted bounding boxes. Our evaluation study shows that object detectors fine-tuned with the asymmetric loss are better aligned with human preference and are preferred over fixed scaling factors. A qualitative evaluation shows that human preference might be influenced by some object characteristics, like object shape.
Authors: Hidehisa Arai, Keita Miwa, Kento Sasaki, Yu Yamaguchi, Kohei Watanabe, Shunsuke Aoki, Issei Yamamoto
Abstract: Autonomous driving, particularly navigating complex and unanticipated scenarios, demands sophisticated reasoning and planning capabilities. While Multi-modal Large Language Models (MLLMs) offer a promising avenue for this, their use has been largely confined to understanding complex environmental contexts or generating high-level driving commands, with few studies extending their application to end-to-end path planning. A major research bottleneck is the lack of large-scale annotated datasets encompassing vision, language, and action. To address this issue, we propose CoVLA (Comprehensive Vision-Language-Action) Dataset, an extensive dataset comprising real-world driving videos spanning more than 80 hours. This dataset leverages a novel, scalable approach based on automated data processing and a caption generation pipeline to generate accurate driving trajectories paired with detailed natural language descriptions of driving environments and maneuvers. This approach utilizes raw in-vehicle sensor data, allowing it to surpass existing datasets in scale and annotation richness. Using CoVLA, we investigate the driving capabilities of MLLMs that can handle vision, language, and action in a variety of driving scenarios. Our results illustrate the strong proficiency of our model in generating coherent language and action outputs, emphasizing the potential of Vision-Language-Action (VLA) models in the field of autonomous driving. This dataset establishes a framework for robust, interpretable, and data-driven autonomous driving systems by providing a comprehensive platform for training and evaluating VLA models, contributing to safer and more reliable self-driving vehicles. The dataset is released for academic purpose.
Authors: Eito Ikuta, Yohan Lee, Akihiro Iohara, Yu Saito, Toshiyuki Tanaka
Abstract: Extracting geometry features from photographic images independently of surface texture and transferring them onto different materials remains a complex challenge. In this study, we introduce Harmonizing Attention, a novel training-free approach that leverages diffusion models for texture-aware geometry transfer. Our method employs a simple yet effective modification of self-attention layers, allowing the model to query information from multiple reference images within these layers. This mechanism is seamlessly integrated into the inversion process as Texture-aligning Attention and into the generation process as Geometry-aligning Attention. This dual-attention approach ensures the effective capture and transfer of material-independent geometry features while maintaining material-specific textural continuity, all without the need for model fine-tuning.
Authors: Yihao Huang, Le Liang, Tianlin Li, Xiaojun Jia, Run Wang, Weikai Miao, Geguang Pu, Yang Liu
Abstract: In recent years, Text-to-Image (T2I) models have garnered significant attention due to their remarkable advancements. However, security concerns have emerged due to their potential to generate inappropriate or Not-Safe-For-Work (NSFW) images. In this paper, inspired by the observation that texts with different semantics can lead to similar human perceptions, we propose an LLM-driven perception-guided jailbreak method, termed PGJ. It is a black-box jailbreak method that requires no specific T2I model (model-free) and generates highly natural attack prompts. Specifically, we propose identifying a safe phrase that is similar in human perception yet inconsistent in text semantics with the target unsafe word and using it as a substitution. The experiments conducted on six open-source models and commercial online services with thousands of prompts have verified the effectiveness of PGJ.
Authors: Natchapon Jongwiriyanurak, Zichao Zeng, June Moh Goo, Xinglei Wang, Ilya Ilyankou, Kerkritt Srirrongvikrai, Meihui Wang, James Haworth
Abstract: Road traffic crashes cause millions of deaths annually and have a significant economic impact, particularly in low- and middle-income countries (LMICs). This paper presents an approach using Vision Language Models (VLMs) for road safety assessment, overcoming the limitations of traditional Convolutional Neural Networks (CNNs). We introduce a new task ,V-RoAst (Visual question answering for Road Assessment), with a real-world dataset. Our approach optimizes prompt engineering and evaluates advanced VLMs, including Gemini-1.5-flash and GPT-4o-mini. The models effectively examine attributes for road assessment. Using crowdsourced imagery from Mapillary, our scalable solution influentially estimates road safety levels. In addition, this approach is designed for local stakeholders who lack resources, as it does not require training data. It offers a cost-effective and automated methods for global road safety assessments, potentially saving lives and reducing economic burdens.
Authors: Xinlong Cheng, Lei Li
Abstract: The capability for open vocabulary perception represents a significant advancement in autonomous driving systems, facilitating the comprehension and interpretation of a wide array of textual inputs in real-time. Despite extensive research in open vocabulary tasks within 2D computer vision, the application of such methodologies to 3D environments, particularly within large-scale outdoor contexts, remains relatively underdeveloped. This paper presents a novel approach that integrates 3D point cloud data, acquired from LIDAR sensors, with textual information. The primary focus is on the utilization of textual data to directly localize and identify objects within the autonomous driving context. We introduce an efficient framework for the fusion of bird's-eye view (BEV) region features with textual features, thereby enabling the system to seamlessly adapt to novel textual inputs and enhancing the robustness of open vocabulary detection tasks. The effectiveness of the proposed methodology is rigorously evaluated through extensive experimentation on the newly introduced NuScenes-T dataset, with additional validation of its zero-shot performance on the Lyft Level 5 dataset. This research makes a substantive contribution to the advancement of autonomous driving technologies by leveraging multimodal data to enhance open vocabulary perception in 3D environments, thereby pushing the boundaries of what is achievable in autonomous navigation and perception.
Authors: Tomoyasu Nanaumi, Kazuhiko Kawamoto, Hiroshi Kera
Abstract: To make an employee roster, photo album, or training dataset of generative models, one needs to collect high-quality images while dismissing low-quality ones. This study addresses a new task of unsupervised detection of low-quality images. We propose a method that not only detects low-quality images with various types of degradation but also provides visual clues of them based on an observation that partial reconstruction by hierarchical variational autoencoders fails for low-quality images. The experiments show that our method outperforms several unsupervised out-of-distribution detection methods and also gives visual clues for low-quality images that help humans recognize them even in thumbnail view.
Authors: Shengzhu Yang, Jiawei Du, Jia Guo, Weihang Zhang, Hanruo Liu, Huiqi Li, Ningli Wang
Abstract: Subtle semantic differences in retinal image and text data present great challenges for pre-training visual-language models. Moreover, false negative samples, i.e., image-text pairs having the same semantics but incorrectly regarded as negatives, disrupt the visual-language pre-training process and affect the model's learning ability. This work aims to develop a retinal foundation model, called ViLReF, by pre-training on a paired dataset comprising 451,956 retinal images and corresponding diagnostic text reports. In our vision-language pre-training strategy, we leverage expert knowledge to facilitate the extraction of labels and propose a novel constraint, the Weighted Similarity Coupling Loss, to adjust the speed of pushing sample pairs further apart dynamically within the feature space. Furthermore, we employ a batch expansion module with dynamic memory queues, maintained by momentum encoders, to supply extra samples and compensate for the vacancies caused by eliminating false negatives. Extensive experiments are conducted on multiple datasets for downstream classification and segmentation tasks. The experimental results demonstrate the powerful zero-shot and transfer learning capabilities of ViLReF, verifying the effectiveness of our pre-training strategy. Our ViLReF model is available at: https://github.com/T6Yang/ViLReF.
Authors: Zhongliang Guo, Lei Fang, Jingyu Lin, Yifei Qian, Shuai Zhao, Zeyu Wang, Junhao Dong, Cunjian Chen, Ognjen Arandjelovi\'c, Chun Pong Lau
Abstract: Recent advancements in generative AI, particularly Latent Diffusion Models (LDMs), have revolutionized image synthesis and manipulation. However, these generative techniques raises concerns about data misappropriation and intellectual property infringement. Adversarial attacks on machine learning models have been extensively studied, and a well-established body of research has extended these techniques as a benign metric to prevent the underlying misuse of generative AI. Current approaches to safeguarding images from manipulation by LDMs are limited by their reliance on model-specific knowledge and their inability to significantly degrade semantic quality of generated images. In response to these shortcomings, we propose the Posterior Collapse Attack (PCA) based on the observation that VAEs suffer from posterior collapse during training. Our method minimizes dependence on the white-box information of target models to get rid of the implicit reliance on model-specific knowledge. By accessing merely a small amount of LDM parameters, in specific merely the VAE encoder of LDMs, our method causes a substantial semantic collapse in generation quality, particularly in perceptual consistency, and demonstrates strong transferability across various model architectures. Experimental results show that PCA achieves superior perturbation effects on image generation of LDMs with lower runtime and VRAM. Our method outperforms existing techniques, offering a more robust and generalizable solution that is helpful in alleviating the socio-technical challenges posed by the rapidly evolving landscape of generative AI.
Authors: Qi Ma, Yue Li, Bin Ren, Nicu Sebe, Ender Konukoglu, Theo Gevers, Luc Van Gool, Danda Pani Paudel
Abstract: 3D Gaussian Splatting (3DGS) has become the de facto method of 3D representation in many vision tasks. This calls for the 3D understanding directly in this representation space. To facilitate the research in this direction, we first build a large-scale dataset of 3DGS using the commonly used ShapeNet and ModelNet datasets. Our dataset ShapeSplat consists of 65K objects from 87 unique categories, whose labels are in accordance with the respective datasets. The creation of this dataset utilized the compute equivalent of 2 GPU years on a TITAN XP GPU. We utilize our dataset for unsupervised pretraining and supervised finetuning for classification and segmentation tasks. To this end, we introduce \textbf{\textit{Gaussian-MAE}}, which highlights the unique benefits of representation learning from Gaussian parameters. Through exhaustive experiments, we provide several valuable insights. In particular, we show that (1) the distribution of the optimized GS centroids significantly differs from the uniformly sampled point cloud (used for initialization) counterpart; (2) this change in distribution results in degradation in classification but improvement in segmentation tasks when using only the centroids; (3) to leverage additional Gaussian parameters, we propose Gaussian feature grouping in a normalized feature space, along with splats pooling layer, offering a tailored solution to effectively group and embed similar Gaussians, which leads to notable improvement in finetuning tasks.
Authors: Zijian Zhao, Tingwei Chen, Zhijie Cai, Xiaoyang Li, Hang Li, Qimei Chen, Guangxu Zhu
Abstract: In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. To facilitate future research, we will release the code for our model upon publication.
Authors: Linlin Hu, Ao Sun, Shijie Hao, Richang Hong, Meng Wang
Abstract: Currently, most low-light image enhancement methods only consider information from a single view, neglecting the correlation between cross-view information. Therefore, the enhancement results produced by these methods are often unsatisfactory. In this context, there have been efforts to develop methods specifically for low-light stereo image enhancement. These methods take into account the cross-view disparities and enable interaction between the left and right views, leading to improved performance. However, these methods still do not fully exploit the interaction between left and right view information. To address this issue, we propose a model called Toward Sufficient Dual-View Interaction for Low-light Stereo Image Enhancement (SDI-Net). The backbone structure of SDI-Net is two encoder-decoder pairs, which are used to learn the mapping function from low-light images to normal-light images. Among the encoders and the decoders, we design a module named Cross-View Sufficient Interaction Module (CSIM), aiming to fully exploit the correlations between the binocular views via the attention mechanism. The quantitative and visual results on public datasets validate the superiority of our method over other related methods. Ablation studies also demonstrate the effectiveness of the key elements in our model.
Authors: Longfei Lu, Huachen Gao, Tao Dai, Yaohua Zha, Zhi Hou, Junta Wu, Shu-Tao Xia
Abstract: Recently, image-to-3D approaches have significantly advanced the generation quality and speed of 3D assets based on large reconstruction models, particularly 3D Gaussian reconstruction models. Existing large 3D Gaussian models directly map 2D image to 3D Gaussian parameters, while regressing 2D image to 3D Gaussian representations is challenging without 3D priors. In this paper, we propose a large Point-to-Gaussian model, that inputs the initial point cloud produced from large 3D diffusion model conditional on 2D image to generate the Gaussian parameters, for image-to-3D generation. The point cloud provides initial 3D geometry prior for Gaussian generation, thus significantly facilitating image-to-3D Generation. Moreover, we present the \textbf{A}ttention mechanism, \textbf{P}rojection mechanism, and \textbf{P}oint feature extractor, dubbed as \textbf{APP} block, for fusing the image features with point cloud features. The qualitative and quantitative experiments extensively demonstrate the effectiveness of the proposed approach on GSO and Objaverse datasets, and show the proposed method achieves state-of-the-art performance.
Authors: Vladislav Li, Georgios Tsoumplekas, Ilias Siniosoglou, Vasileios Argyriou, Anastasios Lytos, Eleftherios Fountoukidis, Panagiotis Sarigiannidis
Abstract: Current methods for low- and few-shot object detection have primarily focused on enhancing model performance for detecting objects. One common approach to achieve this is by combining model finetuning with data augmentation strategies. However, little attention has been given to the energy efficiency of these approaches in data-scarce regimes. This paper seeks to conduct a comprehensive empirical study that examines both model performance and energy efficiency of custom data augmentations and automated data augmentation selection strategies when combined with a lightweight object detector. The methods are evaluated in three different benchmark datasets in terms of their performance and energy consumption, and the Efficiency Factor is employed to gain insights into their effectiveness considering both performance and efficiency. Consequently, it is shown that in many cases, the performance gains of data augmentation strategies are overshadowed by their increased energy usage, necessitating the development of more energy efficient data augmentation strategies to address data scarcity.
Authors: Kazi Hasan Ibn Arif, JinYi Yoon, Dimitrios S. Nikolopoulos, Hans Vandierendonck, Deepu John, Bo Ji
Abstract: High-resolution Vision-Language Models (VLMs) have been widely used in multimodal tasks to enhance accuracy by preserving detailed image information. However, these models often generate excessive visual tokens due to encoding multiple partitions of the input image. Processing these excessive visual tokens is computationally challenging, especially in resource-constrained environments with commodity GPUs. To support high-resolution images while meeting resource constraints, we propose High-Resolution Early Dropping (HiRED), a token-dropping scheme that operates within a fixed token budget before the Large Language Model (LLM) stage. HiRED can be integrated with existing high-resolution VLMs in a plug-and-play manner, as it requires no additional training while still maintaining superior accuracy. We strategically use the vision encoder's attention in the initial layers to assess the visual content of each image partition and allocate the token budget accordingly. Then, using the attention in the final layer, we select the most important visual tokens from each partition within the allocated budget, dropping the rest. Empirically, when applied to LLaVA-Next-7B on NVIDIA TESLA P40 GPU, HiRED with a 20% token budget increases token generation throughput by 4.7, reduces first-token generation latency by 15 seconds, and saves 2.3 GB of GPU memory for a single inference.
Authors: Farhanul Haque, Md. Al-Hasan, Sumaiya Tabssum Mou, Abu Saleh Musa Miah, Jungpil Shin, Md Abdur Rahim
Abstract: The Bengali language is the 5th most spoken native and 7th most spoken language in the world, and Bengali handwritten character recognition has attracted researchers for decades. However, other languages such as English, Arabic, Turkey, and Chinese character recognition have contributed significantly to developing handwriting recognition systems. Still, little research has been done on Bengali character recognition because of the similarity of the character, curvature and other complexities. However, many researchers have used traditional machine learning and deep learning models to conduct Bengali hand-written recognition. The study employed a convolutional neural network (CNN) with ensemble transfer learning and a multichannel attention network. We generated the feature from the two branches of the CNN, including Inception Net and ResNet and then produced an ensemble feature fusion by concatenating them. After that, we applied the attention module to produce the contextual information from the ensemble features. Finally, we applied a classification module to refine the features and classification. We evaluated the proposed model using the CAMTERdb 3.1.2 data set and achieved 92\% accuracy for the raw dataset and 98.00\% for the preprocessed dataset. We believe that our contribution to the Bengali handwritten character recognition domain will be considered a great development.
