Authors: Chang Sun, Hong Yang, Bo Qin
Abstract: Visual Speech Recognition (VSR) tasks are generally recognized to have a lower theoretical performance ceiling than Automatic Speech Recognition (ASR), owing to the inherent limitations of conveying semantic information visually. To mitigate this challenge, this paper introduces an advanced knowledge distillation approach using a Joint-Embedding Predictive Architecture (JEPA), named JEP-KD, designed to more effectively utilize audio features during model training. Central to JEP-KD is the inclusion of a generative network within the embedding layer, which enhances the video encoder's capacity for semantic feature extraction and brings it into closer alignment with the audio features from a pre-trained ASR model's encoder. This approach aims to progressively reduce the performance gap between VSR and ASR. Moreover, a comprehensive multimodal, multistage training regimen for the JEP-KD framework is established, bolstering the robustness and efficacy of the training process. Experiment results demonstrate that JEP-KD significantly improves the performance of VSR models and demonstrates versatility across different VSR platforms, indicating its potential for broader application within other multimodal tasks.
Authors: Md. Simul Hasan Talukder, Sharmin Akter, Abdullah Hafez Nur
Abstract: Sugarcane, a key crop for the world's sugar industry, is prone to several diseases that have a substantial negative influence on both its yield and quality. To effectively manage and implement preventative initiatives, diseases must be detected promptly and accurately. In this study, we present a unique model called sugarcaneNet2024 that outperforms previous methods for automatically and quickly detecting sugarcane disease through leaf image processing. Our proposed model consolidates an optimized weighted average ensemble of seven customized and LASSO-regularized pre-trained models, particularly InceptionV3, InceptionResNetV2, DenseNet201, DenseNet169, Xception, and ResNet152V2. Initially, we added three more dense layers with 0.0001 LASSO regularization, three 30% dropout layers, and three batch normalizations with renorm enabled at the bottom of these pre-trained models to improve the performance. The accuracy of sugarcane leaf disease classification was greatly increased by this addition. Following this, several comparative studies between the average ensemble and individual models were carried out, indicating that the ensemble technique performed better. The average ensemble of all modified pre-trained models produced outstanding outcomes: 100%, 99%, 99%, and 99.45% for f1 score, precision, recall, and accuracy, respectively. Performance was further enhanced by the implementation of an optimized weighted average ensemble technique incorporated with grid search. This optimized sugarcaneNet2024 model performed the best for detecting sugarcane diseases, having achieved accuracy, precision, recall, and F1 score of 99.67%, 100%, 100%, and 100% , respectively.
Authors: Han Yuan, Chuan Hong, Pengtao Jiang, Gangming Zhao, Nguyen Tuan Anh Tran, Xinxing Xu, Yet Yen Yan, Nan Liu
Abstract: Background: Pneumothorax is an acute thoracic disease caused by abnormal air collection between the lungs and chest wall. To address the opaqueness often associated with deep learning (DL) models, explainable artificial intelligence (XAI) methods have been introduced to outline regions related to pneumothorax diagnoses made by DL models. However, these explanations sometimes diverge from actual lesion areas, highlighting the need for further improvement. Method: We propose a template-guided approach to incorporate the clinical knowledge of pneumothorax into model explanations generated by XAI methods, thereby enhancing the quality of these explanations. Utilizing one lesion delineation created by radiologists, our approach first generates a template that represents potential areas of pneumothorax occurrence. This template is then superimposed on model explanations to filter out extraneous explanations that fall outside the template's boundaries. To validate its efficacy, we carried out a comparative analysis of three XAI methods with and without our template guidance when explaining two DL models in two real-world datasets. Results: The proposed approach consistently improved baseline XAI methods across twelve benchmark scenarios built on three XAI methods, two DL models, and two datasets. The average incremental percentages, calculated by the performance improvements over the baseline performance, were 97.8% in Intersection over Union (IoU) and 94.1% in Dice Similarity Coefficient (DSC) when comparing model explanations and ground-truth lesion areas. Conclusions: In the context of pneumothorax diagnoses, we proposed a template-guided approach for improving AI explanations. We anticipate that our template guidance will forge a fresh approach to elucidating AI models by integrating clinical domain expertise.
Authors: Young Seok Jeon, Hongfei Yang, Huazhu Fu, Mengling Feng
Abstract: Imposing key anatomical features, such as the number of organs, their shapes, sizes, and relative positions, is crucial for building a robust multi-organ segmentation model. Current attempts to incorporate anatomical features include broadening effective receptive fields (ERF) size with resource- and data-intensive modules such as self-attention or introducing organ-specific topology regularizers, which may not scale to multi-organ segmentation problems where inter-organ relation also plays a huge role. We introduce a new approach to impose anatomical constraints on any existing encoder-decoder segmentation model by conditioning model prediction with learnable anatomy prior. More specifically, given an abdominal scan, a part of the encoder spatially warps a learnable prior to align with the given input scan using thin plate spline (TPS) grid interpolation. The warped prior is then integrated during the decoding phase to guide the model for more anatomy-informed predictions. Code is available at \hyperlink{https://anonymous.4open.science/r/AIC-UNet-7048}{https://anonymous.4open.science/r/AIC-UNet-7048}.
URLs: https://anonymous.4open.science/r/AIC-UNet-7048, https://anonymous.4open.science/r/AIC-UNet-7048
Authors: Yasuyuki Ihara
Abstract: Multiple object tracking (MOT), a key task in image recognition, presents a persistent challenge in balancing processing speed and tracking accuracy. This study introduces a novel approach that leverages quantum annealing (QA) to expedite computation speed, while enhancing tracking accuracy through the ensembling of object tracking processes. A method to improve the matching integration process is also proposed. By utilizing the sequential nature of MOT, this study further augments the tracking method via reverse annealing (RA). Experimental validation confirms the maintenance of high accuracy with an annealing time of a mere 3 $\mu$s per tracking process. The proposed method holds significant potential for real-time MOT applications, including traffic flow measurement for urban traffic light control, collision prediction for autonomous robots and vehicles, and management of products mass-produced in factories.
Authors: Luigi Piccinelli, Yung-Hsu Yang, Christos Sakaridis, Mattia Segu, Siyuan Li, Luc Van Gool, Fisher Yu
Abstract: Accurate monocular metric depth estimation (MMDE) is crucial to solving downstream tasks in 3D perception and modeling. However, the remarkable accuracy of recent MMDE methods is confined to their training domains. These methods fail to generalize to unseen domains even in the presence of moderate domain gaps, which hinders their practical applicability. We propose a new model, UniDepth, capable of reconstructing metric 3D scenes from solely single images across domains. Departing from the existing MMDE methods, UniDepth directly predicts metric 3D points from the input image at inference time without any additional information, striving for a universal and flexible MMDE solution. In particular, UniDepth implements a self-promptable camera module predicting dense camera representation to condition depth features. Our model exploits a pseudo-spherical output representation, which disentangles camera and depth representations. In addition, we propose a geometric invariance loss that promotes the invariance of camera-prompted depth features. Thorough evaluations on ten datasets in a zero-shot regime consistently demonstrate the superior performance of UniDepth, even when compared with methods directly trained on the testing domains. Code and models are available at: https://github.com/lpiccinelli-eth/unidepth
Authors: Edward Fish, Jon Weinbren, Andrew Gilbert
Abstract: This paper introduces a novel approach to temporal action localization (TAL) in few-shot learning. Our work addresses the inherent limitations of conventional single-prompt learning methods that often lead to overfitting due to the inability to generalize across varying contexts in real-world videos. Recognizing the diversity of camera views, backgrounds, and objects in videos, we propose a multi-prompt learning framework enhanced with optimal transport. This design allows the model to learn a set of diverse prompts for each action, capturing general characteristics more effectively and distributing the representation to mitigate the risk of overfitting. Furthermore, by employing optimal transport theory, we efficiently align these prompts with action features, optimizing for a comprehensive representation that adapts to the multifaceted nature of video data. Our experiments demonstrate significant improvements in action localization accuracy and robustness in few-shot settings on the standard challenging datasets of THUMOS-14 and EpicKitchens100, highlighting the efficacy of our multi-prompt optimal transport approach in overcoming the challenges of conventional few-shot TAL methods.
Authors: Mukund Varma T, Peihao Wang, Zhiwen Fan, Zhangyang Wang, Hao Su, Ravi Ramamoorthi
Abstract: In recent years, there has been an explosion of 2D vision models for numerous tasks such as semantic segmentation, style transfer or scene editing, enabled by large-scale 2D image datasets. At the same time, there has been renewed interest in 3D scene representations such as neural radiance fields from multi-view images. However, the availability of 3D or multiview data is still substantially limited compared to 2D image datasets, making extending 2D vision models to 3D data highly desirable but also very challenging. Indeed, extending a single 2D vision operator like scene editing to 3D typically requires a highly creative method specialized to that task and often requires per-scene optimization. In this paper, we ask the question of whether any 2D vision model can be lifted to make 3D consistent predictions. We answer this question in the affirmative; our new Lift3D method trains to predict unseen views on feature spaces generated by a few visual models (i.e. DINO and CLIP), but then generalizes to novel vision operators and tasks, such as style transfer, super-resolution, open vocabulary segmentation and image colorization; for some of these tasks, there is no comparable previous 3D method. In many cases, we even outperform state-of-the-art methods specialized for the task in question. Moreover, Lift3D is a zero-shot method, in the sense that it requires no task-specific training, nor scene-specific optimization.
Authors: Yanyu Li, Xian Liu, Anil Kag, Ju Hu, Yerlan Idelbayev, Dhritiman Sagar, Yanzhi Wang, Sergey Tulyakov, Jian Ren
Abstract: Diffusion-based text-to-image generative models, e.g., Stable Diffusion, have revolutionized the field of content generation, enabling significant advancements in areas like image editing and video synthesis. Despite their formidable capabilities, these models are not without their limitations. It is still challenging to synthesize an image that aligns well with the input text, and multiple runs with carefully crafted prompts are required to achieve satisfactory results. To mitigate these limitations, numerous studies have endeavored to fine-tune the pre-trained diffusion models, i.e., UNet, utilizing various technologies. Yet, amidst these efforts, a pivotal question of text-to-image diffusion model training has remained largely unexplored: Is it possible and feasible to fine-tune the text encoder to improve the performance of text-to-image diffusion models? Our findings reveal that, instead of replacing the CLIP text encoder used in Stable Diffusion with other large language models, we can enhance it through our proposed fine-tuning approach, TextCraftor, leading to substantial improvements in quantitative benchmarks and human assessments. Interestingly, our technique also empowers controllable image generation through the interpolation of different text encoders fine-tuned with various rewards. We also demonstrate that TextCraftor is orthogonal to UNet finetuning, and can be combined to further improve generative quality.
Authors: Anees Ur Rehman Hashmi, Dwarikanath Mahapatra, Mohammad Yaqub
Abstract: Explaining Deep Learning models is becoming increasingly important in the face of daily emerging multimodal models, particularly in safety-critical domains like medical imaging. However, the lack of detailed investigations into the performance of explainability methods on these models is widening the gap between their development and safe deployment. In this work, we analyze the performance of various explainable AI methods on a vision-language model, MedCLIP, to demystify its inner workings. We also provide a simple methodology to overcome the shortcomings of these methods. Our work offers a different new perspective on the explainability of a recent well-known VLM in the medical domain and our assessment method is generalizable to other current and possible future VLMs.
Authors: Yui Lo, Yuqian Chen, Dongnan Liu, Wan Liu, Leo Zekelman, Fan Zhang, Yogesh Rathi, Nikos Makris, Alexandra J. Golby, Weidong Cai, Lauren J. O'Donnell
Abstract: Shape plays an important role in computer graphics, offering informative features to convey an object's morphology and functionality. Shape analysis in brain imaging can help interpret structural and functionality correlations of the human brain. In this work, we investigate the shape of the brain's 3D white matter connections and its potential predictive relationship to human cognitive function. We reconstruct brain connections as sequences of 3D points using diffusion magnetic resonance imaging (dMRI) tractography. To describe each connection, we extract 12 shape descriptors in addition to traditional dMRI connectivity and tissue microstructure features. We introduce a novel framework, Shape--fused Fiber Cluster Transformer (SFFormer), that leverages a multi-head cross-attention feature fusion module to predict subject-specific language performance based on dMRI tractography. We assess the performance of the method on a large dataset including 1065 healthy young adults. The results demonstrate that both the transformer-based SFFormer model and its inter/intra feature fusion with shape, microstructure, and connectivity are informative, and together, they improve the prediction of subject-specific language performance scores. Overall, our results indicate that the shape of the brain's connections is predictive of human language function.
Authors: Khiem Vuong, N. Dinesh Reddy, Robert Tamburo, Srinivasa G. Narasimhan
Abstract: Current methods for 2D and 3D object understanding struggle with severe occlusions in busy urban environments, partly due to the lack of large-scale labeled ground-truth annotations for learning occlusion. In this work, we introduce a novel framework for automatically generating a large, realistic dataset of dynamic objects under occlusions using freely available time-lapse imagery. By leveraging off-the-shelf 2D (bounding box, segmentation, keypoint) and 3D (pose, shape) predictions as pseudo-groundtruth, unoccluded 3D objects are identified automatically and composited into the background in a clip-art style, ensuring realistic appearances and physically accurate occlusion configurations. The resulting clip-art image with pseudo-groundtruth enables efficient training of object reconstruction methods that are robust to occlusions. Our method demonstrates significant improvements in both 2D and 3D reconstruction, particularly in scenarios with heavily occluded objects like vehicles and people in urban scenes.
Authors: Weizhuo Wang, C. Karen Liu, Monroe Kennedy III
Abstract: Wearable collaborative robots stand to assist human wearers who need fall prevention assistance or wear exoskeletons. Such a robot needs to be able to predict the ego motion of the wearer based on egocentric vision and the surrounding scene. In this work, we leveraged body-mounted cameras and sensors to anticipate the trajectory of human wearers through complex surroundings. To facilitate research in ego-motion prediction, we have collected a comprehensive walking scene navigation dataset centered on the user's perspective. We present a method to predict human motion conditioning on the surrounding static scene. Our method leverages a diffusion model to produce a distribution of potential future trajectories, taking into account the user's observation of the environment. We introduce a compact representation to encode the user's visual memory of the surroundings, as well as an efficient sample-generating technique to speed up real-time inference of a diffusion model. We ablate our model and compare it to baselines, and results show that our model outperforms existing methods on key metrics of collision avoidance and trajectory mode coverage.
Authors: Jorgen Cani, Ioannis Mademlis, Adamantia Anna Rebolledo Chrysochoou, Georgios Th. Papadopoulos
Abstract: Illicit object detection is a critical task performed at various high-security locations, including airports, train stations, subways, and ports. The continuous and tedious work of examining thousands of X-ray images per hour can be mentally taxing. Thus, Deep Neural Networks (DNNs) can be used to automate the X-ray image analysis process, improve efficiency and alleviate the security officers' inspection burden. The neural architectures typically utilized in relevant literature are Convolutional Neural Networks (CNNs), with Vision Transformers (ViTs) rarely employed. In order to address this gap, this paper conducts a comprehensive evaluation of relevant ViT architectures on illicit item detection in X-ray images. This study utilizes both Transformer and hybrid backbones, such as SWIN and NextViT, and detectors, such as DINO and RT-DETR. The results demonstrate the remarkable accuracy of the DINO Transformer detector in the low-data regime, the impressive real-time performance of YOLOv8, and the effectiveness of the hybrid NextViT backbone.
Authors: De-An Huang, Shijia Liao, Subhashree Radhakrishnan, Hongxu Yin, Pavlo Molchanov, Zhiding Yu, Jan Kautz
Abstract: There has been tremendous progress in multimodal Large Language Models (LLMs). Recent works have extended these models to video input with promising instruction following capabilities. However, an important missing piece is temporal localization. These models cannot accurately answer the "When?" questions. We identify three key aspects that limit their temporal localization capabilities: (i) time representation, (ii) architecture, and (iii) data. We address these shortcomings by proposing Language Instructed Temporal-Localization Assistant (LITA) with the following features: (1) We introduce time tokens that encode timestamps relative to the video length to better represent time in videos. (2) We introduce SlowFast tokens in the architecture to capture temporal information at fine temporal resolution. (3) We emphasize temporal localization data for LITA. In addition to leveraging existing video datasets with timestamps, we propose a new task, Reasoning Temporal Localization (RTL), along with the dataset, ActivityNet-RTL, for learning and evaluating this task. Reasoning temporal localization requires both the reasoning and temporal localization of Video LLMs. LITA demonstrates strong performance on this challenging task, nearly doubling the temporal mean intersection-over-union (mIoU) of baselines. In addition, we show that our emphasis on temporal localization also substantially improves video-based text generation compared to existing Video LLMs, including a 36% relative improvement of Temporal Understanding. Code is available at: https://github.com/NVlabs/LITA
Authors: Vishal Purohit, Junjie Luo, Yiheng Chi, Qi Guo, Stanley H. Chan, Qiang Qiu
Abstract: The astonishing development of single-photon cameras has created an unprecedented opportunity for scientific and industrial imaging. However, the high data throughput generated by these 1-bit sensors creates a significant bottleneck for low-power applications. In this paper, we explore the possibility of generating a color image from a single binary frame of a single-photon camera. We evidently find this problem being particularly difficult to standard colorization approaches due to the substantial degree of exposure variation. The core innovation of our paper is an exposure synthesis model framed under a neural ordinary differential equation (Neural ODE) that allows us to generate a continuum of exposures from a single observation. This innovation ensures consistent exposure in binary images that colorizers take on, resulting in notably enhanced colorization. We demonstrate applications of the method in single-image and burst colorization and show superior generative performance over baselines. Project website can be found at https://vishal-s-p.github.io/projects/2023/generative_quanta_color.html.
URLs: https://vishal-s-p.github.io/projects/2023/generative_quanta_color.html.
Authors: Wei Dong, Xing Zhang, Bihui Chen, Dawei Yan, Zhijun Lin, Qingsen Yan, Peng Wang, Yang Yang
Abstract: Parameter-efficient fine-tuning for pre-trained Vision Transformers aims to adeptly tailor a model to downstream tasks by learning a minimal set of new adaptation parameters while preserving the frozen majority of pre-trained parameters. Striking a balance between retaining the generalizable representation capacity of the pre-trained model and acquiring task-specific features poses a key challenge. Currently, there is a lack of focus on guiding this delicate trade-off. In this study, we approach the problem from the perspective of Singular Value Decomposition (SVD) of pre-trained parameter matrices, providing insights into the tuning dynamics of existing methods. Building upon this understanding, we propose a Residual-based Low-Rank Rescaling (RLRR) fine-tuning strategy. This strategy not only enhances flexibility in parameter tuning but also ensures that new parameters do not deviate excessively from the pre-trained model through a residual design. Extensive experiments demonstrate that our method achieves competitive performance across various downstream image classification tasks, all while maintaining comparable new parameters. We believe this work takes a step forward in offering a unified perspective for interpreting existing methods and serves as motivation for the development of new approaches that move closer to effectively considering the crucial trade-off mentioned above. Our code is available at \href{https://github.com/zstarN70/RLRR.git}{https://github.com/zstarN70/RLRR.git}.