Authors: Nitish Shukla, Arun Ross
Abstract: A face morph is created by combining the face images usually pertaining to two distinct identities. The goal is to generate an image that can be matched with two identities thereby undermining the security of a face recognition system. To deal with this problem, several morph attack detection techniques have been developed. But these methods do not extract any information about the underlying bonafides used to create them. Demorphing addresses this limitation. However, current demorphing techniques are mostly reference-based, i.e, they need an image of one of the identities to recover the other. In this work, we treat demorphing as an ill-posed decomposition problem. We propose a novel method that is reference-free and recovers the bonafides with high accuracy. Our method decomposes the morph into several identity-preserving feature components. A merger network then weighs and combines these components to recover the bonafides. Our method is observed to reconstruct high-quality bonafides in terms of definition and fidelity. Experiments on the CASIA-WebFace, SMDD and AMSL datasets demonstrate the effectiveness of our method.
Authors: Jonathan Prexl, Michael Schmitt
Abstract: This paper introduces SenPa-MAE, a transformer architecture that encodes the sensor parameters of an observed multispectral signal into the image embeddings. SenPa-MAE can be pre-trained on imagery of different satellites with non-matching spectral or geometrical sensor characteristics. To incorporate sensor parameters, we propose a versatile sensor parameter encoding module as well as a data augmentation strategy for the diversification of the pre-training dataset. This enables the model to effectively differentiate between various sensors and gain an understanding of sensor parameters and the correlation to the observed signal. Given the rising number of Earth observation satellite missions and the diversity in their sensor specifications, our approach paves the way towards a sensor-independent Earth observation foundation model. This opens up possibilities such as cross-sensor training and sensor-independent inference.
Authors: Haoning Wu, Shaocheng Shen, Qiang Hu, Xiaoyun Zhang, Ya Zhang, Yanfeng Wang
Abstract: Diffusion models have emerged as frontrunners in text-to-image generation for their impressive capabilities. Nonetheless, their fixed image resolution during training often leads to challenges in high-resolution image generation, such as semantic inaccuracies and object replication. This paper introduces MegaFusion, a novel approach that extends existing diffusion-based text-to-image generation models towards efficient higher-resolution generation without additional fine-tuning or extra adaptation. Specifically, we employ an innovative truncate and relay strategy to bridge the denoising processes across different resolutions, allowing for high-resolution image generation in a coarse-to-fine manner. Moreover, by integrating dilated convolutions and noise re-scheduling, we further adapt the model's priors for higher resolution. The versatility and efficacy of MegaFusion make it universally applicable to both latent-space and pixel-space diffusion models, along with other derivative models. Extensive experiments confirm that MegaFusion significantly boosts the capability of existing models to produce images of megapixels and various aspect ratios, while only requiring about 40% of the original computational cost.
Authors: Youjun Zhao, Jiaying Lin, Shuquan Ye, Qianshi Pang, Rynson W. H. Lau
Abstract: Open-vocabulary 3D scene understanding (OV-3D) aims to localize and classify novel objects beyond the closed object classes. However, existing approaches and benchmarks primarily focus on the open vocabulary problem within the context of object classes, which is insufficient to provide a holistic evaluation to what extent a model understands the 3D scene. In this paper, we introduce a more challenging task called Generalized Open-Vocabulary 3D Scene Understanding (GOV-3D) to explore the open vocabulary problem beyond object classes. It encompasses an open and diverse set of generalized knowledge, expressed as linguistic queries of fine-grained and object-specific attributes. To this end, we contribute a new benchmark named OpenScan, which consists of 3D object attributes across eight representative linguistic aspects, including affordance, property, material, and more. We further evaluate state-of-the-art OV-3D methods on our OpenScan benchmark, and discover that these methods struggle to comprehend the abstract vocabularies of the GOV-3D task, a challenge that cannot be addressed by simply scaling up object classes during training. We highlight the limitations of existing methodologies and explore a promising direction to overcome the identified shortcomings. Data and code are available at https://github.com/YoujunZhao/OpenScan
Authors: Yunzhe Xu, Yiyuan Pan, Zhe Liu, Hesheng Wang
Abstract: Large Language Models (LLMs) have demonstrated potential in Vision-and-Language Navigation (VLN) tasks, yet current applications face challenges. While LLMs excel in general conversation scenarios, they struggle with specialized navigation tasks, yielding suboptimal performance compared to specialized VLN models. We introduce FLAME (FLAMingo-Architected Embodied Agent), a novel Multimodal LLM-based agent and architecture designed for urban VLN tasks that efficiently handles multiple observations. Our approach implements a three-phase tuning technique for effective adaptation to navigation tasks, including single perception tuning for street view description, multiple perception tuning for trajectory summarization, and end-to-end training on VLN datasets. The augmented datasets are synthesized automatically. Experimental results demonstrate FLAME's superiority over existing methods, surpassing state-of-the-art methods by a 7.3% increase in task completion rate on Touchdown dataset. This work showcases the potential of Multimodal LLMs (MLLMs) in complex navigation tasks, representing an advancement towards practical applications of MLLMs in embodied AI. Project page: https://flame-sjtu.github.io
Authors: Valentinos Pariza, Mohammadreza Salehi, Gertjan Burghouts, Francesco Locatello, Yuki M. Asano
Abstract: We propose sorting patch representations across views as a novel self-supervised learning signal to improve pretrained representations. To this end, we introduce NeCo: Patch Neighbor Consistency, a novel training loss that enforces patch-level nearest neighbor consistency across a student and teacher model, relative to reference batches. Our method leverages a differentiable sorting method applied on top of pretrained representations, such as DINOv2-registers to bootstrap the learning signal and further improve upon them. This dense post-pretraining leads to superior performance across various models and datasets, despite requiring only 19 hours on a single GPU. We demonstrate that this method generates high-quality dense feature encoders and establish several new state-of-the-art results: +5.5% and + 6% for non-parametric in-context semantic segmentation on ADE20k and Pascal VOC, and +7.2% and +5.7% for linear segmentation evaluations on COCO-Things and -Stuff.
Authors: Satoshi Kosugi
Abstract: In this paper, we delve into the concept of interpretable image enhancement, a technique that enhances image quality by adjusting filter parameters with easily understandable names such as "Exposure" and "Contrast". Unlike using predefined image editing filters, our framework utilizes learnable filters that acquire interpretable names through training. Our contribution is two-fold. Firstly, we introduce a novel filter architecture called an image-adaptive neural implicit lookup table, which uses a multilayer perceptron to implicitly define the transformation from input feature space to output color space. By incorporating image-adaptive parameters directly into the input features, we achieve highly expressive filters. Secondly, we introduce a prompt guidance loss to assign interpretable names to each filter. We evaluate visual impressions of enhancement results, such as exposure and contrast, using a vision and language model along with guiding prompts. We define a constraint to ensure that each filter affects only the targeted visual impression without influencing other attributes, which allows us to obtain the desired filter effects. Experimental results show that our method outperforms existing predefined filter-based methods, thanks to the filters optimized to predict target results. Our source code is available at https://github.com/satoshi-kosugi/PG-IA-NILUT.
Authors: Ruoxi Chen, Haibo Jin, Yixin Liu, Jinyin Chen, Haohan Wang, Lichao Sun
Abstract: Text-to-image diffusion models have emerged as an evolutionary for producing creative content in image synthesis. Based on the impressive generation abilities of these models, instruction-guided diffusion models can edit images with simple instructions and input images. While they empower users to obtain their desired edited images with ease, they have raised concerns about unauthorized image manipulation. Prior research has delved into the unauthorized use of personalized diffusion models; however, this problem of instruction-guided diffusion models remains largely unexplored. In this paper, we first propose a protection method EditShield against unauthorized modifications from such models. Specifically, EditShield works by adding imperceptible perturbations that can shift the latent representation used in the diffusion process, tricking models into generating unrealistic images with mismatched subjects. Our extensive experiments demonstrate EditShield's effectiveness among synthetic and real-world datasets. Besides, we found that EditShield performs robustly against various manipulation settings across editing types and synonymous instruction phrases.
Authors: Wenxin Fan, Jian Cheng, Cheng Li, Jing Yang, Ruoyou Wu, Juan Zou, Shanshan Wang
Abstract: Deep learning has shown great potential in accelerating diffusion tensor imaging (DTI). Nevertheless, existing methods tend to suffer from Rician noise and eddy current, leading to detail loss in reconstructing the DTI-derived parametric maps especially when sparsely sampled q-space data are used. To address this, this paper proposes a novel method, AID-DTI (\textbf{A}ccelerating h\textbf{I}gh fi\textbf{D}elity \textbf{D}iffusion \textbf{T}ensor \textbf{I}maging), to facilitate fast and accurate DTI with only six measurements. AID-DTI is equipped with a newly designed Singular Value Decomposition-based regularizer, which can effectively capture fine details while suppressing noise during network training by exploiting the correlation across DTI-derived parameters. Additionally, we introduce a Nesterov-based adaptive learning algorithm that optimizes the regularization parameter dynamically to enhance the performance. AID-DTI is an extendable framework capable of incorporating flexible network architecture. Experimental results on Human Connectome Project (HCP) data consistently demonstrate that the proposed method estimates DTI parameter maps with fine-grained details and outperforms other state-of-the-art methods both quantitatively and qualitatively.
Authors: Seonghee Lee, Maho Kohga, Steve Landau, Sile O'Modhrain, Hari Subramonyam
Abstract: People with visual impairments often struggle to create content that relies heavily on visual elements, particularly when conveying spatial and structural information. Existing accessible drawing tools, which construct images line by line, are suitable for simple tasks like math but not for more expressive artwork. On the other hand, emerging generative AI-based text-to-image tools can produce expressive illustrations from descriptions in natural language, but they lack precise control over image composition and properties. To address this gap, our work integrates generative AI with a constructive approach that provides users with enhanced control and editing capabilities. Our system, AltCanvas, features a tile-based interface enabling users to construct visual scenes incrementally, with each tile representing an object within the scene. Users can add, edit, move, and arrange objects while receiving speech and audio feedback. Once completed, the scene can be rendered as a color illustration or as a vector for tactile graphic generation. Involving 14 blind or low-vision users in design and evaluation, we found that participants effectively used the AltCanvas workflow to create illustrations.
Authors: An Vuong, Thinh Nguyen
Abstract: Multiplicative noise, also known as speckle or pepper noise, commonly affects images produced by synthetic aperture radar (SAR), lasers, or optical lenses. Unlike additive noise, which typically arises from thermal processes or external factors, multiplicative noise is inherent to the system, originating from the fluctuation in diffuse reflections. These fluctuations result in multiple copies of the same signal with varying magnitudes being combined. Consequently, despeckling, or removing multiplicative noise, necessitates different techniques compared to those used for additive noise removal. In this paper, we propose a novel approach using Stochastic Differential Equations based diffusion models to address multiplicative noise. We demonstrate that multiplicative noise can be effectively modeled as a Geometric Brownian Motion process in the logarithmic domain. Utilizing the Fokker-Planck equation, we derive the corresponding reverse process for image denoising. To validate our method, we conduct extensive experiments on two different datasets, comparing our approach to both classical signal processing techniques and contemporary CNN-based noise removal models. Our results indicate that the proposed method significantly outperforms existing methods on perception-based metrics such as FID and LPIPS, while maintaining competitive performance on traditional metrics like PSNR and SSIM.
Authors: Di Fang, Yinan Zhu, Runze Fang, Cen Chen, Ziqian Zeng, Huiping Zhuang
Abstract: Continual learning enables AI models to learn new data sequentially without retraining in real-world scenarios. Most existing methods assume the training data are balanced, aiming to reduce the catastrophic forgetting problem that models tend to forget previously generated data. However, data imbalance and the mixture of new and old data in real-world scenarios lead the model to ignore categories with fewer training samples. To solve this problem, we propose an analytic imbalance rectifier algorithm (AIR), a novel online exemplar-free continual learning method with an analytic (i.e., closed-form) solution for data-imbalanced class-incremental learning (CIL) and generalized CIL scenarios in real-world continual learning. AIR introduces an analytic re-weighting module (ARM) that calculates a re-weighting factor for each class for the loss function to balance the contribution of each category to the overall loss and solve the problem of imbalanced training data. AIR uses the least squares technique to give a non-discriminatory optimal classifier and its iterative update method in continual learning. Experimental results on multiple datasets show that AIR significantly outperforms existing methods in long-tailed and generalized CIL scenarios. The source code is available at https://github.com/fang-d/AIR.
Authors: Chaitra Hegde, Yashar Kiarashi, Allan I Levey, Amy D Rodriguez, Hyeokhyen Kwon, Gari D Clifford
Abstract: INTRODUCTION: Mild cognitive impairment (MCI) is characterized by a decline in cognitive functions beyond typical age and education-related expectations. Since, MCI has been linked to reduced social interactions and increased aimless movements, we aimed to automate the capture of these behaviors to enhance longitudinal monitoring. METHODS: Using a privacy-preserving distributed camera network, we collected movement and social interaction data from groups of individuals with MCI undergoing therapy within a 1700$m^2$ space. We developed movement and social interaction features, which were then used to train a series of machine learning algorithms to distinguish between higher and lower cognitive functioning MCI groups. RESULTS: A Wilcoxon rank-sum test revealed statistically significant differences between high and low-functioning cohorts in features such as linear path length, walking speed, change in direction while walking, entropy of velocity and direction change, and number of group formations in the indoor space. Despite lacking individual identifiers to associate with specific levels of MCI, a machine learning approach using the most significant features provided a 71% accuracy. DISCUSSION: We provide evidence to show that a privacy-preserving low-cost camera network using edge computing framework has the potential to distinguish between different levels of cognitive impairment from the movements and social interactions captured during group activities.
Authors: Shiyu Yuan, Carlo Lipizzi, Tian Han
Abstract: Multimodal generative models are crucial for various applications. We propose an approach that combines an expressive energy-based model (EBM) prior with Markov Chain Monte Carlo (MCMC) inference in the latent space for multimodal generation. The EBM prior acts as an informative guide, while MCMC inference, specifically through short-run Langevin dynamics, brings the posterior distribution closer to its true form. This method not only provides an expressive prior to better capture the complexity of multimodality but also improves the learning of shared latent variables for more coherent generation across modalities. Our proposed method is supported by empirical experiments, underscoring the effectiveness of our EBM prior with MCMC inference in enhancing cross-modal and joint generative tasks in multimodal contexts.
Authors: Tatsuhiro Baba, Abu Saleh Musa Miah, Jungpil Shin, Md. Al Mehedi Hasan
Abstract: Cervical cancer is a crucial global health concern for women, and the persistent infection of High-risk HPV mainly triggers this remains a global health challenge, with young women diagnosis rates soaring from 10\% to 40\% over three decades. While Pap smear screening is a prevalent diagnostic method, visual image analysis can be lengthy and often leads to mistakes. Early detection of the disease can contribute significantly to improving patient outcomes. In recent decades, many researchers have employed machine learning techniques that achieved promise in cervical cancer detection processes based on medical images. In recent years, many researchers have employed various deep-learning techniques to achieve high-performance accuracy in detecting cervical cancer but are still facing various challenges. This research proposes an innovative and novel approach to automate cervical cancer image classification using Multi-Head Self-Attention (MHSA) and convolutional neural networks (CNNs). The proposed method leverages the strengths of both MHSA mechanisms and CNN to effectively capture both local and global features within cervical images in two streams. MHSA facilitates the model's ability to focus on relevant regions of interest, while CNN extracts hierarchical features that contribute to accurate classification. Finally, we combined the two stream features and fed them into the classification module to refine the feature and the classification. To evaluate the performance of the proposed approach, we used the SIPaKMeD dataset, which classifies cervical cells into five categories. Our model achieved a remarkable accuracy of 98.522\%. This performance has high recognition accuracy of medical image classification and holds promise for its applicability in other medical image recognition tasks.
Authors: Zebang Cheng, Shuyuan Tu, Dawei Huang, Minghan Li, Xiaojiang Peng, Zhi-Qi Cheng, Alexander G. Hauptmann
Abstract: This paper presents our winning approach for the MER-NOISE and MER-OV tracks of the MER2024 Challenge on multimodal emotion recognition. Our system leverages the advanced emotional understanding capabilities of Emotion-LLaMA to generate high-quality annotations for unlabeled samples, addressing the challenge of limited labeled data. To enhance multimodal fusion while mitigating modality-specific noise, we introduce Conv-Attention, a lightweight and efficient hybrid framework. Extensive experimentation vali-dates the effectiveness of our approach. In the MER-NOISE track, our system achieves a state-of-the-art weighted average F-score of 85.30%, surpassing the second and third-place teams by 1.47% and 1.65%, respectively. For the MER-OV track, our utilization of Emotion-LLaMA for open-vocabulary annotation yields an 8.52% improvement in average accuracy and recall compared to GPT-4V, securing the highest score among all participating large multimodal models. The code and model for Emotion-LLaMA are available at https://github.com/ZebangCheng/Emotion-LLaMA.
Authors: Tutian Tang, Minghao Liu, Wenqiang Xu, Cewu Lu
Abstract: Hand-eye calibration involves estimating the transformation between the camera and the robot. Traditional methods rely on fiducial markers, involving much manual labor and careful setup. Recent advancements in deep learning offer markerless techniques, but they present challenges, including the need for retraining networks for each robot, the requirement of accurate mesh models for data generation, and the need to address the sim-to-real gap. In this letter, we propose Kalib, an automatic and universal markerless hand-eye calibration pipeline that leverages the generalizability of visual foundation models to eliminate these barriers. In each calibration process, Kalib uses keypoint tracking and proprioceptive sensors to estimate the transformation between a robot's coordinate space and its corresponding points in camera space. Our method does not require training new networks or access to mesh models. Through evaluations in simulation environments and the real-world dataset DROID, Kalib demonstrates superior accuracy compared to recent baseline methods. This approach provides an effective and flexible calibration process for various robot systems by simplifying setup and removing dependency on precise physical markers.
Authors: Zijian Dong, Yilei Wu, Zijiao Chen, Yichi Zhang, Yueming Jin, Juan Helen Zhou
Abstract: We introduce Scaffold Prompt Tuning (ScaPT), a novel prompt-based framework for adapting large-scale functional magnetic resonance imaging (fMRI) pre-trained models to downstream tasks, with high parameter efficiency and improved performance compared to fine-tuning and baselines for prompt tuning. The full fine-tuning updates all pre-trained parameters, which may distort the learned feature space and lead to overfitting with limited training data which is common in fMRI fields. In contrast, we design a hierarchical prompt structure that transfers the knowledge learned from high-resource tasks to low-resource ones. This structure, equipped with a Deeply-conditioned Input-Prompt (DIP) mapping module, allows for efficient adaptation by updating only 2% of the trainable parameters. The framework enhances semantic interpretability through attention mechanisms between inputs and prompts, and it clusters prompts in the latent space in alignment with prior knowledge. Experiments on public resting state fMRI datasets reveal ScaPT outperforms fine-tuning and multitask-based prompt tuning in neurodegenerative diseases diagnosis/prognosis and personality trait prediction, even with fewer than 20 participants. It highlights ScaPT's efficiency in adapting pre-trained fMRI models to low-resource tasks.
Authors: Kevin Kam Fung Yuen
Abstract: This paper presents a tutorial of an explainable approach using Convolutional Neural Network (CNN) and Gradient-weighted Class Activation Mapping (Grad-CAM) to classify four progressive dementia stages based on open MRI brain images. The detailed implementation steps are demonstrated with an explanation. Whilst the proposed CNN architecture is demonstrated to achieve more than 99% accuracy for the test dataset, the computational procedure of CNN remains a black box. The visualisation based on Grad-CAM is attempted to explain such very high accuracy and may provide useful information for physicians. Future motivation based on this work is discussed.
Authors: Eui Jun Hwang, Sukmin Cho, Junmyeong Lee, Jong C. Park
Abstract: Gloss-free Sign Language Translation (SLT) converts sign videos directly into spoken language sentences without relying on glosses. Recently, Large Language Models (LLMs) have shown remarkable translation performance in gloss-free methods by harnessing their powerful natural language generation capabilities. However, these methods often rely on domain-specific fine-tuning of visual encoders to achieve optimal results. By contrast, this paper emphasizes the importance of capturing the spatial configurations and motion dynamics inherent in sign language. With this in mind, we introduce Spatial and Motion-based Sign Language Translation (SpaMo), a novel LLM-based SLT framework. The core idea of SpaMo is simple yet effective. We first extract spatial and motion features using off-the-shelf visual encoders and then input these features into an LLM with a language prompt. Additionally, we employ a visual-text alignment process as a warm-up before the SLT supervision. Our experiments demonstrate that SpaMo achieves state-of-the-art performance on two popular datasets, PHOENIX14T and How2Sign.
Authors: Hongtian Yu, Yangu Li, Mingrui Wu, Letian Shen, Yue Liu, Yunxuan Song, Qixiang Ye, Xiaorui Lyu, Yajun Mao, Yangheng Zheng, Yunfan Liu
Abstract: In high-energy physics, anti-neutrons ($\bar{n}$) are fundamental particles that frequently appear as final-state particles, and the reconstruction of their kinematic properties provides an important probe for understanding the governing principles. However, this confronts significant challenges instrumentally with the electromagnetic calorimeter (EMC), a typical experimental sensor but recovering the information of incident $\bar{n}$ insufficiently. In this study, we introduce Vision Calorimeter (ViC), a baseline method for anti-neutron reconstruction that leverages deep learning detectors to analyze the implicit relationships between EMC responses and incident $\bar{n}$ characteristics. Our motivation lies in that energy distributions of $\bar{n}$ samples deposited in the EMC cell arrays embody rich contextual information. Converted to 2-D images, such contextual energy distributions can be used to predict the status of $\bar{n}$ ($i.e.$, incident position and momentum) through a deep learning detector along with pseudo bounding boxes and a specified training objective. Experimental results demonstrate that ViC substantially outperforms the conventional reconstruction approach, reducing the prediction error of incident position by 42.81% (from 17.31$^{\circ}$ to 9.90$^{\circ}$). More importantly, this study for the first time realizes the measurement of incident $\bar{n}$ momentum, underscoring the potential of deep learning detectors for particle reconstruction. Code is available at https://github.com/yuhongtian17/ViC.
Authors: Junming Wang, Dong Huang, Xiuxian Guan, Zekai Sun, Tianxiang Shen, Fangming Liu, Heming Cui
Abstract: Air-ground robots (AGRs) are widely used in surveillance and disaster response due to their exceptional mobility and versatility (i.e., flying and driving). Current AGR navigation systems perform well in static occlusion-prone environments (e.g., indoors) by using 3D semantic occupancy networks to predict occlusions for complete local mapping and then computing Euclidean Signed Distance Field (ESDF) for path planning. However, these systems face challenges in dynamic, severe occlusion scenes (e.g., crowds) due to limitations in perception networks' low prediction accuracy and path planners' high computation overhead. In this paper, we propose OMEGA, which contains OccMamba with an Efficient AGR-Planner to address the above-mentioned problems. OccMamba adopts a novel architecture that separates semantic and occupancy prediction into independent branches, incorporating two mamba blocks within these branches. These blocks efficiently extract semantic and geometric features in 3D environments with linear complexity, ensuring that the network can learn long-distance dependencies to improve prediction accuracy. Semantic and geometric features are combined within the Bird's Eye View (BEV) space to minimise computational overhead during feature fusion. The resulting semantic occupancy map is then seamlessly integrated into the local map, providing occlusion awareness of the dynamic environment. Our AGR-Planner utilizes this local map and employs kinodynamic A* search and gradient-based trajectory optimization to guarantee planning is ESDF-free and energy-efficient. Extensive experiments demonstrate that OccMamba outperforms the state-of-the-art 3D semantic occupancy network with 25.0% mIoU. End-to-end navigation experiments in dynamic scenes verify OMEGA's efficiency, achieving a 96% average planning success rate. Code and video are available at https://jmwang0117.github.io/OMEGA/.
Authors: Ruoyu Chen, Kezheng Xu, Kangyan Zheng, Weiyi Zhang, Yan Lu, Danli Shi, Mingguang He
Abstract: Ultrawide-field fluorescein angiography (UWF-FA) facilitates diabetic retinopathy (DR) detection by providing a clear visualization of peripheral retinal lesions. However, the intravenous dye injection with potential risks hamper its application. We aim to acquire dye-free UWF-FA images from noninvasive UWF color fundus (UWF-CF) images using generative artificial intelligence (GenAI) and evaluate its effectiveness in DR screening. A total of 18,321 UWF-FA images of different phases were registered with corresponding UWF-CF images and fed into a generative adversarial networks (GAN)-based model for training. The quality of generated UWF-FA images was evaluated through quantitative metrics and human evaluation. The DeepDRiD dataset was used to externally assess the contribution of generated UWF-FA images to DR classification, using area under the receiver operating characteristic curve (AUROC) as outcome metrics. The generated early, mid, and late phase UWF-FA images achieved high authenticity, with multi-scale similarity scores ranging from 0.70 to 0.91 and qualitative visual scores ranging from 1.64 to 1.98 (1=real UWF-FA quality). In fifty randomly selected images, 56% to 76% of the generated images were difficult to distinguish from real images in the Turing test. Moreover, adding these generated UWF-FA images for DR classification significantly increased the AUROC from 0.869 to 0.904 compared to the baseline model using UWF-CF images (P < .001). The model successfully generates realistic multi-frame UWF-FA images without intravenous dye injection. The generated UWF-FA enhanced DR stratification.
Authors: Lukas Fisch, Nils R. Winter, Janik Goltermann, Carlotta Barkhau, Daniel Emden, Jan Ernsting, Maximilian Konowski, Ramona Leenings, Tiana Borgers, Kira Flinkenfl\"ugel, Dominik Grotegerd, Anna Kraus, Elisabeth J. Leehr, Susanne Meinert, Frederike Stein, Lea Teutenberg, Florian Thomas-Odenthal, Paula Usemann, Marco Hermesdorf, Hamidreza Jamalabadi, Andreas Jansen, Igor Nenadic, Benjamin Straube, Tilo Kircher, Klaus Berger, Benjamin Risse, Udo Dannlowski, Tim Hahn
Abstract: Voxel-based Morphometry (VBM) has emerged as a powerful approach in neuroimaging research, utilized in over 7,000 studies since the year 2000. Using Magnetic Resonance Imaging (MRI) data, VBM assesses variations in the local density of brain tissue and examines its associations with biological and psychometric variables. Here, we present deepmriprep, a neural network-based pipeline that performs all necessary preprocessing steps for VBM analysis of T1-weighted MR images using deep neural networks. Utilizing the Graphics Processing Unit (GPU), deepmriprep is 37 times faster than CAT12, the leading VBM preprocessing toolbox. The proposed method matches CAT12 in accuracy for tissue segmentation and image registration across more than 100 datasets and shows strong correlations in VBM results. Tissue segmentation maps from deepmriprep have over 95% agreement with ground truth maps, and its non-linear registration, using supervised SYMNet, predicts smooth deformation fields comparable to CAT12. The high processing speed of deepmriprep enables rapid preprocessing of extensive datasets and thereby fosters the application of VBM analysis to large-scale neuroimaging studies and opens the door to real-time applications. Finally, deepmripreps straightforward, modular design enables researchers to easily understand, reuse, and advance the underlying methods, fostering further advancements in neuroimaging research. deepmriprep can be conveniently installed as a Python package and is publicly accessible at https://github.com/wwu-mmll/deepmriprep.
Authors: Aliza Subedi, Smriti Regmi, Nisha Regmi, Bhumi Bhusal, Ulas Bagci, Debesh Jha
Abstract: Gastrointestinal cancer is a leading cause of cancer-related incidence and death, making it crucial to develop novel computer-aided diagnosis systems for early detection and enhanced treatment. Traditional approaches rely on the expertise of gastroenterologists to identify diseases; however, this process is subjective, and interpretation can vary even among expert clinicians. Considering recent advancements in classifying gastrointestinal anomalies and landmarks in endoscopic and video capsule endoscopy images, this study proposes a hybrid model that combines the advantages of Transformers and Convolutional Neural Networks (CNNs) to enhance classification performance. Our model utilizes DenseNet201 as a CNN branch to extract local features and integrates a Swin Transformer branch for global feature understanding, combining both to perform the classification task. For the GastroVision dataset, our proposed model demonstrates excellent performance with Precision, Recall, F1 score, Accuracy, and Matthews Correlation Coefficient (MCC) of 0.8320, 0.8386, 0.8324, 0.8386, and 0.8191, respectively, showcasing its robustness against class imbalance and surpassing other CNNs as well as the Swin Transformer model. Similarly, for the Kvasir-Capsule, a large video capsule endoscopy dataset, our model outperforms all others, achieving overall Precision, Recall, F1 score, Accuracy, and MCC of 0.7007, 0.7239, 0.6900, 0.7239, and 0.3871. Moreover, we generated saliency maps to explain our model's focus areas, demonstrating its reliable decision-making process. The results underscore the potential of our hybrid CNN-Transformer model in aiding the early and accurate detection of gastrointestinal (GI) anomalies.
Authors: Karen Sanchez, Carlos Hinojosa, Olinto Mieles, Chen Zhao, Bernard Ghanem, Henry Arguello
Abstract: Chronic wounds pose an ongoing health concern globally, largely due to the prevalence of conditions such as diabetes and leprosy's disease. The standard method of monitoring these wounds involves visual inspection by healthcare professionals, a practice that could present challenges for patients in remote areas with inadequate transportation and healthcare infrastructure. This has led to the development of algorithms designed for the analysis and follow-up of wound images, which perform image-processing tasks such as classification, detection, and segmentation. However, the effectiveness of these algorithms heavily depends on the availability of comprehensive and varied wound image data, which is usually scarce. This paper introduces the CO2Wounds-V2 dataset, an extended collection of RGB wound images from leprosy patients with their corresponding semantic segmentation annotations, aiming to enhance the development and testing of image-processing algorithms in the medical field.