URLs: https://github.com/zstarN70/RLRR.git, https://github.com/zstarN70/RLRR.git
Authors: Liangjian Wen, Xiasi Wang, Jianzhuang Liu, Zenglin Xu
Abstract: Self-supervised learning aims to learn representation that can be effectively generalized to downstream tasks. Many self-supervised approaches regard two views of an image as both the input and the self-supervised signals, assuming that either view contains the same task-relevant information and the shared information is (approximately) sufficient for predicting downstream tasks. Recent studies show that discarding superfluous information not shared between the views can improve generalization. Hence, the ideal representation is sufficient for downstream tasks and contains minimal superfluous information, termed minimal sufficient representation. One can learn this representation by maximizing the mutual information between the representation and the supervised view while eliminating superfluous information. Nevertheless, the computation of mutual information is notoriously intractable. In this work, we propose an objective termed multi-view entropy bottleneck (MVEB) to learn minimal sufficient representation effectively. MVEB simplifies the minimal sufficient learning to maximizing both the agreement between the embeddings of two views and the differential entropy of the embedding distribution. Our experiments confirm that MVEB significantly improves performance. For example, it achieves top-1 accuracy of 76.9\% on ImageNet with a vanilla ResNet-50 backbone on linear evaluation. To the best of our knowledge, this is the new state-of-the-art result with ResNet-50.
Authors: Junjie Wen, Jinqiang Cui, Benyun Zhao, Bingxin Han, Xuchen Liu, Zhi Gao, Ben M. Chen
Abstract: In recent years, significant progress has been made in the field of underwater image enhancement (UIE). However, its practical utility for high-level vision tasks, such as underwater object detection (UOD) in Autonomous Underwater Vehicles (AUVs), remains relatively unexplored. It may be attributed to several factors: (1) Existing methods typically employ UIE as a pre-processing step, which inevitably introduces considerable computational overhead and latency. (2) The process of enhancing images prior to training object detectors may not necessarily yield performance improvements. (3) The complex underwater environments can induce significant domain shifts across different scenarios, seriously deteriorating the UOD performance. To address these challenges, we introduce EnYOLO, an integrated real-time framework designed for simultaneous UIE and UOD with domain-adaptation capability. Specifically, both the UIE and UOD task heads share the same network backbone and utilize a lightweight design. Furthermore, to ensure balanced training for both tasks, we present a multi-stage training strategy aimed at consistently enhancing their performance. Additionally, we propose a novel domain-adaptation strategy to align feature embeddings originating from diverse underwater environments. Comprehensive experiments demonstrate that our framework not only achieves state-of-the-art (SOTA) performance in both UIE and UOD tasks, but also shows superior adaptability when applied to different underwater scenarios. Our efficiency analysis further highlights the substantial potential of our framework for onboard deployment.
Authors: Yanting Wang, Hongye Fu, Wei Zou, Jinyuan Jia
Abstract: Different from a unimodal model whose input is from a single modality, the input (called multi-modal input) of a multi-modal model is from multiple modalities such as image, 3D points, audio, text, etc. Similar to unimodal models, many existing studies show that a multi-modal model is also vulnerable to adversarial perturbation, where an attacker could add small perturbation to all modalities of a multi-modal input such that the multi-modal model makes incorrect predictions for it. Existing certified defenses are mostly designed for unimodal models, which achieve sub-optimal certified robustness guarantees when extended to multi-modal models as shown in our experimental results. In our work, we propose MMCert, the first certified defense against adversarial attacks to a multi-modal model. We derive a lower bound on the performance of our MMCert under arbitrary adversarial attacks with bounded perturbations to both modalities (e.g., in the context of auto-driving, we bound the number of changed pixels in both RGB image and depth image). We evaluate our MMCert using two benchmark datasets: one for the multi-modal road segmentation task and the other for the multi-modal emotion recognition task. Moreover, we compare our MMCert with a state-of-the-art certified defense extended from unimodal models. Our experimental results show that our MMCert outperforms the baseline.
Authors: Yunpeng Zhang, Deheng Qian, Ding Li, Yifeng Pan, Yong Chen, Zhenbao Liang, Zhiyao Zhang, Shurui Zhang, Hongxu Li, Maolei Fu, Yun Ye, Zhujin Liang, Yi Shan, Dalong Du
Abstract: Modeling complicated interactions among the ego-vehicle, road agents, and map elements has been a crucial part for safety-critical autonomous driving. Previous works on end-to-end autonomous driving rely on the attention mechanism for handling heterogeneous interactions, which fails to capture the geometric priors and is also computationally intensive. In this paper, we propose the Interaction Scene Graph (ISG) as a unified method to model the interactions among the ego-vehicle, road agents, and map elements. With the representation of the ISG, the driving agents aggregate essential information from the most influential elements, including the road agents with potential collisions and the map elements to follow. Since a mass of unnecessary interactions are omitted, the more efficient scene-graph-based framework is able to focus on indispensable connections and leads to better performance. We evaluate the proposed method for end-to-end autonomous driving on the nuScenes dataset. Compared with strong baselines, our method significantly outperforms in the full-stack driving tasks, including perception, prediction, and planning. Code will be released at https://github.com/zhangyp15/GraphAD.
Authors: Benhao Huang
Abstract: The emergence of text-to-image models marks a significant milestone in the evolution of AI-generated images (AGIs), expanding their use in diverse domains like design, entertainment, and more. Despite these breakthroughs, the quality of AGIs often remains suboptimal, highlighting the need for effective evaluation methods. These methods are crucial for assessing the quality of images relative to their textual descriptions, and they must accurately mirror human perception. Substantial progress has been achieved in this domain, with innovative techniques such as BLIP and DBCNN contributing significantly. However, recent studies, including AGIQA-3K, reveal a notable discrepancy between current methods and state-of-the-art (SOTA) standards. This gap emphasizes the necessity for a more sophisticated and precise evaluation metric. In response, our objective is to develop a model that could give ratings for metrics, which focuses on parameters like perceptual quality, authenticity, and the correspondence between text and image, that more closely aligns with human perception. In our paper, we introduce a range of effective methods, including prompt designs and the Metric Transformer. The Metric Transformer is a novel structure inspired by the complex interrelationships among various AGI quality metrics. The code is available at https://github.com/huskydoge/CS3324-Digital-Image-Processing/tree/main/Assignment1
URLs: https://github.com/huskydoge/CS3324-Digital-Image-Processing/tree/main/Assignment1
Authors: Yutong He, Alexander Robey, Naoki Murata, Yiding Jiang, Joshua Williams, George J. Pappas, Hamed Hassani, Yuki Mitsufuji, Ruslan Salakhutdinov, J. Zico Kolter
Abstract: Prompt engineering is effective for controlling the output of text-to-image (T2I) generative models, but it is also laborious due to the need for manually crafted prompts. This challenge has spurred the development of algorithms for automated prompt generation. However, these methods often struggle with transferability across T2I models, require white-box access to the underlying model, and produce non-intuitive prompts. In this work, we introduce PRISM, an algorithm that automatically identifies human-interpretable and transferable prompts that can effectively generate desired concepts given only black-box access to T2I models. Inspired by large language model (LLM) jailbreaking, PRISM leverages the in-context learning ability of LLMs to iteratively refine the candidate prompts distribution for given reference images. Our experiments demonstrate the versatility and effectiveness of PRISM in generating accurate prompts for objects, styles and images across multiple T2I models, including Stable Diffusion, DALL-E, and Midjourney.
Authors: Lingjun Zhao, Jingyu Song, Katherine A. Skinner
Abstract: In the field of 3D object detection for autonomous driving, LiDAR-Camera (LC) fusion is the top-performing sensor configuration. Still, LiDAR is relatively high cost, which hinders adoption of this technology for consumer automobiles. Alternatively, camera and radar are commonly deployed on vehicles already on the road today, but performance of Camera-Radar (CR) fusion falls behind LC fusion. In this work, we propose Camera-Radar Knowledge Distillation (CRKD) to bridge the performance gap between LC and CR detectors with a novel cross-modality KD framework. We use the Bird's-Eye-View (BEV) representation as the shared feature space to enable effective knowledge distillation. To accommodate the unique cross-modality KD path, we propose four distillation losses to help the student learn crucial features from the teacher model. We present extensive evaluations on the nuScenes dataset to demonstrate the effectiveness of the proposed CRKD framework. The project page for CRKD is https://song-jingyu.github.io/CRKD.
Authors: John R. McNulty, Lee Kho, Alexandria L. Case, Charlie Fornaca, Drew Johnston, David Slater, Joshua M. Abzug, Sybil A. Russell
Abstract: In medical imaging, access to data is commonly limited due to patient privacy restrictions and the issue that it can be difficult to acquire enough data in the case of rare diseases.[1] The purpose of this investigation was to develop a reusable open-source synthetic image generation pipeline, the GAN Image Synthesis Tool (GIST), that is easy to use as well as easy to deploy. The pipeline helps to improve and standardize AI algorithms in the digital health space by generating high quality synthetic image data that is not linked to specific patients. Its image generation capabilities include the ability to generate imaging of pathologies or injuries with low incidence rates. This improvement of digital health AI algorithms could improve diagnostic accuracy, aid in patient care, decrease medicolegal claims, and ultimately decrease the overall cost of healthcare. The pipeline builds on existing Generative Adversarial Networks (GANs) algorithms, and preprocessing and evaluation steps were included for completeness. For this work, we focused on ensuring the pipeline supports radiography, with a focus on synthetic knee and elbow x-ray images. In designing the pipeline, we evaluated the performance of current GAN architectures, studying the performance on available x-ray data. We show that the pipeline is capable of generating high quality and clinically relevant images based on a lay person's evaluation and the Fr\'echet Inception Distance (FID) metric.
Authors: Hao Shen, Lu Shi, Wanru Xu, Yigang Cen, Linna Zhang, Gaoyun An
Abstract: Video Anomaly Detection (VAD), aiming to identify abnormalities within a specific context and timeframe, is crucial for intelligent Video Surveillance Systems. While recent deep learning-based VAD models have shown promising results by generating high-resolution frames, they often lack competence in preserving detailed spatial and temporal coherence in video frames. To tackle this issue, we propose a self-supervised learning approach for VAD through an inter-patch relationship prediction task. Specifically, we introduce a two-branch vision transformer network designed to capture deep visual features of video frames, addressing spatial and temporal dimensions responsible for modeling appearance and motion patterns, respectively. The inter-patch relationship in each dimension is decoupled into inter-patch similarity and the order information of each patch. To mitigate memory consumption, we convert the order information prediction task into a multi-label learning problem, and the inter-patch similarity prediction task into a distance matrix regression problem. Comprehensive experiments demonstrate the effectiveness of our method, surpassing pixel-generation-based methods by a significant margin across three public benchmarks. Additionally, our approach outperforms other self-supervised learning-based methods.
Authors: Jinhong Wang, Tingting Chen, Jintai Chen, Yixuan Wu, Yuyang Xu, Danny Chen, Haochao Ying, Jian Wu
Abstract: Automatic ophthalmic disease diagnosis on fundus images is important in clinical practice. However, due to complex fundus textures and limited annotated data, developing an effective automatic method for this problem is still challenging. In this paper, we present a self-supervised method via polar transformation based progressive contrastive learning, called PoCo, for ophthalmic disease diagnosis. Specifically, we novelly inject the polar transformation into contrastive learning to 1) promote contrastive learning pre-training to be faster and more stable and 2) naturally capture task-free and rotation-related textures, which provides insights into disease recognition on fundus images. Beneficially, simple normal translation-invariant convolution on transformed images can equivalently replace the complex rotation-invariant and sector convolution on raw images. After that, we develop a progressive contrastive learning method to efficiently utilize large unannotated images and a novel progressive hard negative sampling scheme to gradually reduce the negative sample number for efficient training and performance enhancement. Extensive experiments on three public ophthalmic disease datasets show that our PoCo achieves state-of-the-art performance with good generalization ability, validating that our method can reduce annotation efforts and provide reliable diagnosis. Codes are available at \url{https://github.com/wjh892521292/PoCo}.
Authors: Jianqiang Wan, Sibo Song, Wenwen Yu, Yuliang Liu, Wenqing Cheng, Fei Huang, Xiang Bai, Cong Yao, Zhibo Yang
Abstract: Recently, visually-situated text parsing (VsTP) has experienced notable advancements, driven by the increasing demand for automated document understanding and the emergence of Generative Large Language Models (LLMs) capable of processing document-based questions. Various methods have been proposed to address the challenging problem of VsTP. However, due to the diversified targets and heterogeneous schemas, previous works usually design task-specific architectures and objectives for individual tasks, which inadvertently leads to modal isolation and complex workflow. In this paper, we propose a unified paradigm for parsing visually-situated text across diverse scenarios. Specifically, we devise a universal model, called OmniParser, which can simultaneously handle three typical visually-situated text parsing tasks: text spotting, key information extraction, and table recognition. In OmniParser, all tasks share the unified encoder-decoder architecture, the unified objective: point-conditioned text generation, and the unified input & output representation: prompt & structured sequences. Extensive experiments demonstrate that the proposed OmniParser achieves state-of-the-art (SOTA) or highly competitive performances on 7 datasets for the three visually-situated text parsing tasks, despite its unified, concise design. The code is available at https://github.com/AlibabaResearch/AdvancedLiterateMachinery.
URLs: https://github.com/AlibabaResearch/AdvancedLiterateMachinery.
Authors: Saurav Jha, Dong Gong, Lina Yao
Abstract: Continual learning (CL) aims to help deep neural networks to learn new knowledge while retaining what has been learned. Recently, pre-trained vision-language models such as CLIP, with powerful generalization ability, have been gaining traction as practical CL candidates. However, the domain mismatch between the pre-training and the downstream CL tasks calls for finetuning of the CLIP on the latter. The deterministic nature of the existing finetuning methods makes them overlook the many possible interactions across the modalities and deems them unsafe for high-risk CL tasks requiring reliable uncertainty estimation. To address these, our work proposes Continual LeArning with Probabilistic finetuning (CLAP). CLAP develops probabilistic modeling over task-specific modules with visual-guided text features, providing more reliable fine-tuning in CL. It further alleviates forgetting by exploiting the rich pre-trained knowledge of CLIP for weight initialization and distribution regularization of task-specific modules. Cooperating with the diverse range of existing prompting methods, CLAP can surpass the predominant deterministic finetuning approaches for CL with CLIP. Lastly, we study the superior uncertainty estimation abilities of CLAP for novel data detection and exemplar selection within CL setups. Our code is available at \url{https://github.com/srvCodes/clap4clip}.
Authors: Huanpeng Chu, Wei Wu, Chengjie Zang, Kun Yuan
Abstract: Diffusion models have revolutionized image synthesis, setting new benchmarks in quality and creativity. However, their widespread adoption is hindered by the intensive computation required during the iterative denoising process. Post-training quantization (PTQ) presents a solution to accelerate sampling, aibeit at the expense of sample quality, extremely in low-bit settings. Addressing this, our study introduces a unified Quantization Noise Correction Scheme (QNCD), aimed at minishing quantization noise throughout the sampling process. We identify two primary quantization challenges: intra and inter quantization noise. Intra quantization noise, mainly exacerbated by embeddings in the resblock module, extends activation quantization ranges, increasing disturbances in each single denosing step. Besides, inter quantization noise stems from cumulative quantization deviations across the entire denoising process, altering data distributions step-by-step. QNCD combats these through embedding-derived feature smoothing for eliminating intra quantization noise and an effective runtime noise estimatiation module for dynamicly filtering inter quantization noise. Extensive experiments demonstrate that our method outperforms previous quantization methods for diffusion models, achieving lossless results in W4A8 and W8A8 quantization settings on ImageNet (LDM-4). Code is available at: https://github.com/huanpengchu/QNCD
Authors: Seyeon Kim, Siyoon Jin, Jihye Park, Kihong Kim, Jiyoung Kim, Jisu Nam, Seungryong Kim
Abstract: Conventional GAN-based models for talking head generation often suffer from limited quality and unstable training. Recent approaches based on diffusion models aimed to address these limitations and improve fidelity. However, they still face challenges, including extensive sampling times and difficulties in maintaining temporal consistency due to the high stochasticity of diffusion models. To overcome these challenges, we propose a novel motion-disentangled diffusion model for high-quality talking head generation, dubbed MoDiTalker. We introduce the two modules: audio-to-motion (AToM), designed to generate a synchronized lip motion from audio, and motion-to-video (MToV), designed to produce high-quality head video following the generated motion. AToM excels in capturing subtle lip movements by leveraging an audio attention mechanism. In addition, MToV enhances temporal consistency by leveraging an efficient tri-plane representation. Our experiments conducted on standard benchmarks demonstrate that our model achieves superior performance compared to existing models. We also provide comprehensive ablation studies and user study results.
Authors: Wufei Ma, Jiahao Li, Bin Li, Yan Lu
Abstract: Deep learning-based video compression is a challenging task, and many previous state-of-the-art learning-based video codecs use optical flows to exploit the temporal correlation between successive frames and then compress the residual error. Although these two-stage models are end-to-end optimized, the epistemic uncertainty in the motion estimation and the aleatoric uncertainty from the quantization operation lead to errors in the intermediate representations and introduce artifacts in the reconstructed frames. This inherent flaw limits the potential for higher bit rate savings. To address this issue, we propose an uncertainty-aware video compression model that can effectively capture the predictive uncertainty with deep ensembles. Additionally, we introduce an ensemble-aware loss to encourage the diversity among ensemble members and investigate the benefits of incorporating adversarial training in the video compression task. Experimental results on 1080p sequences show that our model can effectively save bits by more than 20% compared to DVC Pro.
Authors: Yutong Chen, Yifan Zhan, Zhihang Zhong, Wei Wang, Xiao Sun, Yu Qiao, Yinqiang Zheng
Abstract: Neural rendering techniques have significantly advanced 3D human body modeling. However, previous approaches often overlook dynamics induced by factors such as motion inertia, leading to challenges in scenarios like abrupt stops after rotation, where the pose remains static while the appearance changes. This limitation arises from reliance on a single pose as conditional input, resulting in ambiguity in mapping one pose to multiple appearances. In this study, we elucidate that variations in human appearance depend not only on the current frame's pose condition but also on past pose states. Therefore, we introduce Dyco, a novel method utilizing the delta pose sequence representation for non-rigid deformations and canonical space to effectively model temporal appearance variations. To prevent a decrease in the model's generalization ability to novel poses, we further propose low-dimensional global context to reduce unnecessary inter-body part dependencies and a quantization operation to mitigate overfitting of the delta pose sequence by the model. To validate the effectiveness of our approach, we collected a novel dataset named I3D-Human, with a focus on capturing temporal changes in clothing appearance under approximate poses. Through extensive experiments on both I3D-Human and existing datasets, our approach demonstrates superior qualitative and quantitative performance. In addition, our inertia-aware 3D human method can unprecedentedly simulate appearance changes caused by inertia at different velocities.
Authors: Tianhao Zhou, Haipeng Li, Ziyi Wang, Ao Luo, Chen-Lin Zhang, Jiajun Li, Bing Zeng, Shuaicheng Liu
Abstract: Image stitching from different captures often results in non-rectangular boundaries, which is often considered unappealing. To solve non-rectangular boundaries, current solutions involve cropping, which discards image content, inpainting, which can introduce unrelated content, or warping, which can distort non-linear features and introduce artifacts. To overcome these issues, we introduce a novel diffusion-based learning framework, \textbf{RecDiffusion}, for image stitching rectangling. This framework combines Motion Diffusion Models (MDM) to generate motion fields, effectively transitioning from the stitched image's irregular borders to a geometrically corrected intermediary. Followed by Content Diffusion Models (CDM) for image detail refinement. Notably, our sampling process utilizes a weighted map to identify regions needing correction during each iteration of CDM. Our RecDiffusion ensures geometric accuracy and overall visual appeal, surpassing all previous methods in both quantitative and qualitative measures when evaluated on public benchmarks. Code is released at https://github.com/lhaippp/RecDiffusion.