Authors: Zili Liu, Hao Chen, Lei Bai, Wenyuan Li, Wanli Ouyang, Zhengxia Zou, Zhenwei Shi
Abstract: In an era of frequent extreme weather and global warming, obtaining precise, fine-grained near-surface weather forecasts is increasingly essential for human activities. Downscaling (DS), a crucial task in meteorological forecasting, enables the reconstruction of high-resolution meteorological states for target regions from global-scale forecast results. Previous downscaling methods, inspired by CNN and Transformer-based super-resolution models, lacked tailored designs for meteorology and encountered structural limitations. Notably, they failed to efficiently integrate topography, a crucial prior in the downscaling process. In this paper, we address these limitations by pioneering the selective state space model into the meteorological field downscaling and propose a novel model called MambaDS. This model enhances the utilization of multivariable correlations and topography information, unique challenges in the downscaling process while retaining the advantages of Mamba in long-range dependency modeling and linear computational complexity. Through extensive experiments in both China mainland and the continental United States (CONUS), we validated that our proposed MambaDS achieves state-of-the-art results in three different types of meteorological field downscaling settings. We will release the code subsequently.
Authors: Chiara Stuardi, Claudio Gheller, Franco Vazza, Andrea Botteon
Abstract: The forthcoming generation of radio telescope arrays promises significant advancements in sensitivity and resolution, enabling the identification and characterization of many new faint and diffuse radio sources. Conventional manual cataloging methodologies are anticipated to be insufficient to exploit the capabilities of new radio surveys. Radio interferometric images of diffuse sources present a challenge for image segmentation tasks due to noise, artifacts, and embedded radio sources. In response to these challenges, we introduce Radio U-Net, a fully convolutional neural network based on the U-Net architecture. Radio U-Net is designed to detect faint and extended sources in radio surveys, such as radio halos, relics, and cosmic web filaments. Radio U-Net was trained on synthetic radio observations built upon cosmological simulations and then tested on a sample of galaxy clusters, where the detection of cluster diffuse radio sources relied on customized data reduction and visual inspection of LOFAR Two Metre Sky Survey (LoTSS) data. The 83% of clusters exhibiting diffuse radio emission were accurately identified, and the segmentation successfully recovered the morphology of the sources even in low-quality images. In a test sample comprising 246 galaxy clusters, we achieved a 73% accuracy rate in distinguishing between clusters with and without diffuse radio emission. Our results establish the applicability of Radio U-Net to extensive radio survey datasets, probing its efficiency on cutting-edge high-performance computing systems. This approach represents an advancement in optimizing the exploitation of forthcoming large radio surveys for scientific exploration.
Authors: Xinqi Su, Yawen Cui, Ajian Liu, Xun Lin, Yuhao Wang, Haochen Liang, Wenhui Li, Zitong Yu
Abstract: In current web environment, fake news spreads rapidly across online social networks, posing serious threats to society. Existing multimodal fake news detection (MFND) methods can be classified into knowledge-based and semantic-based approaches. However, these methods are overly dependent on human expertise and feedback, lacking flexibility. To address this challenge, we propose a Dynamic Analysis and Adaptive Discriminator (DAAD) approach for fake news detection. For knowledge-based methods, we introduce the Monte Carlo Tree Search (MCTS) algorithm to leverage the self-reflective capabilities of large language models (LLMs) for prompt optimization, providing richer, domain-specific details and guidance to the LLMs, while enabling more flexible integration of LLM comment on news content. For semantic-based methods, we define four typical deceit patterns: emotional exaggeration, logical inconsistency, image manipulation, and semantic inconsistency, to reveal the mechanisms behind fake news creation. To detect these patterns, we carefully design four discriminators and expand them in depth and breadth, using the soft-routing mechanism to explore optimal detection models. Experimental results on three real-world datasets demonstrate the superiority of our approach. The code will be available at: https://github.com/SuXinqi/DAAD.
Authors: Ezequiel de la Rosa, Ruisheng Su, Mauricio Reyes, Roland Wiest, Evamaria O. Riedel, Florian Kofler, Kaiyuan Yang, Hakim Baazaoui, David Robben, Susanne Wegener, Jan S. Kirschke, Benedikt Wiestler, Bjoern Menze
Abstract: Accurate estimation of core (irreversibly damaged tissue) and penumbra (salvageable tissue) volumes is essential for ischemic stroke treatment decisions. Perfusion CT, the clinical standard, estimates these volumes but is affected by variations in deconvolution algorithms, implementations, and thresholds. Core tissue expands over time, with growth rates influenced by thrombus location, collateral circulation, and inherent patient-specific factors. Understanding this tissue growth is crucial for determining the need to transfer patients to comprehensive stroke centers, predicting the benefits of additional reperfusion attempts during mechanical thrombectomy, and forecasting final clinical outcomes. This work presents the ISLES'24 challenge, which addresses final post-treatment stroke infarct prediction from pre-interventional acute stroke imaging and clinical data. ISLES'24 establishes a unique 360-degree setting where all feasibly accessible clinical data are available for participants, including full CT acute stroke imaging, sub-acute follow-up MRI, and clinical tabular data. The contributions of this work are two-fold: first, we introduce a standardized benchmarking of final stroke infarct segmentation algorithms through the ISLES'24 challenge; second, we provide insights into infarct segmentation using multimodal imaging and clinical data strategies by identifying outperforming methods on a finely curated dataset. The outputs of this challenge are anticipated to enhance clinical decision-making and improve patient outcome predictions. All ISLES'24 materials, including data, performance evaluation scripts, and leading algorithmic strategies, are available to the research community following \url{https://isles-24.grand-challenge.org/}.
Authors: Hojat Asgariandehkordi, Sobhan Goudarzi, Mostafa Sharifzadeh, Adrian Basarab, Hassan Rivaz
Abstract: Ultrasound plane wave imaging is a cutting-edge technique that enables high frame-rate imaging. However, one challenge associated with high frame-rate ultrasound imaging is the high noise associated with them, hindering their wider adoption. Therefore, the development of a denoising method becomes imperative to augment the quality of plane wave images. Drawing inspiration from Denoising Diffusion Probabilistic Models (DDPMs), our proposed solution aims to enhance plane wave image quality. Specifically, the method considers the distinction between low-angle and high-angle compounding plane waves as noise and effectively eliminates it by adapting a DDPM to beamformed radiofrequency (RF) data. The method underwent training using only 400 simulated images. In addition, our approach employs natural image segmentation masks as intensity maps for the generated images, resulting in accurate denoising for various anatomy shapes. The proposed method was assessed across simulation, phantom, and in vivo images. The results of the evaluations indicate that our approach not only enhances image quality on simulated data but also demonstrates effectiveness on phantom and in vivo data in terms of image quality. Comparative analysis with other methods underscores the superiority of our proposed method across various evaluation metrics. The source code and trained model will be released along with the dataset at: http://code.sonography.ai
Authors: Vitus Benson, Ana Bastos, Christian Reimers, Alexander J. Winkler, Fanny Yang, Markus Reichstein
Abstract: Accurately describing the distribution of CO$_2$ in the atmosphere with atmospheric tracer transport models is essential for greenhouse gas monitoring and verification support systems to aid implementation of international climate agreements. Large deep neural networks are poised to revolutionize weather prediction, which requires 3D modeling of the atmosphere. While similar in this regard, atmospheric transport modeling is subject to new challenges. Both, stable predictions for longer time horizons and mass conservation throughout need to be achieved, while IO plays a larger role compared to computational costs. In this study we explore four different deep neural networks (UNet, GraphCast, Spherical Fourier Neural Operator and SwinTransformer) which have proven as state-of-the-art in weather prediction to assess their usefulness for atmospheric tracer transport modeling. For this, we assemble the CarbonBench dataset, a systematic benchmark tailored for machine learning emulators of Eulerian atmospheric transport. Through architectural adjustments, we decouple the performance of our emulators from the distribution shift caused by a steady rise in atmospheric CO$_2$. More specifically, we center CO$_2$ input fields to zero mean and then use an explicit flux scheme and a mass fixer to assure mass balance. This design enables stable and mass conserving transport for over 6 months with all four neural network architectures. In our study, the SwinTransformer displays particularly strong emulation skill (90-day $R^2 > 0.99$), with physically plausible emulation even for forward runs of multiple years. This work paves the way forward towards high resolution forward and inverse modeling of inert trace gases with neural networks.
Authors: Chunting Zhou, Lili Yu, Arun Babu, Kushal Tirumala, Michihiro Yasunaga, Leonid Shamis, Jacob Kahn, Xuezhe Ma, Luke Zettlemoyer, Omer Levy
Abstract: We introduce Transfusion, a recipe for training a multi-modal model over discrete and continuous data. Transfusion combines the language modeling loss function (next token prediction) with diffusion to train a single transformer over mixed-modality sequences. We pretrain multiple Transfusion models up to 7B parameters from scratch on a mixture of text and image data, establishing scaling laws with respect to a variety of uni- and cross-modal benchmarks. Our experiments show that Transfusion scales significantly better than quantizing images and training a language model over discrete image tokens. By introducing modality-specific encoding and decoding layers, we can further improve the performance of Transfusion models, and even compress each image to just 16 patches. We further demonstrate that scaling our Transfusion recipe to 7B parameters and 2T multi-modal tokens produces a model that can generate images and text on a par with similar scale diffusion models and language models, reaping the benefits of both worlds.
Authors: Xingyuan Yu, Weicai Ye, Xiyue Guo, Yuhang Ming, Jinyu Li, Hujun Bao, Zhaopeng Cui, Guofeng Zhang
Abstract: In this paper, we introduce a self-supervised deep SLAM method that robustly operates in dynamic scenes while accurately identifying dynamic components. Our method leverages a dual-flow representation for static flow and dynamic flow, facilitating effective scene decomposition in dynamic environments. We propose a dynamic update module based on this representation and develop a dense SLAM system that excels in dynamic scenarios. In addition, we design a self-supervised training scheme using DINO as a prior, enabling label-free training. Our method achieves superior accuracy compared to other self-supervised methods. It also matches or even surpasses the performance of existing supervised methods in some cases. All code and data will be made publicly available upon acceptance.
Authors: Haochen Wang, Yuchao Wang, Yujun Shen, Junsong Fan, Yuxi Wang, Zhaoxiang Zhang
Abstract: The crux of label-efficient semantic segmentation is to produce high-quality pseudo-labels to leverage a large amount of unlabeled or weakly labeled data. A common practice is to select the highly confident predictions as the pseudo-ground-truths for each pixel, but it leads to a problem that most pixels may be left unused due to their unreliability. However, we argue that every pixel matters to the model training, even those unreliable and ambiguous pixels. Intuitively, an unreliable prediction may get confused among the top classes, however, it should be confident about the pixel not belonging to the remaining classes. Hence, such a pixel can be convincingly treated as a negative key to those most unlikely categories. Therefore, we develop an effective pipeline to make sufficient use of unlabeled data. Concretely, we separate reliable and unreliable pixels via the entropy of predictions, push each unreliable pixel to a category-wise queue that consists of negative keys, and manage to train the model with all candidate pixels. Considering the training evolution, we adaptively adjust the threshold for the reliable-unreliable partition. Experimental results on various benchmarks and training settings demonstrate the superiority of our approach over the state-of-the-art alternatives.
Authors: Yuan Liu, Haodong Duan, Yuanhan Zhang, Bo Li, Songyang Zhang, Wangbo Zhao, Yike Yuan, Jiaqi Wang, Conghui He, Ziwei Liu, Kai Chen, Dahua Lin
Abstract: Large vision-language models (VLMs) have recently achieved remarkable progress, exhibiting impressive multimodal perception and reasoning abilities. However, effectively evaluating these large VLMs remains a major challenge, hindering future development in this domain. Traditional benchmarks like VQAv2 or COCO Caption provide quantitative performance measurements but lack fine-grained ability assessment and robust evaluation metrics. Meanwhile, subjective benchmarks, such as OwlEval, offer comprehensive evaluations of a model's abilities by incorporating human labor, which is not scalable and may display significant bias. In response to these challenges, we propose MMBench, a bilingual benchmark for assessing the multi-modal capabilities of VLMs. MMBench methodically develops a comprehensive evaluation pipeline, primarily comprised of the following key features: 1. MMBench is meticulously curated with well-designed quality control schemes, surpassing existing similar benchmarks in terms of the number and variety of evaluation questions and abilities; 2. MMBench introduces a rigorous CircularEval strategy and incorporates large language models to convert free-form predictions into pre-defined choices, which helps to yield accurate evaluation results for models with limited instruction-following capabilities. 3. MMBench incorporates multiple-choice questions in both English and Chinese versions, enabling an apples-to-apples comparison of VLMs' performance under a bilingual context. To summarize, MMBench is a systematically designed objective benchmark for a robust and holistic evaluation of vision-language models. We hope MMBench will assist the research community in better evaluating their models and facilitate future progress in this area. The evalutation code of MMBench has been integrated into VLMEvalKit: https://github.com/open-compass/VLMEvalKit.
Authors: M. M. Amaan Valiuddin, Christiaan G. A. Viviers, Ruud J. G. van Sloun, Peter H. N. de With, Fons van der Sommen
Abstract: Data uncertainties, such as sensor noise, occlusions or limitations in the acquisition method can introduce irreducible ambiguities in images, which result in varying, yet plausible, semantic hypotheses. In Machine Learning, this ambiguity is commonly referred to as aleatoric uncertainty. In image segmentation, latent density models can be utilized to address this problem. The most popular approach is the Probabilistic U-Net (PU-Net), which uses latent Normal densities to optimize the conditional data log-likelihood Evidence Lower Bound. In this work, we demonstrate that the PU-Net latent space is severely sparse and heavily under-utilized. To address this, we introduce mutual information maximization and entropy-regularized Sinkhorn Divergence in the latent space to promote homogeneity across all latent dimensions, effectively improving gradient-descent updates and latent space informativeness. Our results show that by applying this on public datasets of various clinical segmentation problems, our proposed methodology receives up to 11% performance gains compared against preceding latent variable models for probabilistic segmentation on the Hungarian-Matched Intersection over Union. The results indicate that encouraging a homogeneous latent space significantly improves latent density modeling for medical image segmentation.
Authors: Xindi Wu, Byron Zhang, Zhiwei Deng, Olga Russakovsky
Abstract: Dataset distillation methods reduce large-scale datasets to smaller sets of synthetic data, preserving sufficient information to quickly train a new model from scratch. However, prior work on dataset distillation has focused exclusively on image classification datasets, whereas modern large-scale datasets are primarily vision-language datasets. In this work, we design the first vision-language dataset distillation method, building on the idea of trajectory matching. A key challenge is that vision-language datasets do not have a set of discrete classes. To overcome this, our proposed method jointly distills image-text pairs in a contrastive formulation. Further, we leverage Low-Rank Adaptation (LoRA) matching to enable more efficient and effective trajectory matching in complex modern vision-language models. Since there are no existing baselines, we compare our distillation approach with three adapted vision-language coreset selection methods. We demonstrate significant improvements on the challenging Flickr30K and COCO retrieval benchmarks: for example, on Flickr30K, the best coreset selection method selecting 1000 image-text pairs for training achieves only 5.6% image-to-text retrieval accuracy (i.e., recall@1); in contrast, our dataset distillation almost doubles that to 9.9% with just 100 training pairs, an order of magnitude fewer.