Authors: Tianyi Liu, Zhaorui Tan, Kaizhu Huang, Haochuan Jiang
Abstract: Medical image segmentation presents the challenge of segmenting various-size targets, demanding the model to effectively capture both local and global information. Despite recent efforts using CNNs and ViTs to predict annotations of different scales, these approaches often struggle to effectively balance the detection of targets across varying sizes. Simply utilizing local information from CNNs and global relationships from ViTs without considering potential significant divergence in latent feature distributions may result in substantial information loss. To address this issue, in this paper, we will introduce a novel Stagger Network (SNet) and argues that a well-designed fusion structure can mitigate the divergence in latent feature distributions between CNNs and ViTs, thereby reducing information loss. Specifically, to emphasize both global dependencies and local focus, we design a Parallel Module to bridge the semantic gap. Meanwhile, we propose the Stagger Module, trying to fuse the selected features that are more semantically similar. An Information Recovery Module is further adopted to recover complementary information back to the network. As a key contribution, we theoretically analyze that the proposed parallel and stagger strategies would lead to less information loss, thus certifying the SNet's rationale. Experimental results clearly proved that the proposed SNet excels comparisons with recent SOTAs in segmenting on the Synapse dataset where targets are in various sizes. Besides, it also demonstrates superiority on the ACDC and the MoNuSeg datasets where targets are with more consistent dimensions.
Authors: Yiyu Wang, Hao Luo, Jungang Xu, Yingfei Sun, Fan Wang
Abstract: Supervised image captioning approaches have made great progress, but it is challenging to collect high-quality human-annotated image-text data. Recently, large-scale vision and language models (e.g., CLIP) and large-scale generative language models (e.g., GPT-2) have shown strong performances in various tasks, which also provide some new solutions for image captioning with web paired data, unpaired data or even text-only data. Among them, the mainstream solution is to project image embeddings into the text embedding space with the assistance of consistent representations between image-text pairs from the CLIP model. However, the current methods still face several challenges in adapting to the diversity of data configurations in a unified solution, accurately estimating image-text embedding bias, and correcting unsatisfactory prediction results in the inference stage. This paper proposes a new Text data-centric approach with Interactive Prompts for image Captioning, named TIPCap. 1) We consider four different settings which gradually reduce the dependence on paired data. 2) We construct a mapping module driven by multivariate Gaussian distribution to mitigate the modality gap, which is applicable to the above four different settings. 3) We propose a prompt interaction module that can incorporate optional prompt information before generating captions. Extensive experiments show that our TIPCap outperforms other weakly or unsupervised image captioning methods and achieves a new state-of-the-art performance on two widely used datasets, i.e., MS-COCO and Flickr30K.
Authors: Hemanth Saratchandran, Sameera Ramasinghe, Simon Lucey
Abstract: In the realm of computer vision, Neural Fields have gained prominence as a contemporary tool harnessing neural networks for signal representation. Despite the remarkable progress in adapting these networks to solve a variety of problems, the field still lacks a comprehensive theoretical framework. This article aims to address this gap by delving into the intricate interplay between initialization and activation, providing a foundational basis for the robust optimization of Neural Fields. Our theoretical insights reveal a deep-seated connection among network initialization, architectural choices, and the optimization process, emphasizing the need for a holistic approach when designing cutting-edge Neural Fields.
Authors: Weihao Jiang, Zhaozhi Xie, Yuxiang Lu, Longjie Qi, Jingyong Cai, Hiroyuki Uchiyama, Bin Chen, Yue Ding, Hongtao Lu
Abstract: Mask-guided matting networks have achieved significant improvements and have shown great potential in practical applications in recent years. However, simply learning matting representation from synthetic and lack-of-real-world-diversity matting data, these approaches tend to overfit low-level details in wrong regions, lack generalization to objects with complex structures and real-world scenes such as shadows, as well as suffer from interference of background lines or textures. To address these challenges, in this paper, we propose a novel auxiliary learning framework for mask-guided matting models, incorporating three auxiliary tasks: semantic segmentation, edge detection, and background line detection besides matting, to learn different and effective representations from different types of data and annotations. Our framework and model introduce the following key aspects: (1) to learn real-world adaptive semantic representation for objects with diverse and complex structures under real-world scenes, we introduce extra semantic segmentation and edge detection tasks on more diverse real-world data with segmentation annotations; (2) to avoid overfitting on low-level details, we propose a module to utilize the inconsistency between learned segmentation and matting representations to regularize detail refinement; (3) we propose a novel background line detection task into our auxiliary learning framework, to suppress interference of background lines or textures. In addition, we propose a high-quality matting benchmark, Plant-Mat, to evaluate matting methods on complex structures. Extensively quantitative and qualitative results show that our approach outperforms state-of-the-art mask-guided methods.
Authors: Shengjun Zhang, Xin Fei, Yueqi Duan
Abstract: Point clouds captured by different sensors such as RGB-D cameras and LiDAR possess non-negligible domain gaps. Most existing methods design different network architectures and train separately on point clouds from various sensors. Typically, point-based methods achieve outstanding performances on even-distributed dense point clouds from RGB-D cameras, while voxel-based methods are more efficient for large-range sparse LiDAR point clouds. In this paper, we propose geometry-to-voxel auxiliary learning to enable voxel representations to access point-level geometric information, which supports better generalisation of the voxel-based backbone with additional interpretations of multi-sensor point clouds. Specifically, we construct hierarchical geometry pools generated by a voxel-guided dynamic point network, which efficiently provide auxiliary fine-grained geometric information adapted to different stages of voxel features. We conduct experiments on joint multi-sensor datasets to demonstrate the effectiveness of GeoAuxNet. Enjoying elaborate geometric information, our method outperforms other models collectively trained on multi-sensor datasets, and achieve competitive results with the-state-of-art experts on each single dataset.
Authors: Sishuo Chen, Lei Li, Shuhuai Ren, Rundong Gao, Yuanxin Liu, Xiaohan Bi, Xu Sun, Lu Hou
Abstract: Video paragraph captioning (VPC) involves generating detailed narratives for long videos, utilizing supportive modalities such as speech and event boundaries. However, the existing models are constrained by the assumption of constant availability of a single auxiliary modality, which is impractical given the diversity and unpredictable nature of real-world scenarios. To this end, we propose a Missing-Resistant framework MR-VPC that effectively harnesses all available auxiliary inputs and maintains resilience even in the absence of certain modalities. Under this framework, we propose the Multimodal VPC (MVPC) architecture integrating video, speech, and event boundary inputs in a unified manner to process various auxiliary inputs. Moreover, to fortify the model against incomplete data, we introduce DropAM, a data augmentation strategy that randomly omits auxiliary inputs, paired with DistillAM, a regularization target that distills knowledge from teacher models trained on modality-complete data, enabling efficient learning in modality-deficient environments. Through exhaustive experimentation on YouCook2 and ActivityNet Captions, MR-VPC has proven to deliver superior performance on modality-complete and modality-missing test data. This work highlights the significance of developing resilient VPC models and paves the way for more adaptive, robust multimodal video understanding.
Authors: Angchi Xu, Wei-Shi Zheng
Abstract: Weakly-supervised action segmentation is a task of learning to partition a long video into several action segments, where training videos are only accompanied by transcripts (ordered list of actions). Most of existing methods need to infer pseudo segmentation for training by serial alignment between all frames and the transcript, which is time-consuming and hard to be parallelized while training. In this work, we aim to escape from this inefficient alignment with massive but redundant frames, and instead to directly localize a few action transitions for pseudo segmentation generation, where a transition refers to the change from an action segment to its next adjacent one in the transcript. As the true transitions are submerged in noisy boundaries due to intra-segment visual variation, we propose a novel Action-Transition-Aware Boundary Alignment (ATBA) framework to efficiently and effectively filter out noisy boundaries and detect transitions. In addition, to boost the semantic learning in the case that noise is inevitably present in the pseudo segmentation, we also introduce video-level losses to utilize the trusted video-level supervision. Extensive experiments show the effectiveness of our approach on both performance and training speed.
Authors: Junghyup Lee, Bumsub Ham
Abstract: Training-free network architecture search (NAS) aims to discover high-performing networks with zero-cost proxies, capturing network characteristics related to the final performance. However, network rankings estimated by previous training-free NAS methods have shown weak correlations with the performance. To address this issue, we propose AZ-NAS, a novel approach that leverages the ensemble of various zero-cost proxies to enhance the correlation between a predicted ranking of networks and the ground truth substantially in terms of the performance. To achieve this, we introduce four novel zero-cost proxies that are complementary to each other, analyzing distinct traits of architectures in the views of expressivity, progressivity, trainability, and complexity. The proxy scores can be obtained simultaneously within a single forward and backward pass, making an overall NAS process highly efficient. In order to integrate the rankings predicted by our proxies effectively, we introduce a non-linear ranking aggregation method that highlights the networks highly-ranked consistently across all the proxies. Experimental results conclusively demonstrate the efficacy and efficiency of AZ-NAS, outperforming state-of-the-art methods on standard benchmarks, all while maintaining a reasonable runtime cost.
Authors: Haonan Lin, Mengmeng Wang, Yan Chen, Wenbin An, Yuzhe Yao, Guang Dai, Qianying Wang, Yong Liu, Jingdong Wang
Abstract: While large-scale pre-trained text-to-image models can synthesize diverse and high-quality human-centered images, novel challenges arise with a nuanced task of "identity fine editing": precisely modifying specific features of a subject while maintaining its inherent identity and context. Existing personalization methods either require time-consuming optimization or learning additional encoders, adept in "identity re-contextualization". However, they often struggle with detailed and sensitive tasks like human face editing. To address these challenges, we introduce DreamSalon, a noise-guided, staged-editing framework, uniquely focusing on detailed image manipulations and identity-context preservation. By discerning editing and boosting stages via the frequency and gradient of predicted noises, DreamSalon first performs detailed manipulations on specific features in the editing stage, guided by high-frequency information, and then employs stochastic denoising in the boosting stage to improve image quality. For more precise editing, DreamSalon semantically mixes source and target textual prompts, guided by differences in their embedding covariances, to direct the model's focus on specific manipulation areas. Our experiments demonstrate DreamSalon's ability to efficiently and faithfully edit fine details on human faces, outperforming existing methods both qualitatively and quantitatively.
Authors: Sidi Yang, Binxiao Huang, Mingdeng Cao, Yatai Ji, Hanzhong Guo, Ngai Wong, Yujiu Yang
Abstract: The widespread use of high-definition screens in edge devices, such as end-user cameras, smartphones, and televisions, is spurring a significant demand for image enhancement. Existing enhancement models often optimize for high performance while falling short of reducing hardware inference time and power consumption, especially on edge devices with constrained computing and storage resources. To this end, we propose Image Color Enhancement Lookup Table (ICELUT) that adopts LUTs for extremely efficient edge inference, without any convolutional neural network (CNN). During training, we leverage pointwise (1x1) convolution to extract color information, alongside a split fully connected layer to incorporate global information. Both components are then seamlessly converted into LUTs for hardware-agnostic deployment. ICELUT achieves near-state-of-the-art performance and remarkably low power consumption. We observe that the pointwise network structure exhibits robust scalability, upkeeping the performance even with a heavily downsampled 32x32 input image. These enable ICELUT, the first-ever purely LUT-based image enhancer, to reach an unprecedented speed of 0.4ms on GPU and 7ms on CPU, at least one order faster than any CNN solution. Codes are available at https://github.com/Stephen0808/ICELUT.
Authors: Yuqing Huang, Xin Li, Zikun Zhou, Yaowei Wang, Zhenyu He, Ming-Hsuan Yang
Abstract: Existing tracking methods mainly focus on learning better target representation or developing more robust prediction models to improve tracking performance. While tracking performance has significantly improved, the target loss issue occurs frequently due to tracking failures, complete occlusion, or out-of-view situations. However, considerably less attention is paid to the self-recovery issue of tracking methods, which is crucial for practical applications. To this end, we propose a recoverable tracking framework, RTracker, that uses a tree-structured memory to dynamically associate a tracker and a detector to enable self-recovery ability. Specifically, we propose a Positive-Negative Tree-structured memory to chronologically store and maintain positive and negative target samples. Upon the PN tree memory, we develop corresponding walking rules for determining the state of the target and define a set of control flows to unite the tracker and the detector in different tracking scenarios. Our core idea is to use the support samples of positive and negative target categories to establish a relative distance-based criterion for a reliable assessment of target loss. The favorable performance in comparison against the state-of-the-art methods on numerous challenging benchmarks demonstrates the effectiveness of the proposed algorithm.
Authors: Namhyuk Ahn, Wonhyuk Ahn, KiYoon Yoo, Daesik Kim, Seung-Hun Nam
Abstract: Recent progress in diffusion models has profoundly enhanced the fidelity of image generation. However, this has raised concerns about copyright infringements. While prior methods have introduced adversarial perturbations to prevent style imitation, most are accompanied by the degradation of artworks' visual quality. Recognizing the importance of maintaining this, we develop a visually improved protection method that preserves its protection capability. To this end, we create a perceptual map to identify areas most sensitive to human eyes. We then adjust the protection intensity guided by an instance-aware refinement. We also integrate a perceptual constraints bank to further improve the imperceptibility. Results show that our method substantially elevates the quality of the protected image without compromising on protection efficacy.
Authors: Beerend G. A. Gerats, Jelmer M. Wolterink, Seb P. Mol, Ivo A. M. J. Broeders
Abstract: Laparoscopic video tracking primarily focuses on two target types: surgical instruments and anatomy. The former could be used for skill assessment, while the latter is necessary for the projection of virtual overlays. Where instrument and anatomy tracking have often been considered two separate problems, in this paper, we propose a method for joint tracking of all structures simultaneously. Based on a single 2D monocular video clip, we train a neural field to represent a continuous spatiotemporal scene, used to create 3D tracks of all surfaces visible in at least one frame. Due to the small size of instruments, they generally cover a small part of the image only, resulting in decreased tracking accuracy. Therefore, we propose enhanced class weighting to improve the instrument tracks. We evaluate tracking on video clips from laparoscopic cholecystectomies, where we find mean tracking accuracies of 92.4% for anatomical structures and 87.4% for instruments. Additionally, we assess the quality of depth maps obtained from the method's scene reconstructions. We show that these pseudo-depths have comparable quality to a state-of-the-art pre-trained depth estimator. On laparoscopic videos in the SCARED dataset, the method predicts depth with an MAE of 2.9 mm and a relative error of 9.2%. These results show the feasibility of using neural fields for monocular 3D reconstruction of laparoscopic scenes.
Authors: Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, Robby T. Tan
Abstract: Domain adaptive object detection aims to adapt detection models to domains where annotated data is unavailable. Existing methods have been proposed to address the domain gap using the semi-supervised student-teacher framework. However, a fundamental issue arises from the class imbalance in the labelled training set, which can result in inaccurate pseudo-labels. The relationship between classes, especially where one class is a majority and the other minority, has a large impact on class bias. We propose Class-Aware Teacher (CAT) to address the class bias issue in the domain adaptation setting. In our work, we approximate the class relationships with our Inter-Class Relation module (ICRm) and exploit it to reduce the bias within the model. In this way, we are able to apply augmentations to highly related classes, both inter- and intra-domain, to boost the performance of minority classes while having minimal impact on majority classes. We further reduce the bias by implementing a class-relation weight to our classification loss. Experiments conducted on various datasets and ablation studies show that our method is able to address the class bias in the domain adaptation setting. On the Cityscapes to Foggy Cityscapes dataset, we attained a 52.5 mAP, a substantial improvement over the 51.2 mAP achieved by the state-of-the-art method.
Authors: Yiyang Sun, Zhiyuan Xu, Xiaonian Wang, Jing Yao
Abstract: Self-supervised multi-frame methods have currently achieved promising results in depth estimation. However, these methods often suffer from mismatch problems due to the moving objects, which break the static assumption. Additionally, unfairness can occur when calculating photometric errors in high-freq or low-texture regions of the images. To address these issues, existing approaches use additional semantic priori black-box networks to separate moving objects and improve the model only at the loss level. Therefore, we propose FlowDepth, where a Dynamic Motion Flow Module (DMFM) decouples the optical flow by a mechanism-based approach and warps the dynamic regions thus solving the mismatch problem. For the unfairness of photometric errors caused by high-freq and low-texture regions, we use Depth-Cue-Aware Blur (DCABlur) and Cost-Volume sparsity loss respectively at the input and the loss level to solve the problem. Experimental results on the KITTI and Cityscapes datasets show that our method outperforms the state-of-the-art methods.
Authors: Tian Ma, Chuyang Shang, Wanzhu Ren, Yuancheng Li, Jiiayi Yang, Jiali Qian
Abstract: In recent years, research on point weakly supervised object detection (PWSOD) methods in the field of computer vision has attracted people's attention. However, existing pseudo labels generation methods perform poorly in a small amount of supervised annotation data and dense object detection tasks. We consider the generation of weakly supervised pseudo labels as the result of model's sparse output, and propose a method called Sparse Generation to make pseudo labels sparse. It constructs dense tensors through the relationship between data and detector model, optimizes three of its parameters, and obtains a sparse tensor via coordinated calculation, thereby indirectly obtaining higher quality pseudo labels, and solving the model's density problem in the situation of only a small amount of supervised annotation data can be used. On two broadly used open-source datasets (RSOD, SIMD) and a self-built dataset (Bullet-Hole), the experimental results showed that the proposed method has a significant advantage in terms of overall performance metrics, comparing to that state-of-the-art method.
Authors: Xiaoyang Lyu, Chirui Chang, Peng Dai, Yang-tian Sun, Xiaojuang Qi
Abstract: Scene reconstruction from multi-view images is a fundamental problem in computer vision and graphics. Recent neural implicit surface reconstruction methods have achieved high-quality results; however, editing and manipulating the 3D geometry of reconstructed scenes remains challenging due to the absence of naturally decomposed object entities and complex object/background compositions. In this paper, we present Total-Decom, a novel method for decomposed 3D reconstruction with minimal human interaction. Our approach seamlessly integrates the Segment Anything Model (SAM) with hybrid implicit-explicit neural surface representations and a mesh-based region-growing technique for accurate 3D object decomposition. Total-Decom requires minimal human annotations while providing users with real-time control over the granularity and quality of decomposition. We extensively evaluate our method on benchmark datasets and demonstrate its potential for downstream applications, such as animation and scene editing. The code is available at \href{https://github.com/CVMI-Lab/Total-Decom.git}{https://github.com/CVMI-Lab/Total-Decom.git}.
URLs: https://github.com/CVMI-Lab/Total-Decom.git, https://github.com/CVMI-Lab/Total-Decom.git
Authors: Yue Gao, Jiaxuan Lu, Siqi Li, Yipeng Li, Shaoyi Du
Abstract: Action recognition from video data forms a cornerstone with wide-ranging applications. Single-view action recognition faces limitations due to its reliance on a single viewpoint. In contrast, multi-view approaches capture complementary information from various viewpoints for improved accuracy. Recently, event cameras have emerged as innovative bio-inspired sensors, leading to advancements in event-based action recognition. However, existing works predominantly focus on single-view scenarios, leaving a gap in multi-view event data exploitation, particularly in challenges like information deficit and semantic misalignment. To bridge this gap, we introduce HyperMV, a multi-view event-based action recognition framework. HyperMV converts discrete event data into frame-like representations and extracts view-related features using a shared convolutional network. By treating segments as vertices and constructing hyperedges using rule-based and KNN-based strategies, a multi-view hypergraph neural network that captures relationships across viewpoint and temporal features is established. The vertex attention hypergraph propagation is also introduced for enhanced feature fusion. To prompt research in this area, we present the largest multi-view event-based action dataset $\text{THU}^{\text{MV-EACT}}\text{-50}$, comprising 50 actions from 6 viewpoints, which surpasses existing datasets by over tenfold. Experimental results show that HyperMV significantly outperforms baselines in both cross-subject and cross-view scenarios, and also exceeds the state-of-the-arts in frame-based multi-view action recognition.
Authors: Yujin Chen, Yinyu Nie, Benjamin Ummenhofer, Reiner Birkl, Michael Paulitsch, Matthias M\"uller, Matthias Nie{\ss}ner
Abstract: We present Mesh2NeRF, an approach to derive ground-truth radiance fields from textured meshes for 3D generation tasks. Many 3D generative approaches represent 3D scenes as radiance fields for training. Their ground-truth radiance fields are usually fitted from multi-view renderings from a large-scale synthetic 3D dataset, which often results in artifacts due to occlusions or under-fitting issues. In Mesh2NeRF, we propose an analytic solution to directly obtain ground-truth radiance fields from 3D meshes, characterizing the density field with an occupancy function featuring a defined surface thickness, and determining view-dependent color through a reflection function considering both the mesh and environment lighting. Mesh2NeRF extracts accurate radiance fields which provides direct supervision for training generative NeRFs and single scene representation. We validate the effectiveness of Mesh2NeRF across various tasks, achieving a noteworthy 3.12dB improvement in PSNR for view synthesis in single scene representation on the ABO dataset, a 0.69 PSNR enhancement in the single-view conditional generation of ShapeNet Cars, and notably improved mesh extraction from NeRF in the unconditional generation of Objaverse Mugs.
Authors: Jiaxing Chen, Yuxuan Liu, Dehu Li, Xiang An, Ziyong Feng, Yongle Zhao, Yin Xie
Abstract: The surge of Multimodal Large Language Models (MLLMs), given their prominent emergent capabilities in instruction following and reasoning, has greatly advanced the field of visual reasoning. However, constrained by their non-lossless image tokenization, most MLLMs fall short of comprehensively capturing details of text and objects, especially in high-resolution images. To address this, we propose P2G, a novel framework for plug-and-play grounding of reasoning in MLLMs. Specifically, P2G exploits the tool-usage potential of MLLMs to employ expert agents to achieve on-the-fly grounding to critical visual and textual objects of image, thus achieving deliberate reasoning via multimodal prompting. We further create P2GB, a benchmark aimed at assessing MLLMs' ability to understand inter-object relationships and text in challenging high-resolution images. Comprehensive experiments on visual reasoning tasks demonstrate the superiority of P2G. Noteworthy, P2G achieved comparable performance with GPT-4V on P2GB, with a 7B backbone. Our work highlights the potential of plug-and-play grounding of reasoning and opens up a promising alternative beyond model scaling.
Authors: Qianyu Zhou, Ke-Yue Zhang, Taiping Yao, Xuequan Lu, Shouhong Ding, Lizhuang Ma
Abstract: Face Anti-Spoofing (FAS) is pivotal in safeguarding facial recognition systems against presentation attacks. While domain generalization (DG) methods have been developed to enhance FAS performance, they predominantly focus on learning domain-invariant features during training, which may not guarantee generalizability to unseen data that differs largely from the source distributions. Our insight is that testing data can serve as a valuable resource to enhance the generalizability beyond mere evaluation for DG FAS. In this paper, we introduce a novel Test-Time Domain Generalization (TTDG) framework for FAS, which leverages the testing data to boost the model's generalizability. Our method, consisting of Test-Time Style Projection (TTSP) and Diverse Style Shifts Simulation (DSSS), effectively projects the unseen data to the seen domain space. In particular, we first introduce the innovative TTSP to project the styles of the arbitrarily unseen samples of the testing distribution to the known source space of the training distributions. We then design the efficient DSSS to synthesize diverse style shifts via learnable style bases with two specifically designed losses in a hyperspherical feature space. Our method eliminates the need for model updates at the test time and can be seamlessly integrated into not only the CNN but also ViT backbones. Comprehensive experiments on widely used cross-domain FAS benchmarks demonstrate our method's state-of-the-art performance and effectiveness.
Authors: Jiacui Huang, Hongtao Zhang, Mingbo Zhao, Zhou Wu
Abstract: Vision-and-Language Navigation (VLN) is a challenging task that requires a robot to navigate in photo-realistic environments with human natural language promptings. Recent studies aim to handle this task by constructing the semantic spatial map representation of the environment, and then leveraging the strong ability of reasoning in large language models for generalizing code for guiding the robot navigation. However, these methods face limitations in instance-level and attribute-level navigation tasks as they cannot distinguish different instances of the same object. To address this challenge, we propose a new method, namely, Instance-aware Visual Language Map (IVLMap), to empower the robot with instance-level and attribute-level semantic mapping, where it is autonomously constructed by fusing the RGBD video data collected from the robot agent with special-designed natural language map indexing in the bird's-in-eye view. Such indexing is instance-level and attribute-level. In particular, when integrated with a large language model, IVLMap demonstrates the capability to i) transform natural language into navigation targets with instance and attribute information, enabling precise localization, and ii) accomplish zero-shot end-to-end navigation tasks based on natural language commands. Extensive navigation experiments are conducted. Simulation results illustrate that our method can achieve an average improvement of 14.4\% in navigation accuracy. Code and demo are released at https://ivlmap.github.io/.
Authors: Qiankun Liu, Rui Liu, Bolun Zheng, Hongkui Wang, Ying Fu
Abstract: Recently, infrared small target detection (IRSTD) has been dominated by deep-learning-based methods. However, these methods mainly focus on the design of complex model structures to extract discriminative features, leaving the loss functions for IRSTD under-explored. For example, the widely used Intersection over Union (IoU) and Dice losses lack sensitivity to the scales and locations of targets, limiting the detection performance of detectors. In this paper, we focus on boosting detection performance with a more effective loss but a simpler model structure. Specifically, we first propose a novel Scale and Location Sensitive (SLS) loss to handle the limitations of existing losses: 1) for scale sensitivity, we compute a weight for the IoU loss based on target scales to help the detector distinguish targets with different scales: 2) for location sensitivity, we introduce a penalty term based on the center points of targets to help the detector localize targets more precisely. Then, we design a simple Multi-Scale Head to the plain U-Net (MSHNet). By applying SLS loss to each scale of the predictions, our MSHNet outperforms existing state-of-the-art methods by a large margin. In addition, the detection performance of existing detectors can be further improved when trained with our SLS loss, demonstrating the effectiveness and generalization of our SLS loss. The code is available at https://github.com/ying-fu/MSHNet.
Authors: Matteo Caligiuri, Adriano Simonetto, Gianluca Agresti, Pietro Zanuttigh
Abstract: The acquisition of objects outside the Line-of-Sight of cameras is a very intriguing but also extremely challenging research topic. Recent works showed the feasibility of this idea exploiting transient imaging data produced by custom direct Time of Flight sensors. In this paper, for the first time, we tackle this problem using only data from an off-the-shelf indirect Time of Flight sensor without any further hardware requirement. We introduced a Deep Learning model able to re-frame the surfaces where light bounces happen as a virtual mirror. This modeling makes the task easier to handle and also facilitates the construction of annotated training data. From the obtained data it is possible to retrieve the depth information of the hidden scene. We also provide a first-in-its-kind synthetic dataset for the task and demonstrate the feasibility of the proposed idea over it.
Authors: Yanglin Feng, Yang Qin, Dezhong Peng, Hongyuan Zhu, Xi Peng, Peng Hu
Abstract: In this paper, we present and study a new instance-level retrieval task: PointCloud-Text Matching~(PTM), which aims to find the exact cross-modal instance that matches a given point-cloud query or text query. PTM could be applied to various scenarios, such as indoor/urban-canyon localization and scene retrieval. However, there exists no suitable and targeted dataset for PTM in practice. Therefore, we construct three new PTM benchmark datasets, namely 3D2T-SR, 3D2T-NR, and 3D2T-QA. We observe that the data is challenging and with noisy correspondence due to the sparsity, noise, or disorder of point clouds and the ambiguity, vagueness, or incompleteness of texts, which make existing cross-modal matching methods ineffective for PTM. To tackle these challenges, we propose a PTM baseline, named Robust PointCloud-Text Matching method (RoMa). RoMa consists of two modules: a Dual Attention Perception module (DAP) and a Robust Negative Contrastive Learning module (RNCL). Specifically, DAP leverages token-level and feature-level attention to adaptively focus on useful local and global features, and aggregate them into common representations, thereby reducing the adverse impact of noise and ambiguity. To handle noisy correspondence, RNCL divides negative pairs, which are much less error-prone than positive pairs, into clean and noisy subsets, and assigns them forward and reverse optimization directions respectively, thus enhancing robustness against noisy correspondence. We conduct extensive experiments on our benchmarks and demonstrate the superiority of our RoMa.
Authors: Bo Miao, Mohammed Bennamoun, Yongsheng Gao, Mubarak Shah, Ajmal Mian
Abstract: Referring Video Object Segmentation (R-VOS) methods face challenges in maintaining consistent object segmentation due to temporal context variability and the presence of other visually similar objects. We propose an end-to-end R-VOS paradigm that explicitly models temporal instance consistency alongside the referring segmentation. Specifically, we introduce a novel hybrid memory that facilitates inter-frame collaboration for robust spatio-temporal matching and propagation. Features of frames with automatically generated high-quality reference masks are propagated to segment the remaining frames based on multi-granularity association to achieve temporally consistent R-VOS. Furthermore, we propose a new Mask Consistency Score (MCS) metric to evaluate the temporal consistency of video segmentation. Extensive experiments demonstrate that our approach enhances temporal consistency by a significant margin, leading to top-ranked performance on popular R-VOS benchmarks, i.e., Ref-YouTube-VOS (67.1%) and Ref-DAVIS17 (65.6%).
Authors: Hongwei Ren, Jiadong Zhu, Yue Zhou, Haotian FU, Yulong Huang, Bojun Cheng
Abstract: Event cameras exhibit remarkable attributes such as high dynamic range, asynchronicity, and low latency, making them highly suitable for vision tasks that involve high-speed motion in challenging lighting conditions. These cameras implicitly capture movement and depth information in events, making them appealing sensors for Camera Pose Relocalization (CPR) tasks. Nevertheless, existing CPR networks based on events neglect the pivotal fine-grained temporal information in events, resulting in unsatisfactory performance. Moreover, the energy-efficient features are further compromised by the use of excessively complex models, hindering efficient deployment on edge devices. In this paper, we introduce PEPNet, a simple and effective point-based network designed to regress six degrees of freedom (6-DOFs) event camera poses. We rethink the relationship between the event camera and CPR tasks, leveraging the raw Point Cloud directly as network input to harness the high-temporal resolution and inherent sparsity of events. PEPNet is adept at abstracting the spatial and implicit temporal features through hierarchical structure and explicit temporal features by Attentive Bi-directional Long Short-Term Memory (A-Bi-LSTM). By employing a carefully crafted lightweight design, PEPNet delivers state-of-the-art (SOTA) performance on both indoor and outdoor datasets with meager computational resources. Specifically, PEPNet attains a significant 38% and 33% performance improvement on the random split IJRR and M3ED datasets, respectively. Moreover, the lightweight design version PEPNet$_{tiny}$ accomplishes results comparable to the SOTA while employing a mere 0.5% of the parameters.
Authors: Xinyu Zhan, Lixin Yang, Yifei Zhao, Kangrui Mao, Hanlin Xu, Zenan Lin, Kailin Li, Cewu Lu
Abstract: We present OAKINK2, a dataset of bimanual object manipulation tasks for complex daily activities. In pursuit of constructing the complex tasks into a structured representation, OAKINK2 introduces three level of abstraction to organize the manipulation tasks: Affordance, Primitive Task, and Complex Task. OAKINK2 features on an object-centric perspective for decoding the complex tasks, treating them as a sequence of object affordance fulfillment. The first level, Affordance, outlines the functionalities that objects in the scene can afford, the second level, Primitive Task, describes the minimal interaction units that humans interact with the object to achieve its affordance, and the third level, Complex Task, illustrates how Primitive Tasks are composed and interdependent. OAKINK2 dataset provides multi-view image streams and precise pose annotations for the human body, hands and various interacting objects. This extensive collection supports applications such as interaction reconstruction and motion synthesis. Based on the 3-level abstraction of OAKINK2, we explore a task-oriented framework for Complex Task Completion (CTC). CTC aims to generate a sequence of bimanual manipulation to achieve task objectives. Within the CTC framework, we employ Large Language Models (LLMs) to decompose the complex task objectives into sequences of Primitive Tasks and have developed a Motion Fulfillment Model that generates bimanual hand motion for each Primitive Task. OAKINK2 datasets and models are available at https://oakink.net/v2.
URLs: https://oakink.net/v2.
Authors: Kyotaro Tokoro, Kazutoshi Akita, Norimichi Ukita
Abstract: While burst LR images are useful for improving the SR image quality compared with a single LR image, prior SR networks accepting the burst LR images are trained in a deterministic manner, which is known to produce a blurry SR image. In addition, it is difficult to perfectly align the burst LR images, making the SR image more blurry. Since such blurry images are perceptually degraded, we aim to reconstruct the sharp high-fidelity boundaries. Such high-fidelity images can be reconstructed by diffusion models. However, prior SR methods using the diffusion model are not properly optimized for the burst SR task. Specifically, the reverse process starting from a random sample is not optimized for image enhancement and restoration methods, including burst SR. In our proposed method, on the other hand, burst LR features are used to reconstruct the initial burst SR image that is fed into an intermediate step in the diffusion model. This reverse process from the intermediate step 1) skips diffusion steps for reconstructing the global structure of the image and 2) focuses on steps for refining detailed textures. Our experimental results demonstrate that our method can improve the scores of the perceptual quality metrics. Code: https://github.com/placerkyo/BSRD
Authors: Ekkasit Pinyoanuntapong, Muhammad Usama Saleem, Pu Wang, Minwoo Lee, Srijan Das, Chen Chen
Abstract: Generating human motion from text has been dominated by denoising motion models either through diffusion or generative masking process. However, these models face great limitations in usability by requiring prior knowledge of the motion length. Conversely, autoregressive motion models address this limitation by adaptively predicting motion endpoints, at the cost of degraded generation quality and editing capabilities. To address these challenges, we propose Bidirectional Autoregressive Motion Model (BAMM), a novel text-to-motion generation framework. BAMM consists of two key components: (1) a motion tokenizer that transforms 3D human motion into discrete tokens in latent space, and (2) a masked self-attention transformer that autoregressively predicts randomly masked tokens via a hybrid attention masking strategy. By unifying generative masked modeling and autoregressive modeling, BAMM captures rich and bidirectional dependencies among motion tokens, while learning the probabilistic mapping from textual inputs to motion outputs with dynamically-adjusted motion sequence length. This feature enables BAMM to simultaneously achieving high-quality motion generation with enhanced usability and built-in motion editability. Extensive experiments on HumanML3D and KIT-ML datasets demonstrate that BAMM surpasses current state-of-the-art methods in both qualitative and quantitative measures.
Authors: Binyuan Huang, Yuqing Wen, Yucheng Zhao, Yaosi Hu, Yingfei Liu, Fan Jia, Weixin Mao, Tiancai Wang, Chi Zhang, Chang Wen Chen, Zhenzhong Chen, Xiangyu Zhang
Abstract: Autonomous driving progress relies on large-scale annotated datasets. In this work, we explore the potential of generative models to produce vast quantities of freely-labeled data for autonomous driving applications and present SubjectDrive, the first model proven to scale generative data production in a way that could continuously improve autonomous driving applications. We investigate the impact of scaling up the quantity of generative data on the performance of downstream perception models and find that enhancing data diversity plays a crucial role in effectively scaling generative data production. Therefore, we have developed a novel model equipped with a subject control mechanism, which allows the generative model to leverage diverse external data sources for producing varied and useful data. Extensive evaluations confirm SubjectDrive's efficacy in generating scalable autonomous driving training data, marking a significant step toward revolutionizing data production methods in this field.
Authors: Yu Xu, Fan Tang, Juan Cao, Yuxin Zhang, Oliver Deussen, Weiming Dong, Jintao Li, Tong-Yee Lee
Abstract: Personalized generation paradigms empower designers to customize visual intellectual properties with the help of textual descriptions by tuning or adapting pre-trained text-to-image models on a few images. Recent works explore approaches for concurrently customizing both content and detailed visual style appearance. However, these existing approaches often generate images where the content and style are entangled. In this study, we reconsider the customization of content and style concepts from the perspective of parameter space construction. Unlike existing methods that utilize a shared parameter space for content and style, we propose a learning framework that separates the parameter space to facilitate individual learning of content and style, thereby enabling disentangled content and style. To achieve this goal, we introduce "partly learnable projection" (PLP) matrices to separate the original adapters into divided sub-parameter spaces. We propose "break-for-make" customization learning pipeline based on PLP, which is simple yet effective. We break the original adapters into "up projection" and "down projection", train content and style PLPs individually with the guidance of corresponding textual prompts in the separate adapters, and maintain generalization by employing a multi-correspondence projection learning strategy. Based on the adapters broken apart for separate training content and style, we then make the entity parameter space by reconstructing the content and style PLPs matrices, followed by fine-tuning the combined adapter to generate the target object with the desired appearance. Experiments on various styles, including textures, materials, and artistic style, show that our method outperforms state-of-the-art single/multiple concept learning pipelines in terms of content-style-prompt alignment.
Authors: Mingze Sun, Chao Xu, Xinyu Jiang, Yang Liu, Baigui Sun, Ruqi Huang
Abstract: In this paper, we introduce an innovative task focused on human communication, aiming to generate 3D holistic human motions for both speakers and listeners. Central to our approach is the incorporation of factorization to decouple audio features and the combination of textual semantic information, thereby facilitating the creation of more realistic and coordinated movements. We separately train VQ-VAEs with respect to the holistic motions of both speaker and listener. We consider the real-time mutual influence between the speaker and the listener and propose a novel chain-like transformer-based auto-regressive model specifically designed to characterize real-world communication scenarios effectively which can generate the motions of both the speaker and the listener simultaneously. These designs ensure that the results we generate are both coordinated and diverse. Our approach demonstrates state-of-the-art performance on two benchmark datasets. Furthermore, we introduce the HoCo holistic communication dataset, which is a valuable resource for future research. Our HoCo dataset and code will be released for research purposes upon acceptance.
Authors: Tongyan Hua, Lin Wang
Abstract: Implicit neural representation (INR), in combination with geometric rendering, has recently been employed in real-time dense RGB-D SLAM. Despite active research endeavors being made, there lacks a unified protocol for fair evaluation, impeding the evolution of this area. In this work, we establish, to our knowledge, the first open-source benchmark framework to evaluate the performance of a wide spectrum of commonly used INRs and rendering functions for mapping and localization. The goal of our benchmark is to 1) gain an intuition of how different INRs and rendering functions impact mapping and localization and 2) establish a unified evaluation protocol w.r.t. the design choices that may impact the mapping and localization. With the framework, we conduct a large suite of experiments, offering various insights in choosing the INRs and geometric rendering functions: for example, the dense feature grid outperforms other INRs (e.g. tri-plane and hash grid), even when geometric and color features are jointly encoded for memory efficiency. To extend the findings into the practical scenario, a hybrid encoding strategy is proposed to bring the best of the accuracy and completion from the grid-based and decomposition-based INRs. We further propose explicit hybrid encoding for high-fidelity dense grid mapping to comply with the RGB-D SLAM system that puts the premise on robustness and computation efficiency.