Authors: Ali Zoljodi, Sadegh Abadijou, Mina Alibeigi, Masoud Daneshtalab
Abstract: Detecting lane markings in road scenes poses a challenge due to their intricate nature, which is susceptible to unfavorable conditions. While lane markings have strong shape priors, their visibility is easily compromised by lighting conditions, occlusions by other vehicles or pedestrians, and fading of colors over time. The detection process is further complicated by the presence of several lane shapes and natural variations, necessitating large amounts of data to train a robust lane detection model capable of handling various scenarios. In this paper, we present a novel self-supervised learning method termed Contrastive Learning for Lane Detection via cross-similarity (CLLD) to enhance the resilience of lane detection models in real-world scenarios, particularly when the visibility of lanes is compromised. CLLD introduces a contrastive learning (CL) method that assesses the similarity of local features within the global context of the input image. It uses the surrounding information to predict lane markings. This is achieved by integrating local feature contrastive learning with our proposed cross-similar operation. The local feature CL concentrates on extracting features from small patches, a necessity for accurately localizing lane segments. Meanwhile, cross-similarity captures global features, enabling the detection of obscured lane segments based on their surroundings. We enhance cross-similarity by randomly masking portions of input images in the process of augmentation. Extensive experiments on TuSimple and CuLane benchmarks demonstrate that CLLD outperforms SOTA contrastive learning methods, particularly in visibility-impairing conditions like shadows, while it also delivers comparable results under normal conditions. Compared to supervised learning, CLLD still excels in challenging scenarios such as shadows and crowded scenes, which are common in real-world driving.
Authors: Dongshen Han, Chaoning Zhang, Sheng Zheng, Chang Lu, Yang Yang, Heng Tao Shen
Abstract: As Segment Anything Model (SAM) becomes a popular foundation model in computer vision, its adversarial robustness has become a concern that cannot be ignored. This works investigates whether it is possible to attack SAM with image-agnostic Universal Adversarial Perturbation (UAP). In other words, we seek a single perturbation that can fool the SAM to predict invalid masks for most (if not all) images. We demonstrate convetional image-centric attack framework is effective for image-independent attacks but fails for universal adversarial attack. To this end, we propose a novel perturbation-centric framework that results in a UAP generation method based on self-supervised contrastive learning (CL), where the UAP is set to the anchor sample and the positive sample is augmented from the UAP. The representations of negative samples are obtained from the image encoder in advance and saved in a memory bank. The effectiveness of our proposed CL-based UAP generation method is validated by both quantitative and qualitative results. On top of the ablation study to understand various components in our proposed method, we shed light on the roles of positive and negative samples in making the generated UAP effective for attacking SAM.
Authors: Francisco Eiras, Kemal Oksuz, Adel Bibi, Philip H. S. Torr, Puneet K. Dokania
Abstract: Referring Image Segmentation (RIS) - the problem of identifying objects in images through natural language sentences - is a challenging task currently mostly solved through supervised learning. However, while collecting referred annotation masks is a time-consuming process, the few existing weakly-supervised and zero-shot approaches fall significantly short in performance compared to fully-supervised learning ones. To bridge the performance gap without mask annotations, we propose a novel weakly-supervised framework that tackles RIS by decomposing it into three steps: obtaining instance masks for the object mentioned in the referencing instruction (segment), using zero-shot learning to select a potentially correct mask for the given instruction (select), and bootstrapping a model which allows for fixing the mistakes of zero-shot selection (correct). In our experiments, using only the first two steps (zero-shot segment and select) outperforms other zero-shot baselines by as much as 16.5%, while our full method improves upon this much stronger baseline and sets the new state-of-the-art for weakly-supervised RIS, reducing the gap between the weakly-supervised and fully-supervised methods in some cases from around 33% to as little as 7%. Code is available at https://github.com/fgirbal/segment-select-correct.
Authors: Zhe Zhang, Gaochang Wu, Jing Zhang, Xiatian Zhu, Dacheng Tao, Tianyou Chai
Abstract: Unsupervised Domain Adaptive Semantic Segmentation (UDA-SS) aims to transfer the supervision from a labeled source domain to an unlabeled target domain. The majority of existing UDA-SS works typically consider images whilst recent attempts have extended further to tackle videos by modeling the temporal dimension. Although the two lines of research share the major challenges -- overcoming the underlying domain distribution shift, their studies are largely independent, resulting in fragmented insights, a lack of holistic understanding, and missed opportunities for cross-pollination of ideas. This fragmentation prevents the unification of methods, leading to redundant efforts and suboptimal knowledge transfer across image and video domains. Under this observation, we advocate unifying the study of UDA-SS across video and image scenarios, enabling a more comprehensive understanding, synergistic advancements, and efficient knowledge sharing. To that end, we explore the unified UDA-SS from a general data augmentation perspective, serving as a unifying conceptual framework, enabling improved generalization, and potential for cross-pollination of ideas, ultimately contributing to the overall progress and practical impact of this field of research. Specifically, we propose a Quad-directional Mixup (QuadMix) method, characterized by tackling distinct point attributes and feature inconsistencies through four-directional paths for intra- and inter-domain mixing in a feature space. To deal with temporal shifts with videos, we incorporate optical flow-guided feature aggregation across spatial and temporal dimensions for fine-grained domain alignment. Extensive experiments show that our method outperforms the state-of-the-art works by large margins on four challenging UDA-SS benchmarks. Our source code and models will be released at \url{https://github.com/ZHE-SAPI/UDASS}.
Authors: Luigi Riz, Cristiano Saltori, Yiming Wang, Elisa Ricci, Fabio Poiesi
Abstract: The task of Novel Class Discovery (NCD) in semantic segmentation entails training a model able to accurately segment unlabelled (novel) classes, relying on the available supervision from annotated (base) classes. Although extensively investigated in 2D image data, the extension of the NCD task to the domain of 3D point clouds represents a pioneering effort, characterized by assumptions and challenges that are not present in the 2D case. This paper represents an advancement in the analysis of point cloud data in four directions. Firstly, it introduces the novel task of NCD for point cloud semantic segmentation. Secondly, it demonstrates that directly transposing the only existing NCD method for 2D image semantic segmentation to 3D data yields suboptimal results. Thirdly, a new NCD approach based on online clustering, uncertainty estimation, and semantic distillation is presented. Lastly, a novel evaluation protocol is proposed to rigorously assess the performance of NCD in point cloud semantic segmentation. Through comprehensive evaluations on the SemanticKITTI, SemanticPOSS, and S3DIS datasets, the paper demonstrates substantial superiority of the proposed method over the considered baselines.
Authors: Honghao Fu, Zhiqi Shen, Jing Jih Chin, Hao Wang
Abstract: Analyzing and reconstructing visual stimuli from brain signals effectively advances the understanding of human visual system. However, the EEG signals are complex and contain significant noise. This leads to substantial limitations in existing works of visual stimuli reconstruction from EEG, such as difficulties in aligning EEG embeddings with the fine-grained semantic information and a heavy reliance on additional large self-collected dataset for training. To address these challenges, we propose a novel approach called BrainVis. Firstly, we divide the EEG signals into various units and apply a self-supervised approach on them to obtain EEG time-domain features, in an attempt to ease the training difficulty. Additionally, we also propose to utilize the frequency-domain features to enhance the EEG representations. Then, we simultaneously align EEG time-frequency embeddings with the interpolation of the coarse and fine-grained semantics in the CLIP space, to highlight the primary visual components and reduce the cross-modal alignment difficulty. Finally, we adopt the cascaded diffusion models to reconstruct images. Using only 10\% training data of the previous work, our proposed BrainVis outperforms state of the arts in both semantic fidelity reconstruction and generation quality. The code is available at https://github.com/RomGai/BrainVis.
Authors: Hao Yang, Hong-Yu Zhou, Cheng Li, Weijian Huang, Jiarun Liu, Yong Liang, Guangming Shi, Hairong Zheng, Qiegen Liu, Shanshan Wang
Abstract: Multimodal deep learning utilizing imaging and diagnostic reports has made impressive progress in the field of medical imaging diagnostics, demonstrating a particularly strong capability for auxiliary diagnosis in cases where sufficient annotation information is lacking. Nonetheless, localizing diseases accurately without detailed positional annotations remains a challenge. Although existing methods have attempted to utilize local information to achieve fine-grained semantic alignment, their capability in extracting the fine-grained semantics of the comprehensive context within reports is limited. To address this problem, a new method is introduced that takes full sentences from textual reports as the basic units for local semantic alignment. This approach combines chest X-ray images with their corresponding textual reports, performing contrastive learning at both global and local levels. The leading results obtained by this method on multiple datasets confirm its efficacy in the task of lesion localization.
Authors: Xianfu Cheng, Weixiao Zhou, Xiang Li, Jian Yang, Hang Zhang, Tao Sun, Wei Zhang, Yuying Mai, Tongliang Li, Xiaoming Chen, Zhoujun Li
Abstract: Scene Text Recognition (STR) is an important and challenging upstream task for building structured information databases, that involves recognizing text within images of natural scenes. Although current state-of-the-art (SOTA) models for STR exhibit high performance, they typically suffer from low inference efficiency due to their reliance on hybrid architectures comprised of visual encoders and sequence decoders. In this work, we propose a VIsion Permutable extractor for fast and efficient Scene Text Recognition (SVIPTR), which achieves an impressive balance between high performance and rapid inference speeds in the domain of STR. Specifically, SVIPTR leverages a visual-semantic extractor with a pyramid structure, characterized by the Permutation and combination of local and global self-attention layers. This design results in a lightweight and efficient model and its inference is insensitive to input length. Extensive experimental results on various standard datasets for both Chinese and English scene text recognition validate the superiority of SVIPTR. Notably, the SVIPTR-T (Tiny) variant delivers highly competitive accuracy on par with other lightweight models and achieves SOTA inference speeds. Meanwhile, the SVIPTR-L (Large) attains SOTA accuracy in single-encoder-type models, while maintaining a low parameter count and favorable inference speed. Our proposed method provides a compelling solution for the STR challenge, which greatly benefits real-world applications requiring fast and efficient STR. The code is publicly available at https://github.com/cxfyxl/VIPTR.
Authors: Pinxin Liu, Luchuan Song, Daoan Zhang, Hang Hua, Yunlong Tang, Huaijin Tu, Jiebo Luo, Chenliang Xu
Abstract: Existing methods like Neural Radiation Fields (NeRF) and 3D Gaussian Splatting (3DGS) have made significant strides in facial attribute control such as facial animation and components editing, yet they struggle with fine-grained representation and scalability in dynamic head modeling. To address these limitations, we propose GaussianStyle, a novel framework that integrates the volumetric strengths of 3DGS with the powerful implicit representation of StyleGAN. The GaussianStyle preserves structural information, such as expressions and poses, using Gaussian points, while projecting the implicit volumetric representation into StyleGAN to capture high-frequency details and mitigate the over-smoothing commonly observed in neural texture rendering. Experimental outcomes indicate that our method achieves state-of-the-art performance in reenactment, novel view synthesis, and animation.
Authors: Zhaoran Zhao, Peng Lu, Xujun Peng, Wenhao Guo
Abstract: In the domain of image layout representation learning, the critical process of translating image layouts into succinct vector forms is increasingly significant across diverse applications, such as image retrieval, manipulation, and generation. Most approaches in this area heavily rely on costly labeled datasets and notably lack in adapting their modeling and learning methods to the specific nuances of photographic image layouts. This shortfall makes the learning process for photographic image layouts suboptimal. In our research, we directly address these challenges. We innovate by defining basic layout primitives that encapsulate various levels of layout information and by mapping these, along with their interconnections, onto a heterogeneous graph structure. This graph is meticulously engineered to capture the intricate layout information within the pixel domain explicitly. Advancing further, we introduce novel pretext tasks coupled with customized loss functions, strategically designed for effective self-supervised learning of these layout graphs. Building on this foundation, we develop an autoencoder-based network architecture skilled in compressing these heterogeneous layout graphs into precise, dimensionally-reduced layout representations. Additionally, we introduce the LODB dataset, which features a broader range of layout categories and richer semantics, serving as a comprehensive benchmark for evaluating the effectiveness of layout representation learning methods. Our extensive experimentation on this dataset demonstrates the superior performance of our approach in the realm of photographic image layout representation learning.
Authors: Yuhu Bai, Jiangning Zhang, Zhaofeng Chen, Yuhang Dong, Yunkang Cao, Guanzhong Tian
Abstract: Few-shot anomaly detection (FSAD) plays a crucial role in industrial manufacturing. However, existing FSAD methods encounter difficulties leveraging a limited number of normal samples, frequently failing to detect and locate inconspicuous anomalies in the spatial domain. We have further discovered that these subtle anomalies would be more noticeable in the frequency domain. In this paper, we propose a Dual-Path Frequency Discriminators (DFD) network from a frequency perspective to tackle these issues. The original spatial images are transformed into multi-frequency images, making them more conducive to the tailored discriminators in detecting anomalies. Additionally, the discriminators learn a joint representation with forms of pseudo-anomalies. Extensive experiments conducted on MVTec AD and VisA benchmarks demonstrate that our DFD surpasses current state-of-the-art methods. Source code will be available.
Authors: Yibin Wang, Weizhong Zhang, Jianwei Zheng, Cheng Jin
Abstract: Image composition involves seamlessly integrating given objects into a specific visual context. Current training-free methods rely on composing attention weights from several samplers to guide the generator. However, since these weights are derived from disparate contexts, their combination leads to coherence confusion and loss of appearance information. These issues worsen with their excessive focus on background generation, even when unnecessary in this task. This not only impedes their swift implementation but also compromises foreground generation quality. Moreover, these methods introduce unwanted artifacts in the transition area. In this paper, we formulate image composition as a subject-based local editing task, solely focusing on foreground generation. At each step, the edited foreground is combined with the noisy background to maintain scene consistency. To address the remaining issues, we propose PrimeComposer, a faster training-free diffuser that composites the images by well-designed attention steering across different noise levels. This steering is predominantly achieved by our Correlation Diffuser, utilizing its self-attention layers at each step. Within these layers, the synthesized subject interacts with both the referenced object and background, capturing intricate details and coherent relationships. This prior information is encoded into the attention weights, which are then integrated into the self-attention layers of the generator to guide the synthesis process. Besides, we introduce a Region-constrained Cross-Attention to confine the impact of specific subject-related tokens to desired regions, addressing the unwanted artifacts shown in the prior method thereby further improving the coherence in the transition area. Our method exhibits the fastest inference efficiency and extensive experiments demonstrate our superiority both qualitatively and quantitatively.
Authors: Feng Liu, Xiaobin Chang
Abstract: Semantic image synthesis aims to generate high-quality images given semantic conditions, i.e. segmentation masks and style reference images. Existing methods widely adopt generative adversarial networks (GANs). GANs take all conditional inputs and directly synthesize images in a single forward step. In this paper, semantic image synthesis is treated as an image denoising task and is handled with a novel image-to-image diffusion model (IIDM). Specifically, the style reference is first contaminated with random noise and then progressively denoised by IIDM, guided by segmentation masks. Moreover, three techniques, refinement, color-transfer and model ensembles, are proposed to further boost the generation quality. They are plug-in inference modules and do not require additional training. Extensive experiments show that our IIDM outperforms existing state-of-the-art methods by clear margins. Further analysis is provided via detailed demonstrations. We have implemented IIDM based on the Jittor framework; code is available at https://github.com/ader47/jittor-jieke-semantic_images_synthesis.
URLs: https://github.com/ader47/jittor-jieke-semantic_images_synthesis.