Authors: Yaxu Xie, Alain Pagani, Didier Stricker
Abstract: Scene graphs have been recently introduced into 3D spatial understanding as a comprehensive representation of the scene. The alignment between 3D scene graphs is the first step of many downstream tasks such as scene graph aided point cloud registration, mosaicking, overlap checking, and robot navigation. In this work, we treat 3D scene graph alignment as a partial graph-matching problem and propose to solve it with a graph neural network. We reuse the geometric features learned by a point cloud registration method and associate the clustered point-level geometric features with the node-level semantic feature via our designed feature fusion module. Partial matching is enabled by using a learnable method to select the top-k similar node pairs. Subsequent downstream tasks such as point cloud registration are achieved by running a pre-trained registration network within the matched regions. We further propose a point-matching rescoring method, that uses the node-wise alignment of the 3D scene graph to reweight the matching candidates from a pre-trained point cloud registration method. It reduces the false point correspondences estimated especially in low-overlapping cases. Experiments show that our method improves the alignment accuracy by 10~20% in low-overlap and random transformation scenarios and outperforms the existing work in multiple downstream tasks.
Authors: Alireza Ganjdanesh, Shangqian Gao, Heng Huang
Abstract: Structural model pruning is a prominent approach used for reducing the computational cost of Convolutional Neural Networks (CNNs) before their deployment on resource-constrained devices. Yet, the majority of proposed ideas require a pretrained model before pruning, which is costly to secure. In this paper, we propose a novel structural pruning approach to jointly learn the weights and structurally prune architectures of CNN models. The core element of our method is a Reinforcement Learning (RL) agent whose actions determine the pruning ratios of the CNN model's layers, and the resulting model's accuracy serves as its reward. We conduct the joint training and pruning by iteratively training the model's weights and the agent's policy, and we regularize the model's weights to align with the selected structure by the agent. The evolving model's weights result in a dynamic reward function for the agent, which prevents using prominent episodic RL methods with stationary environment assumption for our purpose. We address this challenge by designing a mechanism to model the complex changing dynamics of the reward function and provide a representation of it to the RL agent. To do so, we take a learnable embedding for each training epoch and employ a recurrent model to calculate a representation of the changing environment. We train the recurrent model and embeddings using a decoder model to reconstruct observed rewards. Such a design empowers our agent to effectively leverage episodic observations along with the environment representations to learn a proper policy to determine performant sub-networks of the CNN model. Our extensive experiments on CIFAR-10 and ImageNet using ResNets and MobileNets demonstrate the effectiveness of our method.
Authors: Andrii Kompanets (Bert), Remco Duits (Bert), Davide Leonetti (Bert), Nicky van den Berg (Bert), H. H. (Bert), Snijder
Abstract: Safety-critical infrastructures, such as bridges, are periodically inspected to check for existing damage, such as fatigue cracks and corrosion, and to guarantee the safe use of the infrastructure. Visual inspection is the most frequent type of general inspection, despite the fact that its detection capability is rather limited, especially for fatigue cracks. Machine learning algorithms can be used for augmenting the capability of classical visual inspection of bridge structures, however, the implementation of such an algorithm requires a massive annotated training dataset, which is time-consuming to produce. This paper proposes a semi-automatic crack segmentation tool that eases the manual segmentation of cracks on images needed to create a training dataset for a machine learning algorithm. Also, it can be used to measure the geometry of the crack. This tool makes use of an image processing algorithm, which was initially developed for the analysis of vascular systems on retinal images. The algorithm relies on a multi-orientation wavelet transform, which is applied to the image to construct the so-called "orientation scores", i.e. a modified version of the image. Afterwards, the filtered orientation scores are used to formulate an optimal path problem that identifies the crack. The globally optimal path between manually selected crack endpoints is computed, using a state-of-the-art geometric tracking method. The pixel-wise segmentation is done afterwards using the obtained crack path. The proposed method outperforms fully automatic methods and shows potential to be an adequate alternative to the manual data annotation.
Authors: Avinash Paliwal, Wei Ye, Jinhui Xiong, Dmytro Kotovenko, Rakesh Ranjan, Vikas Chandra, Nima Khademi Kalantari
Abstract: The field of 3D reconstruction from images has rapidly evolved in the past few years, first with the introduction of Neural Radiance Field (NeRF) and more recently with 3D Gaussian Splatting (3DGS). The latter provides a significant edge over NeRF in terms of the training and inference speed, as well as the reconstruction quality. Although 3DGS works well for dense input images, the unstructured point-cloud like representation quickly overfits to the more challenging setup of extremely sparse input images (e.g., 3 images), creating a representation that appears as a jumble of needles from novel views. To address this issue, we propose regularized optimization and depth-based initialization. Our key idea is to introduce a structured Gaussian representation that can be controlled in 2D image space. We then constraint the Gaussians, in particular their position, and prevent them from moving independently during optimization. Specifically, we introduce single and multiview constraints through an implicit convolutional decoder and a total variation loss, respectively. With the coherency introduced to the Gaussians, we further constrain the optimization through a flow-based loss function. To support our regularized optimization, we propose an approach to initialize the Gaussians using monocular depth estimates at each input view. We demonstrate significant improvements compared to the state-of-the-art sparse-view NeRF-based approaches on a variety of scenes.
Authors: Pulkit Khandelwal, Michael Tran Duong, Constanza Fuentes, Amanda Denning, Winifred Trotman, Ranjit Ittyerah, Alejandra Bahena, Theresa Schuck, Marianna Gabrielyan, Karthik Prabhakaran, Daniel Ohm, Gabor Mizsei, John Robinson, Monica Munoz, John Detre, Edward Lee, David Irwin, Corey McMillan, M. Dylan Tisdall, Sandhitsu Das, David Wolk, Paul A. Yushkevich
Abstract: Magnetic resonance imaging (MRI) is the standard modality to understand human brain structure and function in vivo (antemortem). Decades of research in human neuroimaging has led to the widespread development of methods and tools to provide automated volume-based segmentations and surface-based parcellations which help localize brain functions to specialized anatomical regions. Recently ex vivo (postmortem) imaging of the brain has opened-up avenues to study brain structure at sub-millimeter ultra high-resolution revealing details not possible to observe with in vivo MRI. Unfortunately, there has been limited methodological development in ex vivo MRI primarily due to lack of datasets and limited centers with such imaging resources. Therefore, in this work, we present one-of-its-kind dataset of 82 ex vivo T2w whole brain hemispheres MRI at 0.3 mm isotropic resolution spanning Alzheimer's disease and related dementias. We adapted and developed a fast and easy-to-use automated surface-based pipeline to parcellate, for the first time, ultra high-resolution ex vivo brain tissue at the native subject space resolution using the Desikan-Killiany-Tourville (DKT) brain atlas. This allows us to perform vertex-wise analysis in the template space and thereby link morphometry measures with pathology measurements derived from histology. We will open-source our dataset docker container, Jupyter notebooks for ready-to-use out-of-the-box set of tools and command line options to advance ex vivo MRI clinical brain imaging research on the project webpage.
Authors: Ming Yan, Yan Zhang, Shuqiang Cai, Shuqi Fan, Xincheng Lin, Yudi Dai, Siqi Shen, Chenglu Wen, Lan Xu, Yuexin Ma, Cheng Wang
Abstract: Comprehensive capturing of human motions requires both accurate captures of complex poses and precise localization of the human within scenes. Most of the HPE datasets and methods primarily rely on RGB, LiDAR, or IMU data. However, solely using these modalities or a combination of them may not be adequate for HPE, particularly for complex and fast movements. For holistic human motion understanding, we present RELI11D, a high-quality multimodal human motion dataset involves LiDAR, IMU system, RGB camera, and Event camera. It records the motions of 10 actors performing 5 sports in 7 scenes, including 3.32 hours of synchronized LiDAR point clouds, IMU measurement data, RGB videos and Event steams. Through extensive experiments, we demonstrate that the RELI11D presents considerable challenges and opportunities as it contains many rapid and complex motions that require precise location. To address the challenge of integrating different modalities, we propose LEIR, a multimodal baseline that effectively utilizes LiDAR Point Cloud, Event stream, and RGB through our cross-attention fusion strategy. We show that LEIR exhibits promising results for rapid motions and daily motions and that utilizing the characteristics of multiple modalities can indeed improve HPE performance. Both the dataset and source code will be released publicly to the research community, fostering collaboration and enabling further exploration in this field.
Authors: Jie Wen, Zheng Zhang, Yong Xu, Bob Zhang, Lunke Fei, Guo-Sen Xie
Abstract: In recent years, incomplete multi-view clustering, which studies the challenging multi-view clustering problem on missing views, has received growing research interests. Although a series of methods have been proposed to address this issue, the following problems still exist: 1) Almost all of the existing methods are based on shallow models, which is difficult to obtain discriminative common representations. 2) These methods are generally sensitive to noise or outliers since the negative samples are treated equally as the important samples. In this paper, we propose a novel incomplete multi-view clustering network, called Cognitive Deep Incomplete Multi-view Clustering Network (CDIMC-net), to address these issues. Specifically, it captures the high-level features and local structure of each view by incorporating the view-specific deep encoders and graph embedding strategy into a framework. Moreover, based on the human cognition, i.e., learning from easy to hard, it introduces a self-paced strategy to select the most confident samples for model training, which can reduce the negative influence of outliers. Experimental results on several incomplete datasets show that CDIMC-net outperforms the state-of-the-art incomplete multi-view clustering methods.
Authors: Guangyu Wang, Jinzhi Zhang, Fan Wang, Ruqi Huang, Lu Fang
Abstract: We propose XScale-NVS for high-fidelity cross-scale novel view synthesis of real-world large-scale scenes. Existing representations based on explicit surface suffer from discretization resolution or UV distortion, while implicit volumetric representations lack scalability for large scenes due to the dispersed weight distribution and surface ambiguity. In light of the above challenges, we introduce hash featurized manifold, a novel hash-based featurization coupled with a deferred neural rendering framework. This approach fully unlocks the expressivity of the representation by explicitly concentrating the hash entries on the 2D manifold, thus effectively representing highly detailed contents independent of the discretization resolution. We also introduce a novel dataset, namely GigaNVS, to benchmark cross-scale, high-resolution novel view synthesis of realworld large-scale scenes. Our method significantly outperforms competing baselines on various real-world scenes, yielding an average LPIPS that is 40% lower than prior state-of-the-art on the challenging GigaNVS benchmark. Please see our project page at: xscalenvs.github.io.
Authors: Xiao Lin, Wenfei Yang, Yuan Gao, Tianzhu Zhang
Abstract: Category-level 6D object pose estimation aims to estimate the rotation, translation and size of unseen instances within specific categories. In this area, dense correspondence-based methods have achieved leading performance. However, they do not explicitly consider the local and global geometric information of different instances, resulting in poor generalization ability to unseen instances with significant shape variations. To deal with this problem, we propose a novel Instance-Adaptive and Geometric-Aware Keypoint Learning method for category-level 6D object pose estimation (AG-Pose), which includes two key designs: (1) The first design is an Instance-Adaptive Keypoint Detection module, which can adaptively detect a set of sparse keypoints for various instances to represent their geometric structures. (2) The second design is a Geometric-Aware Feature Aggregation module, which can efficiently integrate the local and global geometric information into keypoint features. These two modules can work together to establish robust keypoint-level correspondences for unseen instances, thus enhancing the generalization ability of the model.Experimental results on CAMERA25 and REAL275 datasets show that the proposed AG-Pose outperforms state-of-the-art methods by a large margin without category-specific shape priors.
Authors: Yulin Pan, Chaojie Mao, Zeyinzi Jiang, Zhen Han, Jingfeng Zhang
Abstract: Prior studies have made significant progress in image inpainting guided by either text or subject image. However, the research on editing with their combined guidance is still in the early stages. To tackle this challenge, we present LAR-Gen, a novel approach for image inpainting that enables seamless inpainting of masked scene images, incorporating both the textual prompts and specified subjects. Our approach adopts a coarse-to-fine manner to ensure subject identity preservation and local semantic coherence. The process involves (i) Locate: concatenating the noise with masked scene image to achieve precise regional editing, (ii) Assign: employing decoupled cross-attention mechanism to accommodate multi-modal guidance, and (iii) Refine: using a novel RefineNet to supplement subject details. Additionally, to address the issue of scarce training data, we introduce a novel data construction pipeline. This pipeline extracts substantial pairs of data consisting of local text prompts and corresponding visual instances from a vast image dataset, leveraging publicly available large models. Extensive experiments and varied application scenarios demonstrate the superiority of LAR-Gen in terms of both identity preservation and text semantic consistency. Project page can be found at \url{https://ali-vilab.github.io/largen-page/}.
Authors: Yuzheng Wang, Dingkang Yang, Zhaoyu Chen, Yang Liu, Siao Liu, Wenqiang Zhang, Lihua Zhang, Lizhe Qi
Abstract: Data-Free Knowledge Distillation (DFKD) is a promising task to train high-performance small models to enhance actual deployment without relying on the original training data. Existing methods commonly avoid relying on private data by utilizing synthetic or sampled data. However, a long-overlooked issue is that the severe distribution shifts between their substitution and original data, which manifests as huge differences in the quality of images and class proportions. The harmful shifts are essentially the confounder that significantly causes performance bottlenecks. To tackle the issue, this paper proposes a novel perspective with causal inference to disentangle the student models from the impact of such shifts. By designing a customized causal graph, we first reveal the causalities among the variables in the DFKD task. Subsequently, we propose a Knowledge Distillation Causal Intervention (KDCI) framework based on the backdoor adjustment to de-confound the confounder. KDCI can be flexibly combined with most existing state-of-the-art baselines. Experiments in combination with six representative DFKD methods demonstrate the effectiveness of our KDCI, which can obviously help existing methods under almost all settings, \textit{e.g.}, improving the baseline by up to 15.54\% accuracy on the CIFAR-100 dataset.
Authors: Ganlin Zhang, Erik Sandstr\"om, Youmin Zhang, Manthan Patel, Luc Van Gool, Martin R. Oswald
Abstract: Recent advancements in RGB-only dense Simultaneous Localization and Mapping (SLAM) have predominantly utilized grid-based neural implicit encodings and/or struggle to efficiently realize global map and pose consistency. To this end, we propose an efficient RGB-only dense SLAM system using a flexible neural point cloud scene representation that adapts to keyframe poses and depth updates, without needing costly backpropagation. Another critical challenge of RGB-only SLAM is the lack of geometric priors. To alleviate this issue, with the aid of a monocular depth estimator, we introduce a novel DSPO layer for bundle adjustment which optimizes the pose and depth of keyframes along with the scale of the monocular depth. Finally, our system benefits from loop closure and online global bundle adjustment and performs either better or competitive to existing dense neural RGB SLAM methods in tracking, mapping and rendering accuracy on the Replica, TUM-RGBD and ScanNet datasets. The source code will be made available.
Authors: R. Gnana Praveen, Jahangir Alam
Abstract: In video-based emotion recognition, audio and visual modalities are often expected to have a complementary relationship, which is widely explored using cross-attention. However, they may also exhibit weak complementary relationships, resulting in poor representations of audio-visual features, thus degrading the performance of the system. To address this issue, we propose Dynamic Cross-Attention (DCA) that can dynamically select cross-attended or unattended features on the fly based on their strong or weak complementary relationship with each other, respectively. Specifically, a simple yet efficient gating layer is designed to evaluate the contribution of the cross-attention mechanism and choose cross-attended features only when they exhibit a strong complementary relationship, otherwise unattended features. We evaluate the performance of the proposed approach on the challenging RECOLA and Aff-Wild2 datasets. We also compare the proposed approach with other variants of cross-attention and show that the proposed model consistently improves the performance on both datasets.
Authors: Ozgu Goksu, Nicolas Pugeault
Abstract: The pursuit of learning robust representations without human supervision is a longstanding challenge. The recent advancements in self-supervised contrastive learning approaches have demonstrated high performance across various representation learning challenges. However, current methods depend on the random transformation of training examples, resulting in some cases of unrepresentative positive pairs that can have a large impact on learning. This limitation not only impedes the convergence of the learning process but the robustness of the learnt representation as well as requiring larger batch sizes to improve robustness to such bad batches. This paper attempts to alleviate the influence of false positive and false negative pairs by employing pairwise similarity calculations through the Fr\'echet ResNet Distance (FRD), thereby obtaining robust representations from unlabelled data. The effectiveness of the proposed method is substantiated by empirical results, where a linear classifier trained on self-supervised contrastive representations achieved an impressive 87.74\% top-1 accuracy on STL10 and 99.31\% on the Flower102 dataset. These results emphasize the potential of the proposed approach in pushing the boundaries of the state-of-the-art in self-supervised contrastive learning, particularly for image classification tasks.
Authors: Zhenyu Wang, Yali Li, Taichi Liu, Hengshuang Zhao, Shengjin Wang
Abstract: In the current state of 3D object detection research, the severe scarcity of annotated 3D data, substantial disparities across different data modalities, and the absence of a unified architecture, have impeded the progress towards the goal of universality. In this paper, we propose \textbf{OV-Uni3DETR}, a unified open-vocabulary 3D detector via cycle-modality propagation. Compared with existing 3D detectors, OV-Uni3DETR offers distinct advantages: 1) Open-vocabulary 3D detection: During training, it leverages various accessible data, especially extensive 2D detection images, to boost training diversity. During inference, it can detect both seen and unseen classes. 2) Modality unifying: It seamlessly accommodates input data from any given modality, effectively addressing scenarios involving disparate modalities or missing sensor information, thereby supporting test-time modality switching. 3) Scene unifying: It provides a unified multi-modal model architecture for diverse scenes collected by distinct sensors. Specifically, we propose the cycle-modality propagation, aimed at propagating knowledge bridging 2D and 3D modalities, to support the aforementioned functionalities. 2D semantic knowledge from large-vocabulary learning guides novel class discovery in the 3D domain, and 3D geometric knowledge provides localization supervision for 2D detection images. OV-Uni3DETR achieves the state-of-the-art performance on various scenarios, surpassing existing methods by more than 6\% on average. Its performance using only RGB images is on par with or even surpasses that of previous point cloud based methods. Code and pre-trained models will be released later.
Authors: Zhongliang Zhou, Jielu Zhang, Zihan Guan, Mengxuan Hu, Ni Lao, Lan Mu, Sheng Li, Gengchen Mai
Abstract: Geolocating precise locations from images presents a challenging problem in computer vision and information retrieval.Traditional methods typically employ either classification, which dividing the Earth surface into grid cells and classifying images accordingly, or retrieval, which identifying locations by matching images with a database of image-location pairs. However, classification-based approaches are limited by the cell size and cannot yield precise predictions, while retrieval-based systems usually suffer from poor search quality and inadequate coverage of the global landscape at varied scale and aggregation levels. To overcome these drawbacks, we present Img2Loc, a novel system that redefines image geolocalization as a text generation task. This is achieved using cutting-edge large multi-modality models like GPT4V or LLaVA with retrieval augmented generation. Img2Loc first employs CLIP-based representations to generate an image-based coordinate query database. It then uniquely combines query results with images itself, forming elaborate prompts customized for LMMs. When tested on benchmark datasets such as Im2GPS3k and YFCC4k, Img2Loc not only surpasses the performance of previous state-of-the-art models but does so without any model training.
Authors: Shuai Zhang, Huangxuan Zhao, Zhenghong Zhou, Guanjun Wu, Chuansheng Zheng, Xinggang Wang, Wenyu Liu
Abstract: Four-dimensional Digital Subtraction Angiography (4D DSA) is a medical imaging technique that provides a series of 2D images captured at different stages and angles during the process of contrast agent filling blood vessels. It plays a significant role in the diagnosis of cerebrovascular diseases. Improving the rendering quality and speed under sparse sampling is important for observing the status and location of lesions. The current methods exhibit inadequate rendering quality in sparse views and suffer from slow rendering speed. To overcome these limitations, we propose TOGS, a Gaussian splatting method with opacity offset over time, which can effectively improve the rendering quality and speed of 4D DSA. We introduce an opacity offset table for each Gaussian to model the temporal variations in the radiance of the contrast agent. By interpolating the opacity offset table, the opacity variation of the Gaussian at different time points can be determined. This enables us to render the 2D DSA image at that specific moment. Additionally, we introduced a Smooth loss term in the loss function to mitigate overfitting issues that may arise in the model when dealing with sparse view scenarios. During the training phase, we randomly prune Gaussians, thereby reducing the storage overhead of the model. The experimental results demonstrate that compared to previous methods, this model achieves state-of-the-art reconstruction quality under the same number of training views. Additionally, it enables real-time rendering while maintaining low storage overhead. The code will be publicly available.
Authors: Donghyun Kim, Byeongho Heo, Dongyoon Han
Abstract: This paper revives Densely Connected Convolutional Networks (DenseNets) and reveals the underrated effectiveness over predominant ResNet-style architectures. We believe DenseNets' potential was overlooked due to untouched training methods and traditional design elements not fully revealing their capabilities. Our pilot study shows dense connections through concatenation are strong, demonstrating that DenseNets can be revitalized to compete with modern architectures. We methodically refine suboptimal components - architectural adjustments, block redesign, and improved training recipes towards widening DenseNets and boosting memory efficiency while keeping concatenation shortcuts. Our models, employing simple architectural elements, ultimately surpass Swin Transformer, ConvNeXt, and DeiT-III - key architectures in the residual learning lineage. Furthermore, our models exhibit near state-of-the-art performance on ImageNet-1K, competing with the very recent models and downstream tasks, ADE20k semantic segmentation, and COCO object detection/instance segmentation. Finally, we provide empirical analyses that uncover the merits of the concatenation over additive shortcuts, steering a renewed preference towards DenseNet-style designs. Our code is available at https://github.com/naver-ai/rdnet.
Authors: Bu Jin, Yupeng Zheng, Pengfei Li, Weize Li, Yuhang Zheng, Sujie Hu, Xinyu Liu, Jinwei Zhu, Zhijie Yan, Haiyang Sun, Kun Zhan, Peng Jia, Xiaoxiao Long, Yilun Chen, Hao Zhao
Abstract: 3D dense captioning stands as a cornerstone in achieving a comprehensive understanding of 3D scenes through natural language. It has recently witnessed remarkable achievements, particularly in indoor settings. However, the exploration of 3D dense captioning in outdoor scenes is hindered by two major challenges: 1) the \textbf{domain gap} between indoor and outdoor scenes, such as dynamics and sparse visual inputs, makes it difficult to directly adapt existing indoor methods; 2) the \textbf{lack of data} with comprehensive box-caption pair annotations specifically tailored for outdoor scenes. To this end, we introduce the new task of outdoor 3D dense captioning. As input, we assume a LiDAR point cloud and a set of RGB images captured by the panoramic camera rig. The expected output is a set of object boxes with captions. To tackle this task, we propose the TOD3Cap network, which leverages the BEV representation to generate object box proposals and integrates Relation Q-Former with LLaMA-Adapter to generate rich captions for these objects. We also introduce the TOD3Cap dataset, the largest one to our knowledge for 3D dense captioning in outdoor scenes, which contains 2.3M descriptions of 64.3K outdoor objects from 850 scenes. Notably, our TOD3Cap network can effectively localize and caption 3D objects in outdoor scenes, which outperforms baseline methods by a significant margin (+9.6 CiDEr@0.5IoU). Code, data, and models are publicly available at https://github.com/jxbbb/TOD3Cap.
Authors: Aimon Rahman, Malsha V. Perera, Vishal M. Patel
Abstract: Building on the momentum of image generation diffusion models, there is an increasing interest in video-based diffusion models. However, video generation poses greater challenges due to its higher-dimensional nature, the scarcity of training data, and the complex spatiotemporal relationships involved. Image generation models, due to their extensive data requirements, have already strained computational resources to their limits. There have been instances of these models reproducing elements from the training samples, leading to concerns and even legal disputes over sample replication. Video diffusion models, which operate with even more constrained datasets and are tasked with generating both spatial and temporal content, may be more prone to replicating samples from their training sets. Compounding the issue, these models are often evaluated using metrics that inadvertently reward replication. In our paper, we present a systematic investigation into the phenomenon of sample replication in video diffusion models. We scrutinize various recent diffusion models for video synthesis, assessing their tendency to replicate spatial and temporal content in both unconditional and conditional generation scenarios. Our study identifies strategies that are less likely to lead to replication. Furthermore, we propose new evaluation strategies that take replication into account, offering a more accurate measure of a model's ability to generate the original content.
Authors: Johann Haselberger, Bonifaz Stuhr, Bernhard Schick, Steffen M\"uller
Abstract: There is evidence that the driving style of an autonomous vehicle is important to increase the acceptance and trust of the passengers. The driving situation has been found to have a significant influence on human driving behavior. However, current driving style models only partially incorporate driving environment information, limiting the alignment between an agent and the given situation. Therefore, we propose a situation-aware driving style model based on different visual feature encoders pretrained on fleet data, as well as driving behavior predictors, which are adapted to the driving style of a specific driver. Our experiments show that the proposed method outperforms static driving styles significantly and forms plausible situation clusters. Furthermore, we found that feature encoders pretrained on our dataset lead to more precise driving behavior modeling. In contrast, feature encoders pretrained supervised and unsupervised on different data sources lead to more specific situation clusters, which can be utilized to constrain and control the driving style adaptation for specific situations. Moreover, in a real-world setting, where driving style adaptation is happening iteratively, we found the MLP-based behavior predictors achieve good performance initially but suffer from catastrophic forgetting. In contrast, behavior predictors based on situationdependent statistics can learn iteratively from continuous data streams by design. Overall, our experiments show that important information for driving behavior prediction is contained within the visual feature encoder. The dataset is publicly available at huggingface.co/datasets/jHaselberger/SADC-Situation-Awareness-for-Driver-Centric-Driving-Style-Adaptation.
Authors: Bo Wan, Michael Tschannen, Yongqin Xian, Filip Pavetic, Ibrahim Alabdulmohsin, Xiao Wang, Andr\'e Susano Pinto, Andreas Steiner, Lucas Beyer, Xiaohua Zhai
Abstract: Image captioning has been shown as an effective pretraining method similar to contrastive pretraining. However, the incorporation of location-aware information into visual pretraining remains an area with limited research. In this paper, we propose a simple visual pretraining method with location-aware captioners (LocCa). LocCa uses a simple image captioner task interface, to teach a model to read out rich information, i.e. bounding box coordinates, and captions, conditioned on the image pixel input. Thanks to the multitask capabilities of an encoder-decoder architecture, we show that an image captioner can easily handle multiple tasks during pretraining. Our experiments demonstrate that LocCa outperforms standard captioners significantly on localization downstream tasks while maintaining comparable performance on holistic tasks.
Authors: Zhicai Wang, Longhui Wei, Tan Wang, Heyu Chen, Yanbin Hao, Xiang Wang, Xiangnan He, Qi Tian
Abstract: Text-to-image (T2I) generative models have recently emerged as a powerful tool, enabling the creation of photo-realistic images and giving rise to a multitude of applications. However, the effective integration of T2I models into fundamental image classification tasks remains an open question. A prevalent strategy to bolster image classification performance is through augmenting the training set with synthetic images generated by T2I models. In this study, we scrutinize the shortcomings of both current generative and conventional data augmentation techniques. Our analysis reveals that these methods struggle to produce images that are both faithful (in terms of foreground objects) and diverse (in terms of background contexts) for domain-specific concepts. To tackle this challenge, we introduce an innovative inter-class data augmentation method known as Diff-Mix (https://github.com/Zhicaiwww/Diff-Mix), which enriches the dataset by performing image translations between classes. Our empirical results demonstrate that Diff-Mix achieves a better balance between faithfulness and diversity, leading to a marked improvement in performance across diverse image classification scenarios, including few-shot, conventional, and long-tail classifications for domain-specific datasets.
Authors: Evan Matthews, Nicolas Prate
Abstract: Given a set of ordered pixel data in the form of an image, our goal is to perform upsampling on the data such that: the resulting resolution is improved by some factor, the final result passes the human test, having added new, believable, and realistic information and detail to the image, the time complexity for upscaling is relatively close to that of lossy upscaling implementations.
Authors: Dmitrii Zhemchuzhnikov, Sergei Grudinin
Abstract: Effective recognition of spatial patterns and learning their hierarchy is crucial in modern spatial data analysis. Volumetric data applications seek techniques ensuring invariance not only to shifts but also to pattern rotations. While traditional methods can readily achieve translational invariance, rotational invariance possesses multiple challenges and remains an active area of research. Here, we present ILPO-Net (Invariant to Local Patterns Orientation Network), a novel approach that handles arbitrarily shaped patterns with the convolutional operation inherently invariant to local spatial pattern orientations using the Wigner matrix expansions. Our architecture seamlessly integrates the new convolution operator and, when benchmarked on diverse volumetric datasets such as MedMNIST and CATH, demonstrates superior performance over the baselines with significantly reduced parameter counts - up to 1000 times fewer in the case of MedMNIST. Beyond these demonstrations, ILPO-Net's rotational invariance paves the way for other applications across multiple disciplines. Our code is publicly available at https://gricad-gitlab.univ-grenoble-alpes.fr/GruLab/ILPONet.
URLs: https://gricad-gitlab.univ-grenoble-alpes.fr/GruLab/ILPONet.
Authors: Xiaowei Song, Jv Zheng, Shiran Yuan, Huan-ang Gao, Jingwei Zhao, Xiang He, Weihao Gu, Hao Zhao
Abstract: In this paper, we present a Scale-adaptive method for Anti-aliasing Gaussian Splatting (SA-GS). While the state-of-the-art method Mip-Splatting needs modifying the training procedure of Gaussian splatting, our method functions at test-time and is training-free. Specifically, SA-GS can be applied to any pretrained Gaussian splatting field as a plugin to significantly improve the field's anti-alising performance. The core technique is to apply 2D scale-adaptive filters to each Gaussian during test time. As pointed out by Mip-Splatting, observing Gaussians at different frequencies leads to mismatches between the Gaussian scales during training and testing. Mip-Splatting resolves this issue using 3D smoothing and 2D Mip filters, which are unfortunately not aware of testing frequency. In this work, we show that a 2D scale-adaptive filter that is informed of testing frequency can effectively match the Gaussian scale, thus making the Gaussian primitive distribution remain consistent across different testing frequencies. When scale inconsistency is eliminated, sampling rates smaller than the scene frequency result in conventional jaggedness, and we propose to integrate the projected 2D Gaussian within each pixel during testing. This integration is actually a limiting case of super-sampling, which significantly improves anti-aliasing performance over vanilla Gaussian Splatting. Through extensive experiments using various settings and both bounded and unbounded scenes, we show SA-GS performs comparably with or better than Mip-Splatting. Note that super-sampling and integration are only effective when our scale-adaptive filtering is activated. Our codes, data and models are available at https://github.com/zsy1987/SA-GS.
Authors: Chongjie Ye, Yinyu Nie, Jiahao Chang, Yuantao Chen, Yihao Zhi, Xiaoguang Han
Abstract: We present GauStudio, a novel modular framework for modeling 3D Gaussian Splatting (3DGS) to provide standardized, plug-and-play components for users to easily customize and implement a 3DGS pipeline. Supported by our framework, we propose a hybrid Gaussian representation with foreground and skyball background models. Experiments demonstrate this representation reduces artifacts in unbounded outdoor scenes and improves novel view synthesis. Finally, we propose Gaussian Splatting Surface Reconstruction (GauS), a novel render-then-fuse approach for high-fidelity mesh reconstruction from 3DGS inputs without fine-tuning. Overall, our GauStudio framework, hybrid representation, and GauS approach enhance 3DGS modeling and rendering capabilities, enabling higher-quality novel view synthesis and surface reconstruction.
Authors: Yan-Bo Lin, Gedas Bertasius
Abstract: Traditional audio-visual methods rely on independent audio and visual backbones, which is costly and not scalable. In this work, we investigate using an audio-visual siamese network (AVSiam) for efficient and scalable audio-visual pretraining. Our framework uses a single shared vision transformer backbone to process audio and visual inputs, improving its parameter efficiency, reducing the GPU memory footprint, and allowing us to scale our method to larger datasets and model sizes. We pretrain our model using a contrastive audio-visual matching objective with a multi-ratio random masking scheme, which enables our model to process larger audio-visual instance batches, helpful for contrastive learning. Unlike prior audio-visual methods, our method can robustly handle audio, visual, and audio-visual inputs with a single shared ViT backbone. Furthermore, despite using the shared backbone for both modalities, AVSiam achieves competitive or even better results than prior methods on AudioSet and VGGSound for audio-visual classification and retrieval. Our code is available at https://github.com/GenjiB/AVSiam
Authors: Yusuf Dalva, Hidir Yesiltepe, Pinar Yanardag
Abstract: The rapid advancement in image generation models has predominantly been driven by diffusion models, which have demonstrated unparalleled success in generating high-fidelity, diverse images from textual prompts. Despite their success, diffusion models encounter substantial challenges in the domain of image editing, particularly in executing disentangled edits-changes that target specific attributes of an image while leaving irrelevant parts untouched. In contrast, Generative Adversarial Networks (GANs) have been recognized for their success in disentangled edits through their interpretable latent spaces. We introduce GANTASTIC, a novel framework that takes existing directions from pre-trained GAN models-representative of specific, controllable attributes-and transfers these directions into diffusion-based models. This novel approach not only maintains the generative quality and diversity that diffusion models are known for but also significantly enhances their capability to perform precise, targeted image edits, thereby leveraging the best of both worlds.
Authors: Chenyang Liu, Keyan Chen, Haotian Zhang, Zipeng Qi, Zhengxia Zou, Zhenwei Shi
Abstract: Monitoring changes in the Earth's surface is crucial for understanding natural processes and human impacts, necessitating precise and comprehensive interpretation methodologies. Remote sensing satellite imagery offers a unique perspective for monitoring these changes, leading to the emergence of remote sensing image change interpretation (RSICI) as a significant research focus. Current RSICI technology encompasses change detection and change captioning, each with its limitations in providing comprehensive interpretation. To address this, we propose an interactive Change-Agent which integrates a multi-level change interpretation (MCI) model as eyes and a large language model (LLM) as the brain. Our Change-Agent can follow user instructions to achieve comprehensive change interpretation and insightful analysis according to user instructions, such as change detection and change captioning, change object counting, change cause analysis, etc. Our proposed MCI model contains two branches of pixel-level change detection and semantic-level change captioning, in which multiple BI-temporal Iterative Interaction (BI3) layers utilize Local Perception Enhancement (LPE) and the Global Difference Fusion Attention (GDFA) modules to enhance the model's discriminative feature representation capabilities. To train the MCI model, we build the LEVIR-MCI dataset with change masks and captions of bi-temporal images. Extensive experiments demonstrate the effectiveness of the proposed change interpretation model and highlight the promising potential of our Change-Agent in facilitating comprehensive and intelligent interpretation of surface changes. We will make our dataset and codebase of the change interpretation model and Change-Agent publicly available to facilitate future research at https://github.com/Chen-Yang-Liu/Change-Agent
Authors: Kai Zhang, Yi Luan, Hexiang Hu, Kenton Lee, Siyuan Qiao, Wenhu Chen, Yu Su, Ming-Wei Chang
Abstract: Image retrieval, i.e., finding desired images given a reference image, inherently encompasses rich, multi-faceted search intents that are difficult to capture solely using image-based measures. Recent work leverages text instructions to allow users to more freely express their search intents. However, existing work primarily focuses on image pairs that are visually similar and/or can be characterized by a small set of pre-defined relations. The core thesis of this paper is that text instructions can enable retrieving images with richer relations beyond visual similarity. To show this, we introduce MagicLens, a series of self-supervised image retrieval models that support open-ended instructions. MagicLens is built on a key novel insight: image pairs that naturally occur on the same web pages contain a wide range of implicit relations (e.g., inside view of), and we can bring those implicit relations explicit by synthesizing instructions via large multimodal models (LMMs) and large language models (LLMs). Trained on 36.7M (query image, instruction, target image) triplets with rich semantic relations mined from the web, MagicLens achieves comparable or better results on eight benchmarks of various image retrieval tasks than prior state-of-the-art (SOTA) methods. Remarkably, it outperforms previous SOTA but with a 50X smaller model size on multiple benchmarks. Additional human analyses on a 1.4M-image unseen corpus further demonstrate the diversity of search intents supported by MagicLens.
Authors: Sirui Xu, Ziyin Wang, Yu-Xiong Wang, Liang-Yan Gui
Abstract: Text-conditioned human motion generation has experienced significant advancements with diffusion models trained on extensive motion capture data and corresponding textual annotations. However, extending such success to 3D dynamic human-object interaction (HOI) generation faces notable challenges, primarily due to the lack of large-scale interaction data and comprehensive descriptions that align with these interactions. This paper takes the initiative and showcases the potential of generating human-object interactions without direct training on text-interaction pair data. Our key insight in achieving this is that interaction semantics and dynamics can be decoupled. Being unable to learn interaction semantics through supervised training, we instead leverage pre-trained large models, synergizing knowledge from a large language model and a text-to-motion model. While such knowledge offers high-level control over interaction semantics, it cannot grasp the intricacies of low-level interaction dynamics. To overcome this issue, we further introduce a world model designed to comprehend simple physics, modeling how human actions influence object motion. By integrating these components, our novel framework, InterDreamer, is able to generate text-aligned 3D HOI sequences in a zero-shot manner. We apply InterDreamer to the BEHAVE and CHAIRS datasets, and our comprehensive experimental analysis demonstrates its capability to generate realistic and coherent interaction sequences that seamlessly align with the text directives.
Authors: Katherine Xu, Lingzhi Zhang, Jianbo Shi
Abstract: Modern text-to-image (T2I) diffusion models can generate images with remarkable realism and creativity. These advancements have sparked research in fake image detection and attribution, yet prior studies have not fully explored the practical and scientific dimensions of this task. In addition to attributing images to 12 state-of-the-art T2I generators, we provide extensive analyses on what inference stage hyperparameters and image modifications are discernible. Our experiments reveal that initialization seeds are highly detectable, along with other subtle variations in the image generation process to some extent. We further investigate what visual traces are leveraged in image attribution by perturbing high-frequency details and employing mid-level representations of image style and structure. Notably, altering high-frequency information causes only slight reductions in accuracy, and training an attributor on style representations outperforms training on RGB images. Our analyses underscore that fake images are detectable and attributable at various levels of visual granularity than previously explored.