Authors: Haoxing Chen, Yaohui Li, Zizheng Huang, Yan Hong, Zhuoer Xu, Zhangxuan Gu, Jun Lan, Huijia Zhu, Weiqiang Wang
Abstract: Pre-trained large-scale vision-language models (VLMs) have acquired profound understanding of general visual concepts. Recent advancements in efficient transfer learning (ETL) have shown remarkable success in fine-tuning VLMs within the scenario of limited data, introducing only a few parameters to harness task-specific insights from VLMs. Despite significant progress, current leading ETL methods tend to overfit the narrow distributions of base classes seen during training and encounter two primary challenges: (i) only utilizing uni-modal information to modeling task-specific knowledge; and (ii) using costly and time-consuming methods to supplement knowledge. To address these issues, we propose a Conditional Prototype Rectification Prompt Learning (CPR) method to correct the bias of base examples and augment limited data in an effective way. Specifically, we alleviate overfitting on base classes from two aspects. First, each input image acquires knowledge from both textual and visual prototypes, and then generates sample-conditional text tokens. Second, we extract utilizable knowledge from unlabeled data to further refine the prototypes. These two strategies mitigate biases stemming from base classes, yielding a more effective classifier. Extensive experiments on 11 benchmark datasets show that our CPR achieves state-of-the-art performance on both few-shot classification and base-to-new generalization tasks. Our code is avaliable at \url{https://github.com/chenhaoxing/CPR}.
Authors: Hanchi Sun, Xiaohong Liu, Xinyang Jiang, Yifei Shen, Dongsheng Li, Xiongkuo Min, Guangtao Zhai
Abstract: This paper focuses on the task of quality enhancement for compressed videos. Although deep network-based video restorers achieve impressive progress, most of the existing methods lack a structured design to optimally leverage the priors within compression codecs. Since the quality degradation of the video is primarily induced by the compression algorithm, a new paradigm is urgently needed for a more ``conscious'' process of quality enhancement. As a result, we propose the Compression-Realized Deep Structural Network (CRDS), introducing three inductive biases aligned with the three primary processes in the classic compression codec, merging the strengths of classical encoder architecture with deep network capabilities. Inspired by the residual extraction and domain transformation process in the codec, a pre-trained Latent Degradation Residual Auto-Encoder is proposed to transform video frames into a latent feature space, and the mutual neighborhood attention mechanism is integrated for precise motion estimation and residual extraction. Furthermore, drawing inspiration from the quantization noise distribution of the codec, CRDS proposes a novel Progressive Denoising framework with intermediate supervision that decomposes the quality enhancement into a series of simpler denoising sub-tasks. Experimental results on datasets like LDV 2.0 and MFQE 2.0 indicate our approach surpasses state-of-the-art models.
Authors: Tomoya Sugihara, Shuntaro Masuda, Ling Xiao, Toshihiko Yamasaki
Abstract: Current video summarization methods rely heavily on supervised computer vision techniques, which demands time-consuming and subjective manual annotations. To overcome these limitations, we investigated self-supervised video summarization. Inspired by the success of Large Language Models (LLMs), we explored the feasibility in transforming the video summarization task into a Natural Language Processing (NLP) task. By leveraging the advantages of LLMs in context understanding, we aim to enhance the effectiveness of self-supervised video summarization. Our method begins by generating captions for individual video frames, which are then synthesized into text summaries by LLMs. Subsequently, we measure semantic distance between the captions and the text summary. Notably, we propose a novel loss function to optimize our model according to the diversity of the video. Finally, the summarized video can be generated by selecting the frames with captions similar to the text summary. Our method achieves state-of-the-art performance on the SumMe dataset in rank correlation coefficients. In addition, our method has a novel feature of being able to achieve personalized summarization.
Authors: Haonan An, Guang Hua, Zhiping Lin, Yuguang Fang
Abstract: Box-free model watermarking is an emerging technique to safeguard the intellectual property of deep learning models, particularly those for low-level image processing tasks. Existing works have verified and improved its effectiveness in several aspects. However, in this paper, we reveal that box-free model watermarking is prone to removal attacks, even under the real-world threat model such that the protected model and the watermark extractor are in black boxes. Under this setting, we carry out three studies. 1) We develop an extractor-gradient-guided (EGG) remover and show its effectiveness when the extractor uses ReLU activation only. 2) More generally, for an unknown extractor, we leverage adversarial attacks and design the EGG remover based on the estimated gradients. 3) Under the most stringent condition that the extractor is inaccessible, we design a transferable remover based on a set of private proxy models. In all cases, the proposed removers can successfully remove embedded watermarks while preserving the quality of the processed images, and we also demonstrate that the EGG remover can even replace the watermarks. Extensive experimental results verify the effectiveness and generalizability of the proposed attacks, revealing the vulnerabilities of the existing box-free methods and calling for further research.
Authors: Chuanhao Li, Zhen Li, Chenchen Jing, Shuo Liu, Wenqi Shao, Yuwei Wu, Ping Luo, Yu Qiao, Kaipeng Zhang
Abstract: Large vision-language models (LVLMs) are ignorant of the up-to-date knowledge, such as LLaVA series, because they cannot be updated frequently due to the large amount of resources required, and therefore fail in many cases. For example, if a LVLM was released on January 2024, and it wouldn't know the singer of the theme song for the new Detective Conan movie, which wasn't released until April 2024. To solve the problem, a promising solution motivated by retrieval-augmented generation (RAG) is to provide LVLMs with up-to-date knowledge via internet search during inference, i.e., internet-augmented generation (IAG), which is already integrated in some closed-source commercial LVLMs such as GPT-4V. However, the specific mechanics underpinning them remain a mystery. In this paper, we propose a plug-and-play framework, for augmenting existing LVLMs in handling visual question answering (VQA) about up-to-date knowledge, dubbed SearchLVLMs. A hierarchical filtering model is trained to effectively and efficiently find the most helpful content from the websites returned by a search engine to prompt LVLMs with up-to-date knowledge. To train the model and evaluate our framework's performance, we propose a pipeline to automatically generate news-related VQA samples to construct a dataset, dubbed UDK-VQA. A multi-model voting mechanism is introduced to label the usefulness of website/content for VQA samples to construct the training set. Experimental results demonstrate the effectiveness of our framework, outperforming GPT-4V by about 25% in accuracy.
Authors: Chen Wang, Jiatao Gu, Xiaoxiao Long, Yuan Liu, Lingjie Liu
Abstract: Recent years have seen significant advancements in 3D generation. While methods like score distillation achieve impressive results, they often require extensive per-scene optimization, which limits their time efficiency. On the other hand, reconstruction-based approaches are more efficient but tend to compromise quality due to their limited ability to handle uncertainty. We introduce GECO, a novel method for high-quality 3D generative modeling that operates within a second. Our approach addresses the prevalent issues of uncertainty and inefficiency in existing methods through a two-stage approach. In the first stage, we train a single-step multi-view generative model with score distillation. Then, a second-stage distillation is applied to address the challenge of view inconsistency in the multi-view generation. This two-stage process ensures a balanced approach to 3D generation, optimizing both quality and efficiency. Our comprehensive experiments demonstrate that GECO achieves high-quality image-to-3D mesh generation with an unprecedented level of efficiency. We will make the code and model publicly available.
Authors: Yudi Ruan, Hao Ma, Weikai Li, Xiao Wang
Abstract: Low-light image enhancement (LLIE) is critical in computer vision. Existing LLIE methods often fail to discover the underlying relationships between different sub-components, causing the loss of complementary information between multiple modules and network layers, ultimately resulting in the loss of image details. To beat this shortage, we design a hierarchical mutual Enhancement via a Cross Attention transformer (ECAFormer), which introduces an architecture that enables concurrent propagation and interaction of multiple features. The model preserves detailed information by introducing a Dual Multi-head self-attention (DMSA), which leverages visual and semantic features across different scales, allowing them to guide and complement each other. Besides, a Cross-Scale DMSA block is introduced to capture the residual connection, integrating cross-layer information to further enhance image detail. Experimental results show that ECAFormer reaches competitive performance across multiple benchmarks, yielding nearly a 3% improvement in PSNR over the suboptimal method, demonstrating the effectiveness of information interaction in LLIE.
Authors: Xiaolin Hong, Hongwei Yi, Fazhi He, Qiong Cao
Abstract: Generating 3D scenes from human motion sequences supports numerous applications, including virtual reality and architectural design. However, previous auto-regression-based human-aware 3D scene generation methods have struggled to accurately capture the joint distribution of multiple objects and input humans, often resulting in overlapping object generation in the same space. To address this limitation, we explore the potential of diffusion models that simultaneously consider all input humans and the floor plan to generate plausible 3D scenes. Our approach not only satisfies all input human interactions but also adheres to spatial constraints with the floor plan. Furthermore, we introduce two spatial collision guidance mechanisms: human-object collision avoidance and object-room boundary constraints. These mechanisms help avoid generating scenes that conflict with human motions while respecting layout constraints. To enhance the diversity and accuracy of human-guided scene generation, we have developed an automated pipeline that improves the variety and plausibility of human-object interactions in the existing 3D FRONT HUMAN dataset. Extensive experiments on both synthetic and real-world datasets demonstrate that our framework can generate more natural and plausible 3D scenes with precise human-scene interactions, while significantly reducing human-object collisions compared to previous state-of-the-art methods. Our code and data will be made publicly available upon publication of this work.
Authors: Risa Shinoda, Kaede Shiohara
Abstract: Automated animal face identification plays a crucial role in the monitoring of behaviors, conducting of surveys, and finding of lost animals. Despite the advancements in human face identification, the lack of datasets and benchmarks in the animal domain has impeded progress. In this paper, we introduce the PetFace dataset, a comprehensive resource for animal face identification encompassing 257,484 unique individuals across 13 animal families and 319 breed categories, including both experimental and pet animals. This large-scale collection of individuals facilitates the investigation of unseen animal face verification, an area that has not been sufficiently explored in existing datasets due to the limited number of individuals. Moreover, PetFace also has fine-grained annotations such as sex, breed, color, and pattern. We provide multiple benchmarks including re-identification for seen individuals and verification for unseen individuals. The models trained on our dataset outperform those trained on prior datasets, even for detailed breed variations and unseen animal families. Our result also indicates that there is some room to improve the performance of integrated identification on multiple animal families. We hope the PetFace dataset will facilitate animal face identification and encourage the development of non-invasive animal automatic identification methods.
Authors: Feichi Lu, Zijian Dong, Jie Song, Otmar Hilliges
Abstract: Despite progress in human motion capture, existing multi-view methods often face challenges in estimating the 3D pose and shape of multiple closely interacting people. This difficulty arises from reliance on accurate 2D joint estimations, which are hard to obtain due to occlusions and body contact when people are in close interaction. To address this, we propose a novel method leveraging the personalized implicit neural avatar of each individual as a prior, which significantly improves the robustness and precision of this challenging pose estimation task. Concretely, the avatars are efficiently reconstructed via layered volume rendering from sparse multi-view videos. The reconstructed avatar prior allows for the direct optimization of 3D poses based on color and silhouette rendering loss, bypassing the issues associated with noisy 2D detections. To handle interpenetration, we propose a collision loss on the overlapping shape regions of avatars to add penetration constraints. Moreover, both 3D poses and avatars are optimized in an alternating manner. Our experimental results demonstrate state-of-the-art performance on several public datasets.
Authors: Mahadev Prasad Panda, Matteo Tiezzi, Martina Vilas, Gemma Roig, Bjoern M. Eskofier, Dario Zanca
Abstract: We introduce Foveation-based Explanations (FovEx), a novel human-inspired visual explainability (XAI) method for Deep Neural Networks. Our method achieves state-of-the-art performance on both transformer (on 4 out of 5 metrics) and convolutional models (on 3 out of 5 metrics), demonstrating its versatility. Furthermore, we show the alignment between the explanation map produced by FovEx and human gaze patterns (+14\% in NSS compared to RISE, +203\% in NSS compared to gradCAM), enhancing our confidence in FovEx's ability to close the interpretation gap between humans and machines.
Authors: Bin Hu, Xinggang Wang, Wenyu Liu
Abstract: Person Re-Identification (ReID) aims to retrieve relevant individuals in non-overlapping camera images and has a wide range of applications in the field of public safety. In recent years, with the development of Vision Transformer (ViT) and self-supervised learning techniques, the performance of person ReID based on self-supervised pre-training has been greatly improved. Person ReID requires extracting highly discriminative local fine-grained features of the human body, while traditional ViT is good at extracting context-related global features, making it difficult to focus on local human body features. To this end, this article introduces the recently emerged Masked Image Modeling (MIM) self-supervised learning method into person ReID, and effectively extracts high-quality global and local features through large-scale unsupervised pre-training by combining masked image modeling and discriminative contrastive learning, and then conducts supervised fine-tuning training in the person ReID task. This person feature extraction method based on ViT with masked image modeling (PersonViT) has the good characteristics of unsupervised, scalable, and strong generalization capabilities, overcoming the problem of difficult annotation in supervised person ReID, and achieves state-of-the-art results on publicly available benchmark datasets, including MSMT17, Market1501, DukeMTMC-reID, and Occluded-Duke. The code and pre-trained models of the PersonViT method are released at \url{https://github.com/hustvl/PersonViT} to promote further research in the person ReID field.
Authors: Yiying Yang, Fukun Yin, Jiayuan Fan, Xin Chen, Wanzhang Li, Gang Yu
Abstract: As Artificial Intelligence Generated Content (AIGC) advances, a variety of methods have been developed to generate text, images, videos, and 3D objects from single or multimodal inputs, contributing efforts to emulate human-like cognitive content creation. However, generating realistic large-scale scenes from a single input presents a challenge due to the complexities involved in ensuring consistency across extrapolated views generated by models. Benefiting from recent video generation models and implicit neural representations, we propose Scene123, a 3D scene generation model, that not only ensures realism and diversity through the video generation framework but also uses implicit neural fields combined with Masked Autoencoders (MAE) to effectively ensures the consistency of unseen areas across views. Specifically, we initially warp the input image (or an image generated from text) to simulate adjacent views, filling the invisible areas with the MAE model. However, these filled images usually fail to maintain view consistency, thus we utilize the produced views to optimize a neural radiance field, enhancing geometric consistency. Moreover, to further enhance the details and texture fidelity of generated views, we employ a GAN-based Loss against images derived from the input image through the video generation model. Extensive experiments demonstrate that our method can generate realistic and consistent scenes from a single prompt. Both qualitative and quantitative results indicate that our approach surpasses existing state-of-the-art methods. We show encourage video examples at https://yiyingyang12.github.io/Scene123.github.io/.
Authors: Jakub Kol\'a\v{r}, Radim \v{S}petl\'ik, Ji\v{r}\'i Matas
Abstract: We present a novel method for measuring the period of phenomena like rotation, flicker and vibration, by an event camera, a device asynchronously reporting brightness changes at independently operating pixels with high temporal resolution. The approach assumes that for a periodic phenomenon, a highly similar set of events is generated within a spatio-temporal window at a time difference corresponding to its period. The sets of similar events are detected by a correlation in the spatio-temporal event stream space. The proposed method, EEPPR, is evaluated on a dataset of 12 sequences of periodic phenomena, i.e. flashing light and vibration, and periodic motion, e.g., rotation, ranging from 3.2 Hz to 2 kHz (equivalent to 192 - 120 000 RPM). EEPPR significantly outperforms published methods on this dataset, achieving the mean relative error of 0.1%. The dataset and codes are publicly available on GitHub.
Authors: Hossein Jafarinia, Alireza Alipanah, Danial Hamdi, Saeed Razavi, Nahal Mirzaie, Mohammad Hossein Rohban
Abstract: Whole Slide Image (WSI) classification with multiple instance learning (MIL) in digital pathology faces significant computational challenges. Current methods mostly rely on extensive self-supervised learning (SSL) for satisfactory performance, requiring long training periods and considerable computational resources. At the same time, no pre-training affects performance due to domain shifts from natural images to WSIs. We introduce Snuffy architecture, a novel MIL-pooling method based on sparse transformers that mitigates performance loss with limited pre-training and enables continual few-shot pre-training as a competitive option. Our sparsity pattern is tailored for pathology and is theoretically proven to be a universal approximator with the tightest probabilistic sharp bound on the number of layers for sparse transformers, to date. We demonstrate Snuffy's effectiveness on CAMELYON16 and TCGA Lung cancer datasets, achieving superior WSI and patch-level accuracies. The code is available on https://github.com/jafarinia/snuffy.
Authors: Rongxuan Wang, Xin Lu, Xiaoyang Liu, Xiaoyi Zou, Tongyi Cao, Ying Li
Abstract: Online vectorized High-Definition (HD) map construction is crucial for subsequent prediction and planning tasks in autonomous driving. Following MapTR paradigm, recent works have made noteworthy achievements. However, reference points are randomly initialized in mainstream methods, leading to unstable matching between predictions and ground truth. To address this issue, we introduce PriorMapNet to enhance online vectorized HD map construction with priors. We propose the PPS-Decoder, which provides reference points with position and structure priors. Fitted from the map elements in the dataset, prior reference points lower the learning difficulty and achieve stable matching. Furthermore, we propose the PF-Encoder to enhance the image-to-BEV transformation with BEV feature priors. Besides, we propose the DMD cross-attention, which decouples cross-attention along multi-scale and multi-sample respectively to achieve efficiency. Our proposed PriorMapNet achieves state-of-the-art performance in the online vectorized HD map construction task on nuScenes and Argoverse2 datasets. The code will be released publicly soon.
Authors: Sheng Ye, Zhen-Hui Dong, Yubin Hu, Yu-Hui Wen, Yong-Jin Liu
Abstract: 3D Gaussian Splatting has recently emerged as a powerful representation that can synthesize remarkable novel views using consistent multi-view images as input. However, we notice that images captured in dark environments where the scenes are not fully illuminated can exhibit considerable brightness variations and multi-view inconsistency, which poses great challenges to 3D Gaussian Splatting and severely degrades its performance. To tackle this problem, we propose Gaussian-DK. Observing that inconsistencies are mainly caused by camera imaging, we represent a consistent radiance field of the physical world using a set of anisotropic 3D Gaussians, and design a camera response module to compensate for multi-view inconsistencies. We also introduce a step-based gradient scaling strategy to constrain Gaussians near the camera, which turn out to be floaters, from splitting and cloning. Experiments on our proposed benchmark dataset demonstrate that Gaussian-DK produces high-quality renderings without ghosting and floater artifacts and significantly outperforms existing methods. Furthermore, we can also synthesize light-up images by controlling exposure levels that clearly show details in shadow areas.
Authors: Zhaoli Deng, Wen Liu, Fanyi Wang, Junkang Zhang, Fan Chen, Meng Zhang, Wendong Zhang, Zhenpeng Mi
Abstract: Portrait Fidelity Generation is a prominent research area in generative models, with a primary focus on enhancing both controllability and fidelity. Current methods face challenges in generating high-fidelity portrait results when faces occupy a small portion of the image with a low resolution, especially in multi-person group photo settings. To tackle these issues, we propose a systematic solution called MagicID, based on a self-constructed million-level multi-modal dataset named IDZoom. MagicID consists of Multi-Mode Fusion training strategy (MMF) and DDIM Inversion based ID Restoration inference framework (DIIR). During training, MMF iteratively uses the skeleton and landmark modalities from IDZoom as conditional guidance. By introducing the Clone Face Tuning in training stage and Mask Guided Multi-ID Cross Attention (MGMICA) in inference stage, explicit constraints on face positional features are achieved for multi-ID group photo generation. The DIIR aims to address the issue of artifacts. The DDIM Inversion is used in conjunction with face landmarks, global and local face features to achieve face restoration while keeping the background unchanged. Additionally, DIIR is plug-and-play and can be applied to any diffusion-based portrait generation method. To validate the effectiveness of MagicID, we conducted extensive comparative and ablation experiments. The experimental results demonstrate that MagicID has significant advantages in both subjective and objective metrics, and achieves controllable generation in multi-person scenarios.
Authors: Anusree P. S., Bikram Keshari Parida, Seong Yong Moon, Wonsang You
Abstract: Cone Beam Computed Tomography (CBCT) and Panoramic X-rays are the most commonly used imaging modalities in dental health care. CBCT can produce three-dimensional views of a patient's head, providing clinicians with better diagnostic capability, whereas Panoramic X-ray can capture the entire maxillofacial region in a single image. If the CBCT is already available, it can be beneficial to synthesize a Panoramic X-ray, thereby avoiding an immediate additional scan and extra radiation exposure. Existing methods focus on delineating an approximate dental arch and creating orthogonal projections along this arch. However, no golden standard is available for such dental arch extractions, and this choice can affect the quality of synthesized X-rays. To avoid such issues, we propose a novel method for synthesizing Panoramic X-rays from diverse head CBCTs, employing a simulated projection geometry and dynamic rotation centers. Our method effectively synthesized panoramic views from CBCT, even for patients with missing or nonexistent teeth and in the presence of severe metal implants. Our results demonstrate that this method can generate high-quality panoramic images irrespective of the CBCT scanner geometry.
Authors: Haoyu Zhao, Hao Wang, Chen Yang, Wei Shen
Abstract: Recent advancements in human avatar synthesis have utilized radiance fields to reconstruct photo-realistic animatable human avatars. However, both NeRFs-based and 3DGS-based methods struggle with maintaining 3D consistency and exhibit suboptimal detail reconstruction, especially with sparse inputs. To address this challenge, we propose CHASE, which introduces supervision from intrinsic 3D consistency across poses and 3D geometry contrastive learning, achieving performance comparable with sparse inputs to that with full inputs. Following previous work, we first integrate a skeleton-driven rigid deformation and a non-rigid cloth dynamics deformation to coordinate the movements of individual Gaussians during animation, reconstructing basic avatar with coarse 3D consistency. To improve 3D consistency under sparse inputs, we design Dynamic Avatar Adjustment(DAA) to adjust deformed Gaussians based on a selected similar pose/image from the dataset. Minimizing the difference between the image rendered by adjusted Gaussians and the image with the similar pose serves as an additional form of supervision for avatar. Furthermore, we propose a 3D geometry contrastive learning strategy to maintain the 3D global consistency of generated avatars. Though CHASE is designed for sparse inputs, it surprisingly outperforms current SOTA methods \textbf{in both full and sparse settings} on the ZJU-MoCap and H36M datasets, demonstrating that our CHASE successfully maintains avatar's 3D consistency, hence improving rendering quality.
Authors: Jialu Wang, Kaichen Zhou, Andrew Markham, Niki Trigoni
Abstract: Location information is pivotal for the automation and intelligence of terminal devices and edge-cloud IoT systems, such as autonomous vehicles and augmented reality. However, achieving reliable positioning across diverse IoT applications remains challenging due to significant training costs and the necessity of densely collected data. To tackle these issues, we have innovatively applied the selective state space (SSM) model to visual localization, introducing a new model named MambaLoc. The proposed model demonstrates exceptional training efficiency by capitalizing on the SSM model's strengths in efficient feature extraction, rapid computation, and memory optimization, and it further ensures robustness in sparse data environments due to its parameter sparsity. Additionally, we propose the Global Information Selector (GIS), which leverages selective SSM to implicitly achieve the efficient global feature extraction capabilities of Non-local Neural Networks. This design leverages the computational efficiency of the SSM model alongside the Non-local Neural Networks' capacity to capture long-range dependencies with minimal layers. Consequently, the GIS enables effective global information capture while significantly accelerating convergence. Our extensive experimental validation using public indoor and outdoor datasets first demonstrates our model's effectiveness, followed by evidence of its versatility with various existing localization models. Our code and models are publicly available to support further research and development in this area.
Authors: Liyuan Zhu, Yue Li, Erik Sandstr\"om, Shengyu Huang, Konrad Schindler, Iro Armeni
Abstract: Simultaneous Localization and Mapping (SLAM) based on 3D Gaussian Splats (3DGS) has recently shown promise towards more accurate, dense 3D scene maps. However, existing 3DGS-based methods fail to address the global consistency of the scene via loop closure and/or global bundle adjustment. To this end, we propose LoopSplat, which takes RGB-D images as input and performs dense mapping with 3DGS submaps and frame-to-model tracking. LoopSplat triggers loop closure online and computes relative loop edge constraints between submaps directly via 3DGS registration, leading to improvements in efficiency and accuracy over traditional global-to-local point cloud registration. It uses a robust pose graph optimization formulation and rigidly aligns the submaps to achieve global consistency. Evaluation on the synthetic Replica and real-world TUM-RGBD, ScanNet, and ScanNet++ datasets demonstrates competitive or superior tracking, mapping, and rendering compared to existing methods for dense RGB-D SLAM. Code is available at loopsplat.github.io.
Authors: Fuzhao Xue, Yukang Chen, Dacheng Li, Qinghao Hu, Ligeng Zhu, Xiuyu Li, Yunhao Fang, Haotian Tang, Shang Yang, Zhijian Liu, Ethan He, Hongxu Yin, Pavlo Molchanov, Jan Kautz, Linxi Fan, Yuke Zhu, Yao Lu, Song Han
Abstract: Long-context capability is critical for multi-modal foundation models, especially for long video understanding. We introduce LongVILA, a full-stack solution for long-context visual-language models by co-designing the algorithm and system. For model training, we upgrade existing VLMs to support long video understanding by incorporating two additional stages, i.e., long context extension and long supervised fine-tuning. However, training on long video is computationally and memory intensive. We introduce the long-context Multi-Modal Sequence Parallelism (MM-SP) system that efficiently parallelizes long video training and inference, enabling 2M context length training on 256 GPUs without any gradient checkpointing. LongVILA efficiently extends the number of video frames of VILA from 8 to 1024, improving the long video captioning score from 2.00 to 3.26 (out of 5), achieving 99.5% accuracy in 1400-frame (274k context length) video needle-in-a-haystack. LongVILA-8B demonstrates consistent accuracy improvements on long videos in the VideoMME benchmark as the number of frames increases. Besides, MM-SP is 2.1x - 5.7x faster than ring sequence parallelism and 1.1x - 1.4x faster than Megatron with context parallelism + tensor parallelism. Moreover, it seamlessly integrates with Hugging Face Transformers.
Authors: Zalan Fabian, Berk Tinaz, Mahdi Soltanolkotabi
Abstract: Diffusion models have established new state of the art in a multitude of computer vision tasks, including image restoration. Diffusion-based inverse problem solvers generate reconstructions of exceptional visual quality from heavily corrupted measurements. However, in what is widely known as the perception-distortion trade-off, the price of perceptually appealing reconstructions is often paid in declined distortion metrics, such as PSNR. Distortion metrics measure faithfulness to the observation, a crucial requirement in inverse problems. In this work, we propose a novel framework for inverse problem solving, namely we assume that the observation comes from a stochastic degradation process that gradually degrades and noises the original clean image. We learn to reverse the degradation process in order to recover the clean image. Our technique maintains consistency with the original measurement throughout the reverse process, and allows for great flexibility in trading off perceptual quality for improved distortion metrics and sampling speedup via early-stopping. We demonstrate the efficiency of our method on different high-resolution datasets and inverse problems, achieving great improvements over other state-of-the-art diffusion-based methods with respect to both perceptual and distortion metrics.
Authors: Evgeney Bogatyrev, Ivan Molodetskikh, Dmitriy Vatolin
Abstract: In recent years, there has been significant interest in Super-Resolution (SR), which focuses on generating a high-resolution image from a low-resolution input. Deep learning-based methods for super-resolution have been particularly popular and have shown impressive results on various benchmarks. However, research indicates that these methods may not perform as well on strongly compressed videos. We developed a super-resolution benchmark to analyze SR's capacity to upscale compressed videos. Our dataset employed video codecs based on five widely-used compression standards: H.264, H.265, H.266, AV1, and AVS3. We assessed 19 popular SR models using our benchmark and evaluated their ability to restore details and their susceptibility to compression artifacts. To get an accurate perceptual ranking of SR models, we conducted a crowd-sourced side-by-side comparison of their outputs. We found that some SR models, combined with compression, allow us to reduce the video bitrate without significant loss of quality. We also compared a range of image and video quality metrics with subjective scores to evaluate their accuracy on super-resolved compressed videos. The benchmark is publicly available at https://videoprocessing.ai/benchmarks/super-resolution-for-video-compression.html
URLs: https://videoprocessing.ai/benchmarks/super-resolution-for-video-compression.html
Authors: Xijie Huang, Zechun Liu, Shih-Yang Liu, Kwang-Ting Cheng
Abstract: Quantization-aware training (QAT) is a representative model compression method to reduce redundancy in weights and activations. However, most existing QAT methods require end-to-end training on the entire dataset, which suffers from long training time and high energy costs. In addition, the potential label noise in the training data undermines the robustness of QAT. We propose two metrics based on analysis of loss and gradient of quantized weights: error vector score and disagreement score, to quantify the importance of each sample during training. Guided by these two metrics, we proposed a quantization-aware Adaptive Coreset Selection (ACS) method to select the data for the current training epoch. We evaluate our method on various networks (ResNet-18, MobileNetV2, RetinaNet), datasets(CIFAR-10, CIFAR-100, ImageNet-1K, COCO), and under different quantization settings. Specifically, our method can achieve an accuracy of 68.39\% of 4-bit quantized ResNet-18 on the ImageNet-1K dataset with only a 10\% subset, which has an absolute gain of 4.24\% compared to the baseline. Our method can also improve the robustness of QAT by removing noisy samples in the training set.
Authors: Sophie Starck, Yadunandan Vivekanand Kini, Jessica Johanna Maria Ritter, Rickmer Braren, Daniel Rueckert, Tamara Mueller
Abstract: Age prediction is an important part of medical assessments and research. It can aid in detecting diseases as well as abnormal ageing by highlighting potential discrepancies between chronological and biological age. To improve understanding of age-related changes in various body parts, we investigate the ageing of the human body on a large scale by using whole-body 3D images. We utilise the Grad-CAM method to determine the body areas most predictive of a person's age. In order to expand our analysis beyond individual subjects, we employ registration techniques to generate population-wide importance maps that show the most predictive areas in the body for a whole cohort of subjects. We show that the investigation of the full 3D volume of the whole body and the population-wide analysis can give important insights into which body parts play the most important roles in predicting a person's age. Our findings reveal three primary areas of interest: the spine, the autochthonous back muscles, and the cardiac region, which exhibits the highest importance. Finally, we investigate differences between subjects that show accelerated and decelerated ageing.
Authors: Cian Eastwood, Julius von K\"ugelgen, Linus Ericsson, Diane Bouchacourt, Pascal Vincent, Bernhard Sch\"olkopf, Mark Ibrahim
Abstract: Self-supervised representation learning often uses data augmentations to induce some invariance to "style" attributes of the data. However, with downstream tasks generally unknown at training time, it is difficult to deduce a priori which attributes of the data are indeed "style" and can be safely discarded. To deal with this, current approaches try to retain some style information by tuning the degree of invariance to some particular task, such as ImageNet object classification. However, prior work has shown that such task-specific tuning can lead to significant performance degradation on other tasks that rely on the discarded style. To address this, we introduce a more principled approach that seeks to disentangle style features rather than discard them. The key idea is to add multiple style embedding spaces where: (i) each is invariant to all-but-one augmentation; and (ii) joint entropy is maximized. We formalize our structured data-augmentation procedure from a causal latent-variable-model perspective, and prove identifiability of both content and individual style variables. We empirically demonstrate the benefits of our approach on both synthetic and real-world data.