Authors: Keyan Chen, Bowen Chen, Chenyang Liu, Wenyuan Li, Zhengxia Zou, Zhenwei Shi
Abstract: Remote sensing image classification forms the foundation of various understanding tasks, serving a crucial function in remote sensing image interpretation. The recent advancements of Convolutional Neural Networks (CNNs) and Transformers have markedly enhanced classification accuracy. Nonetheless, remote sensing scene classification remains a significant challenge, especially given the complexity and diversity of remote sensing scenarios and the variability of spatiotemporal resolutions. The capacity for whole-image understanding can provide more precise semantic cues for scene discrimination. In this paper, we introduce RSMamba, a novel architecture for remote sensing image classification. RSMamba is based on the State Space Model (SSM) and incorporates an efficient, hardware-aware design known as the Mamba. It integrates the advantages of both a global receptive field and linear modeling complexity. To overcome the limitation of the vanilla Mamba, which can only model causal sequences and is not adaptable to two-dimensional image data, we propose a dynamic multi-path activation mechanism to augment Mamba's capacity to model non-causal data. Notably, RSMamba maintains the inherent modeling mechanism of the vanilla Mamba, yet exhibits superior performance across multiple remote sensing image classification datasets. This indicates that RSMamba holds significant potential to function as the backbone of future visual foundation models. The code will be available at \url{https://github.com/KyanChen/RSMamba}.
Authors: Bowen Zhang, Yiji Cheng, Jiaolong Yang, Chunyu Wang, Feng Zhao, Yansong Tang, Dong Chen, Baining Guo
Abstract: 3D Gaussian Splatting (GS) have achieved considerable improvement over Neural Radiance Fields in terms of 3D fitting fidelity and rendering speed. However, this unstructured representation with scattered Gaussians poses a significant challenge for generative modeling. To address the problem, we introduce GaussianCube, a structured GS representation that is both powerful and efficient for generative modeling. We achieve this by first proposing a modified densification-constrained GS fitting algorithm which can yield high-quality fitting results using a fixed number of free Gaussians, and then re-arranging the Gaussians into a predefined voxel grid via Optimal Transport. The structured grid representation allows us to use standard 3D U-Net as our backbone in diffusion generative modeling without elaborate designs. Extensive experiments conducted on ShapeNet and OmniObject3D show that our model achieves state-of-the-art generation results both qualitatively and quantitatively, underscoring the potential of GaussianCube as a powerful and versatile 3D representation.
Authors: Cynthia Maldonado-Garcia, Rodrigo Bonazzola, Enzo Ferrante, Thomas H Julian, Panagiotis I Sergouniotis, Nishant Ravikumara, Alejandro F Frangi
Abstract: We investigated the potential of optical coherence tomography (OCT) as an additional imaging technique to predict future cardiovascular disease (CVD). We utilised a self-supervised deep learning approach based on Variational Autoencoders (VAE) to learn low-dimensional representations of high-dimensional 3D OCT images and to capture distinct characteristics of different retinal layers within the OCT image. A Random Forest (RF) classifier was subsequently trained using the learned latent features and participant demographic and clinical data, to differentiate between patients at risk of CVD events (MI or stroke) and non-CVD cases. Our predictive model, trained on multimodal data, was assessed based on its ability to correctly identify individuals likely to suffer from a CVD event(MI or stroke), within a 5-year interval after image acquisition. Our self-supervised VAE feature selection and multimodal Random Forest classifier differentiate between patients at risk of future CVD events and the control group with an AUC of 0.75, outperforming the clinically established QRISK3 score (AUC= 0.597). The choroidal layer visible in OCT images was identified as an important predictor of future CVD events using a novel approach to model explanability. Retinal OCT imaging provides a cost-effective and non-invasive alternative to predict the risk of cardiovascular disease and is readily accessible in optometry practices and hospitals.
Authors: Huiyi Wang, Haodong Lu, Lina Yao, Dong Gong
Abstract: Continual learning aims to learn from a stream of continuously arriving data with minimum forgetting of previously learned knowledge. While previous works have explored the effectiveness of leveraging the generalizable knowledge from pre-trained models in continual learning, existing parameter-efficient fine-tuning approaches focus on the use of a predetermined or task-wise set of adapters or prompts. However, these approaches still suffer from forgetting due to task interference on jointly used parameters or restricted flexibility. The reliance on a static model architecture may lead to the allocation of excessive parameters that are not essential or, conversely, inadequate adaptation for downstream tasks, given that the scale and distribution of incoming data are unpredictable in continual learning. We propose Self-Expansion of pre-trained models with Modularized Adaptation (SEMA), a novel fine-tuning approach which automatically decides to reuse or add adapter modules on demand in continual learning, depending on whether drastic distribution shift that could not be handled by existing modules is detected at different representation levels. We design each adapter module to consist of an adapter and a representation descriptor, specifically, implemented as an autoencoder. The representation descriptor functions as a distributional shift indicator during training and triggers adapter expansion. For better usage of the adapters, an expandable weighting router is learned jointly for mixture of adapter outputs. By comparing with vision-transformer-based continual learning adaptation methods, we demonstrate that the proposed framework outperforms the state-of-the-art without memory rehearsal.
Authors: Hamidreza Kamkari, Brendan Leigh Ross, Jesse C. Cresswell, Anthony L. Caterini, Rahul G. Krishnan, Gabriel Loaiza-Ganem
Abstract: Likelihood-based deep generative models (DGMs) commonly exhibit a puzzling behaviour: when trained on a relatively complex dataset, they assign higher likelihood values to out-of-distribution (OOD) data from simpler sources. Adding to the mystery, OOD samples are never generated by these DGMs despite having higher likelihoods. This two-pronged paradox has yet to be conclusively explained, making likelihood-based OOD detection unreliable. Our primary observation is that high-likelihood regions will not be generated if they contain minimal probability mass. We demonstrate how this seeming contradiction of large densities yet low probability mass can occur around data confined to low-dimensional manifolds. We also show that this scenario can be identified through local intrinsic dimension (LID) estimation, and propose a method for OOD detection which pairs the likelihoods and LID estimates obtained from a pre-trained DGM. Our method can be applied to normalizing flows and score-based diffusion models, and obtains results which match or surpass state-of-the-art OOD detection benchmarks using the same DGM backbones. Our code is available at https://github.com/layer6ai-labs/dgm_ood_detection.
Authors: Aditya Golatkar, Alessandro Achille, Luca Zancato, Yu-Xiang Wang, Ashwin Swaminathan, Stefano Soatto
Abstract: Retrieval Augmented Generation (RAG) is emerging as a flexible and robust technique to adapt models to private users data without training, to handle credit attribution, and to allow efficient machine unlearning at scale. However, RAG techniques for image generation may lead to parts of the retrieved samples being copied in the model's output. To reduce risks of leaking private information contained in the retrieved set, we introduce Copy-Protected generation with Retrieval (CPR), a new method for RAG with strong copyright protection guarantees in a mixed-private setting for diffusion models.CPR allows to condition the output of diffusion models on a set of retrieved images, while also guaranteeing that unique identifiable information about those example is not exposed in the generated outputs. In particular, it does so by sampling from a mixture of public (safe) distribution and private (user) distribution by merging their diffusion scores at inference. We prove that CPR satisfies Near Access Freeness (NAF) which bounds the amount of information an attacker may be able to extract from the generated images. We provide two algorithms for copyright protection, CPR-KL and CPR-Choose. Unlike previously proposed rejection-sampling-based NAF methods, our methods enable efficient copyright-protected sampling with a single run of backward diffusion. We show that our method can be applied to any pre-trained conditional diffusion model, such as Stable Diffusion or unCLIP. In particular, we empirically show that applying CPR on top of unCLIP improves quality and text-to-image alignment of the generated results (81.4 to 83.17 on TIFA benchmark), while enabling credit attribution, copy-right protection, and deterministic, constant time, unlearning.
Authors: Petros Toupas, Zhewen Yu, Christos-Savvas Bouganis, Dimitrios Tzovaras
Abstract: Convolutional Neural Networks (CNNs) have demonstrated their effectiveness in numerous vision tasks. However, their high processing requirements necessitate efficient hardware acceleration to meet the application's performance targets. In the space of FPGAs, streaming-based dataflow architectures are often adopted by users, as significant performance gains can be achieved through layer-wise pipelining and reduced off-chip memory access by retaining data on-chip. However, modern topologies, such as the UNet, YOLO, and X3D models, utilise long skip connections, requiring significant on-chip storage and thus limiting the performance achieved by such system architectures. The paper addresses the above limitation by introducing weight and activation eviction mechanisms to off-chip memory along the computational pipeline, taking into account the available compute and memory resources. The proposed mechanism is incorporated into an existing toolflow, expanding the design space by utilising off-chip memory as a buffer. This enables the mapping of such modern CNNs to devices with limited on-chip memory, under the streaming architecture design approach. SMOF has demonstrated the capacity to deliver competitive and, in some cases, state-of-the-art performance across a spectrum of computer vision tasks, achieving up to 10.65 X throughput improvement compared to previous works.
Authors: Xun Wang, John Rachwan, Stephan G\"unnemann, Bertrand Charpentier
Abstract: Neural network pruning serves as a critical technique for enhancing the efficiency of deep learning models. Unlike unstructured pruning, which only sets specific parameters to zero, structured pruning eliminates entire channels, thus yielding direct computational and storage benefits. However, the diverse patterns for coupling parameters, such as residual connections and group convolutions, the diverse deep learning frameworks, and the various time stages at which pruning can be performed make existing pruning methods less adaptable to different architectures, frameworks, and pruning criteria. To address this, we introduce Structurally Prune Anything (SPA), a versatile structured pruning framework that can prune neural networks with any architecture, from any framework, and at any stage of training. SPA leverages a standardized computational graph and ONNX representation to prune diverse neural network architectures without the need for manual intervention. SPA employs a group-level importance estimation method, which groups dependent computational operators, estimates their importance, and prunes unimportant coupled channels. This enables the transfer of various existing pruning criteria into a structured group style. As a result, SPA supports pruning at any time, either before training, after training with fine-tuning, or after training without fine-tuning. In the context of the latter, we introduce Optimal Brain SPA (OBSPA), an algorithm that achieves state-of-the-art pruning results needing neither fine-tuning nor calibration data. In extensive experiments, SPA shows competitive to state-of-the-art pruning performance across various architectures, from popular frameworks, at different pruning times.
Authors: Soumyendu Sarkar, Ashwin Ramesh Babu, Sajad Mousavi, Vineet Gundecha, Avisek Naug, Sahand Ghorbanpour
Abstract: We present a generic Reinforcement Learning (RL) framework optimized for crafting adversarial attacks on different model types spanning from ECG signal analysis (1D), image classification (2D), and video classification (3D). The framework focuses on identifying sensitive regions and inducing misclassifications with minimal distortions and various distortion types. The novel RL method outperforms state-of-the-art methods for all three applications, proving its efficiency. Our RL approach produces superior localization masks, enhancing interpretability for image classification and ECG analysis models. For applications such as ECG analysis, our platform highlights critical ECG segments for clinicians while ensuring resilience against prevalent distortions. This comprehensive tool aims to bolster both resilience with adversarial training and transparency across varied applications and data types.
Authors: Siva Sai Nagender Vasireddy, Chenxu Zhang, Xiaohu Guo, Yapeng Tian
Abstract: This paper addresses the issue of active speaker detection (ASD) in noisy environments and formulates a robust active speaker detection (rASD) problem. Existing ASD approaches leverage both audio and visual modalities, but non-speech sounds in the surrounding environment can negatively impact performance. To overcome this, we propose a novel framework that utilizes audio-visual speech separation as guidance to learn noise-free audio features. These features are then utilized in an ASD model, and both tasks are jointly optimized in an end-to-end framework. Our proposed framework mitigates residual noise and audio quality reduction issues that can occur in a naive cascaded two-stage framework that directly uses separated speech for ASD, and enables the two tasks to be optimized simultaneously. To further enhance the robustness of the audio features and handle inherent speech noises, we propose a dynamic weighted loss approach to train the speech separator. We also collected a real-world noise audio dataset to facilitate investigations. Experiments demonstrate that non-speech audio noises significantly impact ASD models, and our proposed approach improves ASD performance in noisy environments. The framework is general and can be applied to different ASD approaches to improve their robustness. Our code, models, and data will be released.
Authors: Ji Lin, Ligeng Zhu, Wei-Ming Chen, Wei-Chen Wang, Song Han
Abstract: Tiny Machine Learning (TinyML) is a new frontier of machine learning. By squeezing deep learning models into billions of IoT devices and microcontrollers (MCUs), we expand the scope of AI applications and enable ubiquitous intelligence. However, TinyML is challenging due to hardware constraints: the tiny memory resource makes it difficult to hold deep learning models designed for cloud and mobile platforms. There is also limited compiler and inference engine support for bare-metal devices. Therefore, we need to co-design the algorithm and system stack to enable TinyML. In this review, we will first discuss the definition, challenges, and applications of TinyML. We then survey the recent progress in TinyML and deep learning on MCUs. Next, we will introduce MCUNet, showing how we can achieve ImageNet-scale AI applications on IoT devices with system-algorithm co-design. We will further extend the solution from inference to training and introduce tiny on-device training techniques. Finally, we present future directions in this area. Today's large model might be tomorrow's tiny model. The scope of TinyML should evolve and adapt over time.
Authors: Chenshuang Zhang, Chaoning Zhang, Kang Zhang, Axi Niu, Junmo Kim, In So Kweon
Abstract: There is a growing concern about applying batch normalization (BN) in adversarial training (AT), especially when the model is trained on both adversarial samples and clean samples (termed Hybrid-AT). With the assumption that adversarial and clean samples are from two different domains, a common practice in prior works is to adopt Dual BN, where BN and BN are used for adversarial and clean branches, respectively. A popular belief for motivating Dual BN is that estimating normalization statistics of this mixture distribution is challenging and thus disentangling it for normalization achieves stronger robustness. In contrast to this belief, we reveal that disentangling statistics plays a less role than disentangling affine parameters in model training. This finding aligns with prior work (Rebuffi et al., 2023), and we build upon their research for further investigations. We demonstrate that the domain gap between adversarial and clean samples is not very large, which is counter-intuitive considering the significant influence of adversarial perturbation on the model accuracy. We further propose a two-task hypothesis which serves as the empirical foundation and a unified framework for Hybrid-AT improvement. We also investigate Dual BN in test-time and reveal that affine parameters characterize the robustness during inference. Overall, our work sheds new light on understanding the mechanism of Dual BN in Hybrid-AT and its underlying justification.
Authors: Cameron Gordon, Lachlan Ewen MacDonald, Hemanth Saratchandran, Simon Lucey
Abstract: Deep implicit functions have been found to be an effective tool for efficiently encoding all manner of natural signals. Their attractiveness stems from their ability to compactly represent signals with little to no off-line training data. Instead, they leverage the implicit bias of deep networks to decouple hidden redundancies within the signal. In this paper, we explore the hypothesis that additional compression can be achieved by leveraging the redundancies that exist between layers. We propose to use a novel run-time decoder-only hypernetwork - that uses no offline training data - to better model this cross-layer parameter redundancy. Previous applications of hyper-networks with deep implicit functions have applied feed-forward encoder/decoder frameworks that rely on large offline datasets that do not generalize beyond the signals they were trained on. We instead present a strategy for the initialization of run-time deep implicit functions for single-instance signals through a Decoder-Only randomly projected Hypernetwork (D'OH). By directly changing the dimension of a latent code to approximate a target implicit neural architecture, we provide a natural way to vary the memory footprint of neural representations without the costly need for neural architecture search on a space of alternative low-rate structures.
Authors: Louie S{\o}s Meyer, Johanne Engel Aaen, Anitamalina Regitse Tranberg, Peter Kun, Matthias Freiberger, Sebastian Risi, Anders Sundnes L{\o}vlie
Abstract: This Research through Design paper explores how object detection may be applied to a large digital art museum collection to facilitate new ways of encountering and experiencing art. We present the design and evaluation of an interactive application called SMKExplore, which allows users to explore a museum's digital collection of paintings by browsing through objects detected in the images, as a novel form of open-ended exploration. We provide three contributions. First, we show how an object detection pipeline can be integrated into a design process for visual exploration. Second, we present the design and development of an app that enables exploration of an art museum's collection. Third, we offer reflections on future possibilities for museums and HCI researchers to incorporate object detection techniques into the digitalization of museums.
Authors: Peng Tang, Tobias Lasser
Abstract: In this study, we introduce a multi-modal approach that efficiently integrates multi-scale clinical and dermoscopy features within a single network, thereby substantially reducing model parameters. The proposed method includes three novel fusion schemes. Firstly, unlike current methods that usually employ two individual models for for clinical and dermoscopy modalities, we verified that multimodal feature can be learned by sharing the parameters of encoder while leaving the individual modal-specific classifiers. Secondly, the shared cross-attention module can replace the individual one to efficiently interact between two modalities at multiple layers. Thirdly, different from current methods that equally optimize dermoscopy and clinical branches, inspired by prior knowledge that dermoscopy images play a more significant role than clinical images, we propose a novel biased loss. This loss guides the single-shared network to prioritize dermoscopy information over clinical information, implicitly learning a better joint feature representation for the modal-specific task. Extensive experiments on a well-recognized Seven-Point Checklist (SPC) dataset and a collected dataset demonstrate the effectiveness of our method on both CNN and Transformer structures. Furthermore, our method exhibits superiority in both accuracy and model parameters compared to currently advanced methods.
Authors: Yiping Ji, Hemanth Saratchandran, Cameron Gordon, Zeyu Zhang, Simon Lucey
Abstract: Low-rank decomposition has emerged as a vital tool for enhancing parameter efficiency in neural network architectures, gaining traction across diverse applications in machine learning. These techniques significantly lower the number of parameters, striking a balance between compactness and performance. However, a common challenge has been the compromise between parameter efficiency and the accuracy of the model, where reduced parameters often lead to diminished accuracy compared to their full-rank counterparts. In this work, we propose a novel theoretical framework that integrates a sinusoidal function within the low-rank decomposition process. This approach not only preserves the benefits of the parameter efficiency characteristic of low-rank methods but also increases the decomposition's rank, thereby enhancing model accuracy. Our method proves to be an adaptable enhancement for existing low-rank models, as evidenced by its successful application in Vision Transformers (ViT), Large Language Models (LLMs), Neural Radiance Fields (NeRF), and 3D shape modeling. This demonstrates the wide-ranging potential and efficiency of our proposed technique.
Authors: Hyejin Park, Jeongyeon Hwang, Sunung Mun, Sangdon Park, Jungseul Ok
Abstract: Test-time adaptation (TTA) has emerged as a promising solution to address performance decay due to unforeseen distribution shifts between training and test data. While recent TTA methods excel in adapting to test data variations, such adaptability exposes a model to vulnerability against malicious examples, an aspect that has received limited attention. Previous studies have uncovered security vulnerabilities within TTA even when a small proportion of the test batch is maliciously manipulated. In response to the emerging threat, we propose median batch normalization (MedBN), leveraging the robustness of the median for statistics estimation within the batch normalization layer during test-time inference. Our method is algorithm-agnostic, thus allowing seamless integration with existing TTA frameworks. Our experimental results on benchmark datasets, including CIFAR10-C, CIFAR100-C and ImageNet-C, consistently demonstrate that MedBN outperforms existing approaches in maintaining robust performance across different attack scenarios, encompassing both instant and cumulative attacks. Through extensive experiments, we show that our approach sustains the performance even in the absence of attacks, achieving a practical balance between robustness and performance.