Authors: Shijun Liang, Van Hoang Minh Nguyen, Jinghan Jia, Ismail Alkhouri, Sijia Liu, Saiprasad Ravishankar
Abstract: As the popularity of deep learning (DL) in the field of magnetic resonance imaging (MRI) continues to rise, recent research has indicated that DL-based MRI reconstruction models might be excessively sensitive to minor input disturbances, including worst-case additive perturbations. This sensitivity often leads to unstable, aliased images. This raises the question of how to devise DL techniques for MRI reconstruction that can be robust to train-test variations. To address this problem, we propose a novel image reconstruction framework, termed Smoothed Unrolling (SMUG), which advances a deep unrolling-based MRI reconstruction model using a randomized smoothing (RS)-based robust learning approach. RS, which improves the tolerance of a model against input noises, has been widely used in the design of adversarial defense approaches for image classification tasks. Yet, we find that the conventional design that applies RS to the entire DL-based MRI model is ineffective. In this paper, we show that SMUG and its variants address the above issue by customizing the RS process based on the unrolling architecture of a DL-based MRI reconstruction model. Compared to the vanilla RS approach, we show that SMUG improves the robustness of MRI reconstruction with respect to a diverse set of instability sources, including worst-case and random noise perturbations to input measurements, varying measurement sampling rates, and different numbers of unrolling steps. Furthermore, we theoretically analyze the robustness of our method in the presence of perturbations.
Authors: Wenxuan Wang, Haonan Bai, Jen-tse Huang, Yuxuan Wan, Youliang Yuan, Haoyi Qiu, Nanyun Peng, Michael R. Lyu
Abstract: Image generation models can generate or edit images from a given text. Recent advancements in image generation technology, exemplified by DALL-E and Midjourney, have been groundbreaking. These advanced models, despite their impressive capabilities, are often trained on massive Internet datasets, making them susceptible to generating content that perpetuates social stereotypes and biases, which can lead to severe consequences. Prior research on assessing bias within image generation models suffers from several shortcomings, including limited accuracy, reliance on extensive human labor, and lack of comprehensive analysis. In this paper, we propose BiasPainter, a novel evaluation framework that can accurately, automatically and comprehensively trigger social bias in image generation models. BiasPainter uses a diverse range of seed images of individuals and prompts the image generation models to edit these images using gender, race, and age-neutral queries. These queries span 62 professions, 39 activities, 57 types of objects, and 70 personality traits. The framework then compares the edited images to the original seed images, focusing on the significant changes related to gender, race, and age. BiasPainter adopts a key insight that these characteristics should not be modified when subjected to neutral prompts. Built upon this design, BiasPainter can trigger the social bias and evaluate the fairness of image generation models. We use BiasPainter to evaluate six widely-used image generation models, such as stable diffusion and Midjourney. Experimental results show that BiasPainter can successfully trigger social bias in image generation models. According to our human evaluation, BiasPainter can achieve 90.8% accuracy on automatic bias detection, which is significantly higher than the results reported in previous work.
Authors: Xin Ming, Jiawei Li, Jingwang Ling, Libo Zhang, Feng Xu
Abstract: Readily editable mesh blendshapes have been widely used in animation pipelines, while recent advancements in neural geometry and appearance representations have enabled high-quality inverse rendering. Building upon these observations, we introduce a novel technique that reconstructs mesh-based blendshape rigs from single or sparse multi-view videos, leveraging state-of-the-art neural inverse rendering. We begin by constructing a deformation representation that parameterizes vertex displacements into differential coordinates with tetrahedral connections, allowing for high-quality vertex deformation on high-resolution meshes. By constructing a set of semantic regulations in this representation, we achieve joint optimization of blendshapes and expression coefficients. Furthermore, to enable a user-friendly multi-view setup with unsynchronized cameras, we propose a neural regressor to model time-varying motion parameters. This approach implicitly considers the time difference across multiple cameras, enhancing the accuracy of motion modeling. Experiments demonstrate that, with the flexible input of single or sparse multi-view videos, we reconstruct personalized high-fidelity blendshapes. These blendshapes are both geometrically and semantically accurate, and they are compatible with industrial animation pipelines. Code and data are available at https://github.com/grignarder/high-quality-blendshape-generation.
URLs: https://github.com/grignarder/high-quality-blendshape-generation.
Authors: Jiayu Huo, Xi Ouyang, S\'ebastien Ourselin, Rachel Sparks
Abstract: Rapid advancements in medical image segmentation performance have been significantly driven by the development of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). These models follow the discriminative pixel-wise classification learning paradigm and often have limited ability to generalize across diverse medical imaging datasets. In this manuscript, we introduce Generative Medical Segmentation (GMS), a novel approach leveraging a generative model to perform image segmentation. Concretely, GMS employs a robust pre-trained vision foundation model to extract latent representations for images and corresponding ground truth masks, followed by a model that learns a mapping function from the image to the mask in the latent space. Once trained, the model generates an estimated segmentation mask using the pre-trained vision foundation model to decode the predicted latent representation back into the image space. The design of GMS leads to fewer trainable parameters in the model which reduces the risk of overfitting and enhances its generalization capability. Our experimental analysis across five public datasets in different medical imaging domains demonstrates GMS outperforms existing discriminative and generative segmentation models. Furthermore, GMS is able to generalize well across datasets from different centers within the same imaging modality. Our experiments suggest GMS offers a scalable and effective solution for medical image segmentation. GMS implementation and trained model weights are available at https://github.com/King-HAW/GMS.
Authors: Anna Breger, Clemens Karner, Ian Selby, Janek Gr\"ohl, S\"oren Dittmer, Edward Lilley, Judith Babar, Jake Beckford, Thomas R Else, Timothy J Sadler, Shahab Shahipasand, Arthikkaa Thavakumar, Michael Roberts, Carola-Bibiane Sch\"onlieb
Abstract: Image quality assessment (IQA) is standard practice in the development stage of novel machine learning algorithms that operate on images. The most commonly used IQA measures have been developed and tested for natural images, but not in the medical setting. Reported inconsistencies arising in medical images are not surprising, as they have different properties than natural images. In this study, we test the applicability of common IQA measures for medical image data by comparing their assessment to manually rated chest X-ray (5 experts) and photoacoustic image data (2 experts). Moreover, we include supplementary studies on grayscale natural images and accelerated brain MRI data. The results of all experiments show a similar outcome in line with previous findings for medical imaging: PSNR and SSIM in the default setting are in the lower range of the result list and HaarPSI outperforms the other tested measures in the overall performance. Also among the top performers in our medical experiments are the full reference measures FSIM, GMSD, LPIPS and MS-SSIM. Generally, the results on natural images yield considerably higher correlations, suggesting that the additional employment of tailored IQA measures for medical imaging algorithms is needed.
Authors: Xuan-Bac Nguyen, Hoang-Quan Nguyen, Hugh Churchill, Samee U. Khan, Khoa Luu
Abstract: Although quantum machine learning has been introduced for a while, its applications in computer vision are still limited. This paper, therefore, revisits the quantum visual encoding strategies, the initial step in quantum machine learning. Investigating the root cause, we uncover that the existing quantum encoding design fails to ensure information preservation of the visual features after the encoding process, thus complicating the learning process of the quantum machine learning models. In particular, the problem, termed "Quantum Information Gap" (QIG), leads to a gap of information between classical and corresponding quantum features. We provide theoretical proof and practical demonstrations of that found and underscore the significance of QIG, as it directly impacts the performance of quantum machine learning algorithms. To tackle this challenge, we introduce a simple but efficient new loss function named Quantum Information Preserving (QIP) to minimize this gap, resulting in enhanced performance of quantum machine learning algorithms. Extensive experiments validate the effectiveness of our approach, showcasing superior performance compared to current methodologies and consistently achieving state-of-the-art results in quantum modeling.
Authors: Jiuhong Xiao, Ning Zhang, Daniel Tortei, Giuseppe Loianno
Abstract: Accurate geo-localization of Unmanned Aerial Vehicles (UAVs) is crucial for outdoor applications including search and rescue operations, power line inspections, and environmental monitoring. The vulnerability of Global Navigation Satellite Systems (GNSS) signals to interference and spoofing necessitates the development of additional robust localization methods for autonomous navigation. Visual Geo-localization (VG), leveraging onboard cameras and reference satellite maps, offers a promising solution for absolute localization. Specifically, Thermal Geo-localization (TG), which relies on image-based matching between thermal imagery with satellite databases, stands out by utilizing infrared cameras for effective nighttime localization. However, the efficiency and effectiveness of current TG approaches, are hindered by dense sampling on satellite maps and geometric noises in thermal query images. To overcome these challenges, we introduce STHN, a novel UAV thermal geo-localization approach that employs a coarse-to-fine deep homography estimation method. This method attains reliable thermal geo-localization within a 512-meter radius of the UAV's last known location even with a challenging 11\% size ratio between thermal and satellite images, despite the presence of indistinct textures and self-similar patterns. We further show how our research significantly enhances UAV thermal geo-localization performance and robustness against geometric noises under low-visibility conditions in the wild. The code is made publicly available.
Authors: Dmitry Nechaev, Alexey Pchelnikov, Ekaterina Ivanova
Abstract: Pathology, the microscopic examination of diseased tissue, is critical for diagnosing various medical conditions, particularly cancers. Traditional methods are labor-intensive and prone to human error. Digital pathology, which converts glass slides into high-resolution digital images for analysis by computer algorithms, revolutionizes the field by enhancing diagnostic accuracy, consistency, and efficiency through automated image analysis and large-scale data processing. Foundational transformer pretraining is crucial for developing robust, generalizable models as it enables learning from vast amounts of unannotated data. This paper introduces the Hibou family of foundational vision transformers for pathology, leveraging the DINOv2 framework to pretrain two model variants, Hibou-B and Hibou-L, on a proprietary dataset of over 1 million whole slide images (WSIs) representing diverse tissue types and staining techniques. Our pretrained models demonstrate superior performance on both patch-level and slide-level benchmarks, surpassing existing state-of-the-art methods. Notably, Hibou-L achieves the highest average accuracy across multiple benchmark datasets. To support further research and application in the field, we have open-sourced the Hibou models, which can be accessed at https://github.com/HistAI/hibou.
Authors: Seon-Hoon Kim, Dae-won Chung
Abstract: Synthetic Aperture Radar (SAR) imaging technology provides the unique advantage of being able to collect data regardless of weather conditions and time. However, SAR images exhibit complex backscatter patterns and speckle noise, which necessitate expertise for interpretation. Research on translating SAR images into optical-like representations has been conducted to aid the interpretation of SAR data. Nevertheless, existing studies have predominantly utilized low-resolution satellite imagery datasets and have largely been based on Generative Adversarial Network (GAN) which are known for their training instability and low fidelity. To overcome these limitations of low-resolution data usage and GAN-based approaches, this paper introduces a conditional image-to-image translation approach based on Brownian Bridge Diffusion Model (BBDM). We conducted comprehensive experiments on the MSAW dataset, a paired SAR and optical images collection of 0.5m Very-High-Resolution (VHR). The experimental results indicate that our method surpasses both the Conditional Diffusion Models (CDMs) and the GAN-based models in diverse perceptual quality metrics.
Authors: Cho Yang
Abstract: CycleGAN was trained on SynthRAD Grand Challenge Dataset using the single-epoch modification (SEM) method proposed in this paper which is referred to as (CycleGAN-single) compared to the usual method of training CycleGAN on around 200 epochs (CycleGAN-multi). Model performance were evaluated qualitatively and quantitatively with quantitative performance metrics like PSNR, SSIM, MAE and MSE. The consideration of both quantitative and qualitative performance when evaluating a model is unique to certain image-translation tasks like medical imaging as detailed in this paper. Also, this paper shows that good quantitative performance does not always imply good qualitative performance and the converse is also not always True (i.e. good qualitative performance does not always imply good quantitative performance). This paper also proposes FQGA (Fast Paired Image-to-Image Translation Quarter-Generator Adversary) Model which has 1/4 the number of parameters compared to CycleGAN (when comparing their Generator Models). FQGA outperforms CycleGAN qualitatively and quantitatively even only after training on 20 epochs. Finally, using SEM method on FQGA allowed it to again outperform CycleGAN both quantitatively and qualitatively. These performance gains with fewer model parameters and time savings from running fewer epochs may also be applicable to other image-to-image translation tasks in Machine Learning apart from the Medical image-translation task discussed in this paper between Cone Beam Computed Tomography (CBCT) and Computed Tomography (CT) images.
Authors: Yi Kan Wang, Ludmila Tydlitatova, Jeremy D. Kunz, Gerard Oakley, Ran A. Godrich, Matthew C. H. Lee, Chad Vanderbilt, Razik Yousfi, Thomas Fuchs, David S. Klimstra, Siqi Liu
Abstract: Many molecular alterations serve as clinically prognostic or therapy-predictive biomarkers, typically detected using single or multi-gene molecular assays. However, these assays are expensive, tissue destructive and often take weeks to complete. Using AI on routine H&E WSIs offers a fast and economical approach to screen for multiple molecular biomarkers. We present a high-throughput AI-based system leveraging Virchow2, a foundation model pre-trained on 3 million slides, to interrogate genomic features previously determined by an next-generation sequencing (NGS) assay, using 47,960 scanned hematoxylin and eosin (H&E) whole slide images (WSIs) from 38,984 cancer patients. Unlike traditional methods that train individual models for each biomarker or cancer type, our system employs a unified model to simultaneously predict a wide range of clinically relevant molecular biomarkers across cancer types. By training the network to replicate the MSK-IMPACT targeted biomarker panel of 505 genes, it identified 80 high performing biomarkers with a mean AU-ROC of 0.89 in 15 most common cancer types. In addition, 40 biomarkers demonstrated strong associations with specific cancer histologic subtypes. Furthermore, 58 biomarkers were associated with targets frequently assayed clinically for therapy selection and response prediction. The model can also predict the activity of five canonical signaling pathways, identify defects in DNA repair mechanisms, and predict genomic instability measured by tumor mutation burden, microsatellite instability (MSI), and chromosomal instability (CIN). The proposed model can offer potential to guide therapy selection, improve treatment efficacy, accelerate patient screening for clinical trials and provoke the interrogation of new therapeutic targets.
Authors: Zhi Qiao, Linbin Han, Xiantong Zhen, Jia-Hong Gao, Zhen Qian
Abstract: In light of the inherent entailment relations between images and text, hyperbolic point vector embeddings, leveraging the hierarchical modeling advantages of hyperbolic space, have been utilized for visual semantic representation learning. However, point vector embedding approaches fail to address the issue of semantic uncertainty, where an image may have multiple interpretations, and text may refer to different images, a phenomenon particularly prevalent in the medical domain. Therefor, we propose \textbf{HYDEN}, a novel hyperbolic density embedding based image-text representation learning approach tailored for specific medical domain data. This method integrates text-aware local features alongside global features from images, mapping image-text features to density features in hyperbolic space via using hyperbolic pseudo-Gaussian distributions. An encapsulation loss function is employed to model the partial order relations between image-text density distributions. Experimental results demonstrate the interpretability of our approach and its superior performance compared to the baseline methods across various zero-shot tasks and different datasets.