Authors: Baris Imre, Elina Thibeau-Sutre, Jorieke Reimer, Kuan Kho, Jelmer M. Wolterink
Abstract: Chronic subdural hematoma (cSDH) is a common neurological condition characterized by the accumulation of blood between the brain and the dura mater. This accumulation of blood can exert pressure on the brain, potentially leading to fatal outcomes. Treatment options for cSDH are limited to invasive surgery or non-invasive management. Traditionally, the midline shift, hand-measured by experts from an ideal sagittal plane, and the hematoma volume have been the primary metrics for quantifying and analyzing cSDH. However, these approaches do not quantify the local 3D brain deformation caused by cSDH. We propose a novel method using anatomy-aware unsupervised diffeomorphic pseudo-healthy synthesis to generate brain deformation fields. The deformation fields derived from this process are utilized to extract biomarkers that quantify the shift in the brain due to cSDH. We use CT scans of 121 patients for training and validation of our method and find that our metrics allow the identification of patients who require surgery. Our results indicate that automatically obtained brain deformation fields might contain prognostic value for personalized cSDH treatment. Our implementation is available on: github.com/Barisimre/brain-morphing
Authors: Ezequiel de la Rosa, Mauricio Reyes, Sook-Lei Liew, Alexandre Hutton, Roland Wiest, Johannes Kaesmacher, Uta Hanning, Arsany Hakim, Richard Zubal, Waldo Valenzuela, David Robben, Diana M. Sima, Vincenzo Anania, Arne Brys, James A. Meakin, Anne Mickan, Gabriel Broocks, Christian Heitkamp, Shengbo Gao, Kongming Liang, Ziji Zhang, Md Mahfuzur Rahman Siddiquee, Andriy Myronenko, Pooya Ashtari, Sabine Van Huffel, Hyun-su Jeong, Chi-ho Yoon, Chulhong Kim, Jiayu Huo, Sebastien Ourselin, Rachel Sparks, Albert Cl\`erigues, Arnau Oliver, Xavier Llad\'o, Liam Chalcroft, Ioannis Pappas, Jeroen Bertels, Ewout Heylen, Juliette Moreau, Nima Hatami, Carole Frindel, Abdul Qayyum, Moona Mazher, Domenec Puig, Shao-Chieh Lin, Chun-Jung Juan, Tianxi Hu, Lyndon Boone, Maged Goubran, Yi-Jui Liu, Susanne Wegener, Florian Kofler, Ivan Ezhov, Suprosanna Shit, Moritz R. Hernandez Petzsche, Bjoern Menze, Jan S. Kirschke, Benedikt Wiestler
Abstract: Diffusion-weighted MRI (DWI) is essential for stroke diagnosis, treatment decisions, and prognosis. However, image and disease variability hinder the development of generalizable AI algorithms with clinical value. We address this gap by presenting a novel ensemble algorithm derived from the 2022 Ischemic Stroke Lesion Segmentation (ISLES) challenge. ISLES'22 provided 400 patient scans with ischemic stroke from various medical centers, facilitating the development of a wide range of cutting-edge segmentation algorithms by the research community. Through collaboration with leading teams, we combined top-performing algorithms into an ensemble model that overcomes the limitations of individual solutions. Our ensemble model achieved superior ischemic lesion detection and segmentation accuracy on our internal test set compared to individual algorithms. This accuracy generalized well across diverse image and disease variables. Furthermore, the model excelled in extracting clinical biomarkers. Notably, in a Turing-like test, neuroradiologists consistently preferred the algorithm's segmentations over manual expert efforts, highlighting increased comprehensiveness and precision. Validation using a real-world external dataset (N=1686) confirmed the model's generalizability. The algorithm's outputs also demonstrated strong correlations with clinical scores (admission NIHSS and 90-day mRS) on par with or exceeding expert-derived results, underlining its clinical relevance. This study offers two key findings. First, we present an ensemble algorithm (https://github.com/Tabrisrei/ISLES22_Ensemble) that detects and segments ischemic stroke lesions on DWI across diverse scenarios on par with expert (neuro)radiologists. Second, we show the potential for biomedical challenge outputs to extend beyond the challenge's initial objectives, demonstrating their real-world clinical applicability.
Authors: Amy Rafferty, Rishi Ramaesh, Ajitha Rajan
Abstract: The rapidly advancing field of Explainable Artificial Intelligence (XAI) aims to tackle the issue of trust regarding the use of complex black-box deep learning models in real-world applications. Existing post-hoc XAI techniques have recently been shown to have poor performance on medical data, producing unreliable explanations which are infeasible for clinical use. To address this, we propose an ante-hoc approach based on concept bottleneck models which introduces for the first time clinical concepts into the classification pipeline, allowing the user valuable insight into the decision-making process. On a large public dataset of chest X-rays and associated medical reports, we focus on the binary classification task of lung cancer detection. Our approach yields improved classification performance in lung cancer detection when compared to baseline deep learning models (F1 > 0.9), while also generating clinically relevant and more reliable explanations than existing techniques. We evaluate our approach against post-hoc image XAI techniques LIME and SHAP, as well as CXR-LLaVA, a recent textual XAI tool which operates in the context of question answering on chest X-rays.
Authors: Grzegorz Skorupko, Richard Osuala, Zuzanna Szafranowska, Kaisar Kushibar, Nay Aung, Steffen E Petersen, Karim Lekadir, Polyxeni Gkontra
Abstract: The progress in deep learning solutions for disease diagnosis and prognosis based on cardiac magnetic resonance imaging is hindered by highly imbalanced and biased training data. To address this issue, we propose a method to alleviate imbalances inherent in datasets through the generation of synthetic data based on sensitive attributes such as sex, age, body mass index, and health condition. We adopt ControlNet based on a denoising diffusion probabilistic model to condition on text assembled from patient metadata and cardiac geometry derived from segmentation masks using a large-cohort study, specifically, the UK Biobank. We assess our method by evaluating the realism of the generated images using established quantitative metrics. Furthermore, we conduct a downstream classification task aimed at debiasing a classifier by rectifying imbalances within underrepresented groups through synthetically generated samples. Our experiments demonstrate the effectiveness of the proposed approach in mitigating dataset imbalances, such as the scarcity of younger patients or individuals with normal BMI level suffering from heart failure. This work represents a major step towards the adoption of synthetic data for the development of fair and generalizable models for medical classification tasks. Notably, we conduct all our experiments using a single, consumer-level GPU to highlight the feasibility of our approach within resource-constrained environments. Our code is available at https://github.com/faildeny/debiasing-cardiac-mri.
Authors: Dong-Hwan Jang, Sangdoo Yun, Dongyoon Han
Abstract: This paper introduces an efficient fine-tuning method for large pre-trained models, offering strong in-distribution (ID) and out-of-distribution (OOD) performance. Breaking away from traditional practices that need a multitude of fine-tuned models for averaging, our approach employs significantly fewer models to achieve final weights yet yield superior accuracy. Drawing from key insights in the weight space of fine-tuned weights, we uncover a strong link between the performance and proximity to the center of weight space. Based on this, we introduce a method that approximates a center-close weight using only two fine-tuned models, applicable during or after training. Our innovative layer-wise weight averaging technique surpasses state-of-the-art model methods such as Model Soup, utilizing only two fine-tuned models. This strategy can be aptly coined Model Stock, highlighting its reliance on selecting a minimal number of models to draw a more optimized-averaged model. We demonstrate the efficacy of Model Stock with fine-tuned models based upon pre-trained CLIP architectures, achieving remarkable performance on both ID and OOD tasks on the standard benchmarks, all while barely bringing extra computational demands. Our code and pre-trained models are available at https://github.com/naver-ai/model-stock.
Authors: Chengzu Li, Chao Zhang, Simone Teufel, Rama Sanand Doddipatla, Svetlana Stoyanchev
Abstract: We are interested in the generation of navigation instructions, either in their own right or as training material for robotic navigation task. In this paper, we propose a new approach to navigation instruction generation by framing the problem as an image captioning task using semantic maps as visual input. Conventional approaches employ a sequence of panorama images to generate navigation instructions. Semantic maps abstract away from visual details and fuse the information in multiple panorama images into a single top-down representation, thereby reducing computational complexity to process the input. We present a benchmark dataset for instruction generation using semantic maps, propose an initial model and ask human subjects to manually assess the quality of generated instructions. Our initial investigations show promise in using semantic maps for instruction generation instead of a sequence of panorama images, but there is vast scope for improvement. We release the code for data preparation and model training at https://github.com/chengzu-li/VLGen.
Authors: Avinash Ummadisingu, Jongkeum Choi, Koki Yamane, Shimpei Masuda, Naoki Fukaya, Kuniyuki Takahashi
Abstract: Acquiring accurate depth information of transparent objects using off-the-shelf RGB-D cameras is a well-known challenge in Computer Vision and Robotics. Depth estimation/completion methods are typically employed and trained on datasets with quality depth labels acquired from either simulation, additional sensors or specialized data collection setups and known 3d models. However, acquiring reliable depth information for datasets at scale is not straightforward, limiting training scalability and generalization. Neural Radiance Fields (NeRFs) are learning-free approaches and have demonstrated wide success in novel view synthesis and shape recovery. However, heuristics and controlled environments (lights, backgrounds, etc) are often required to accurately capture specular surfaces. In this paper, we propose using Visual Foundation Models (VFMs) for segmentation in a zero-shot, label-free way to guide the NeRF reconstruction process for these objects via the simultaneous reconstruction of semantic fields and extensions to increase robustness. Our proposed method Segmentation-AIDed NeRF (SAID-NeRF) shows significant performance on depth completion datasets for transparent objects and robotic grasping.
Authors: Ole Hall, Anil Yaman
Abstract: Generative Adversarial Networks (GANs) have shown great success in generating high quality images and are thus used as one of the main approaches to generate art images. However, usually the image generation process involves sampling from the latent space of the learned art representations, allowing little control over the output. In this work, we first employ GANs that are trained to produce creative images using an architecture known as Creative Adversarial Networks (CANs), then, we employ an evolutionary approach to navigate within the latent space of the models to discover images. We use automatic aesthetic and collaborative interactive human evaluation metrics to assess the generated images. In the human interactive evaluation case, we propose a collaborative evaluation based on the assessments of several participants. Furthermore, we also experiment with an intelligent mutation operator that aims to improve the quality of the images through local search based on an aesthetic measure. We evaluate the effectiveness of this approach by comparing the results produced by the automatic and collaborative interactive evolution. The results show that the proposed approach can generate highly attractive art images when the evolution is guided by collaborative human feedback.
Authors: Zeren Chen, Zhelun Shi, Xiaoya Lu, Lehan He, Sucheng Qian, Hao Shu Fang, Zhenfei Yin, Wanli Ouyang, Jing Shao, Yu Qiao, Cewu Lu, Lu Sheng
Abstract: The ultimate goals of robotic learning is to acquire a comprehensive and generalizable robotic system capable of performing both seen skills within the training distribution and unseen skills in novel environments. Recent progress in utilizing language models as high-level planners has demonstrated that the complexity of tasks can be reduced through decomposing them into primitive-level plans, making it possible to generalize on novel robotic tasks in a composable manner. Despite the promising future, the community is not yet adequately prepared for composable generalization agents, particularly due to the lack of primitive-level real-world robotic datasets. In this paper, we propose a primitive-level robotic dataset, namely RH20T-P, which contains about 33000 video clips covering 44 diverse and complicated robotic tasks. Each clip is manually annotated according to a set of meticulously designed primitive skills, facilitating the future development of composable generalization agents. To validate the effectiveness of RH20T-P, we also construct a potential and scalable agent based on RH20T-P, called RA-P. Equipped with two planners specialized in task decomposition and motion planning, RA-P can adapt to novel physical skills through composable generalization. Our website and videos can be found at https://sites.google.com/view/rh20t-primitive/main. Dataset and code will be made available soon.
Authors: Hui Zhang, Sammy Christen, Zicong Fan, Otmar Hilliges, Jie Song
Abstract: Human hands possess the dexterity to interact with diverse objects such as grasping specific parts of the objects and/or approaching them from desired directions. More importantly, humans can grasp objects of any shape without object-specific skills. Recent works synthesize grasping motions following single objectives such as a desired approach heading direction or a grasping area. Moreover, they usually rely on expensive 3D hand-object data during training and inference, which limits their capability to synthesize grasping motions for unseen objects at scale. In this paper, we unify the generation of hand-object grasping motions across multiple motion objectives, diverse object shapes and dexterous hand morphologies in a policy learning framework GraspXL. The objectives are composed of the graspable area, heading direction during approach, wrist rotation, and hand position. Without requiring any 3D hand-object interaction data, our policy trained with 58 objects can robustly synthesize diverse grasping motions for more than 500k unseen objects with a success rate of 82.2%. At the same time, the policy adheres to objectives, which enables the generation of diverse grasps per object. Moreover, we show that our framework can be deployed to different dexterous hands and work with reconstructed or generated objects. We quantitatively and qualitatively evaluate our method to show the efficacy of our approach. Our model and code will be available.
Authors: Amy Tabb, Henry Medeiros, Mitchell J. Feldmann, Thiago T. Santos
Abstract: This paper describes CALICO, a method for multi-camera calibration suitable for challenging contexts: stationary and mobile multi-camera systems, cameras without overlapping fields of view, and non-synchronized cameras. Recent approaches are roughly divided into infrastructure- and pattern-based. Infrastructure-based approaches use the scene's features to calibrate, while pattern-based approaches use calibration patterns. Infrastructure-based approaches are not suitable for stationary camera systems, and pattern-based approaches may constrain camera placement because shared fields of view or extremely large patterns are required. CALICO is a pattern-based approach, where the multi-calibration problem is formulated using rigidity constraints between patterns and cameras. We use a {\it pattern rig}: several patterns rigidly attached to each other or some structure. We express the calibration problem as that of algebraic and reprojection error minimization problems. Simulated and real experiments demonstrate the method in a variety of settings. CALICO compared favorably to Kalibr. Mean reconstruction accuracy error was $\le 0.71$ mm for real camera rigs, and $\le 1.11$ for simulated camera rigs. Code and data releases are available at \cite{tabb_amy_2019_3520866} and \url{https://github.com/amy-tabb/calico}.
Authors: Yuyin Zhou, Xianhang Li, Fengze Liu, Qingyue Wei, Xuxi Chen, Lequan Yu, Cihang Xie, Matthew P. Lungren, Lei Xing
Abstract: Deep neural networks have shown great success in representation learning. However, when learning with noisy labels (LNL), they can easily overfit and fail to generalize to new data. This paper introduces a simple and effective method, named Learning to Bootstrap (L2B), which enables models to bootstrap themselves using their own predictions without being adversely affected by erroneous pseudo-labels. It achieves this by dynamically adjusting the importance weight between real observed and generated labels, as well as between different samples through meta-learning. Unlike existing instance reweighting methods, the key to our method lies in a new, versatile objective that enables implicit relabeling concurrently, leading to significant improvements without incurring additional costs. L2B offers several benefits over the baseline methods. It yields more robust models that are less susceptible to the impact of noisy labels by guiding the bootstrapping procedure more effectively. It better exploits the valuable information contained in corrupted instances by adapting the weights of both instances and labels. Furthermore, L2B is compatible with existing LNL methods and delivers competitive results spanning natural and medical imaging tasks including classification and segmentation under both synthetic and real-world noise. Extensive experiments demonstrate that our method effectively mitigates the challenges of noisy labels, often necessitating few to no validation samples, and is well generalized to other tasks such as image segmentation. This not only positions it as a robust complement to existing LNL techniques but also underscores its practical applicability. The code and models are available at https://github.com/yuyinzhou/l2b.
Authors: Saemi Moon, Seunghyuk Cho, Dongwoo Kim
Abstract: We tackle the problem of feature unlearning from a pre-trained image generative model: GANs and VAEs. Unlike a common unlearning task where an unlearning target is a subset of the training set, we aim to unlearn a specific feature, such as hairstyle from facial images, from the pre-trained generative models. As the target feature is only presented in a local region of an image, unlearning the entire image from the pre-trained model may result in losing other details in the remaining region of the image. To specify which features to unlearn, we collect randomly generated images that contain the target features. We then identify a latent representation corresponding to the target feature and then use the representation to fine-tune the pre-trained model. Through experiments on MNIST, CelebA, and FFHQ datasets, we show that target features are successfully removed while keeping the fidelity of the original models. Further experiments with an adversarial attack show that the unlearned model is more robust under the presence of malicious parties.
Authors: Han Liang, Wenqian Zhang, Wenxuan Li, Jingyi Yu, Lan Xu
Abstract: We have recently seen tremendous progress in diffusion advances for generating realistic human motions. Yet, they largely disregard the multi-human interactions. In this paper, we present InterGen, an effective diffusion-based approach that incorporates human-to-human interactions into the motion diffusion process, which enables layman users to customize high-quality two-person interaction motions, with only text guidance. We first contribute a multimodal dataset, named InterHuman. It consists of about 107M frames for diverse two-person interactions, with accurate skeletal motions and 23,337 natural language descriptions. For the algorithm side, we carefully tailor the motion diffusion model to our two-person interaction setting. To handle the symmetry of human identities during interactions, we propose two cooperative transformer-based denoisers that explicitly share weights, with a mutual attention mechanism to further connect the two denoising processes. Then, we propose a novel representation for motion input in our interaction diffusion model, which explicitly formulates the global relations between the two performers in the world frame. We further introduce two novel regularization terms to encode spatial relations, equipped with a corresponding damping scheme during the training of our interaction diffusion model. Extensive experiments validate the effectiveness and generalizability of InterGen. Notably, it can generate more diverse and compelling two-person motions than previous methods and enables various downstream applications for human interactions.
Authors: Romain Loiseau, Elliot Vincent, Mathieu Aubry, Loic Landrieu
Abstract: We propose an unsupervised method for parsing large 3D scans of real-world scenes with easily-interpretable shapes. This work aims to provide a practical tool for analyzing 3D scenes in the context of aerial surveying and mapping, without the need for user annotations. Our approach is based on a probabilistic reconstruction model that decomposes an input 3D point cloud into a small set of learned prototypical 3D shapes. The resulting reconstruction is visually interpretable and can be used to perform unsupervised instance and low-shot semantic segmentation of complex scenes. We demonstrate the usefulness of our model on a novel dataset of seven large aerial LiDAR scans from diverse real-world scenarios. Our approach outperforms state-of-the-art unsupervised methods in terms of decomposition accuracy while remaining visually interpretable. Our code and dataset are available at https://romainloiseau.fr/learnable-earth-parser/
Authors: Baao Xie, Bohan Li, Zequn Zhang, Junting Dong, Xin Jin, Jingyu Yang, Wenjun Zeng
Abstract: 3D representation disentanglement aims to identify, decompose, and manipulate the underlying explanatory factors of 3D data, which helps AI fundamentally understand our 3D world. This task is currently under-explored and poses great challenges: (i) the 3D representations are complex and in general contains much more information than 2D image; (ii) many 3D representations are not well suited for gradient-based optimization, let alone disentanglement. To address these challenges, we use NeRF as a differentiable 3D representation, and introduce a self-supervised Navigation to identify interpretable semantic directions in the latent space. To our best knowledge, this novel method, dubbed NaviNeRF, is the first work to achieve fine-grained 3D disentanglement without any priors or supervisions. Specifically, NaviNeRF is built upon the generative NeRF pipeline, and equipped with an Outer Navigation Branch and an Inner Refinement Branch. They are complementary -- the outer navigation is to identify global-view semantic directions, and the inner refinement dedicates to fine-grained attributes. A synergistic loss is further devised to coordinate two branches. Extensive experiments demonstrate that NaviNeRF has a superior fine-grained 3D disentanglement ability than the previous 3D-aware models. Its performance is also comparable to editing-oriented models relying on semantic or geometry priors.
Authors: Xingzhe Su, Changwen Zheng, Wenwen Qiang, Fengge Wu, Junsuo Zhao, Fuchun Sun, Hui Xiong
Abstract: Generative Adversarial Networks (GANs) have shown notable accomplishments in remote sensing domain. However, this paper reveals that their performance on remote sensing images falls short when compared to their impressive results with natural images. This study identifies a previously overlooked issue: GANs exhibit a heightened susceptibility to overfitting on remote sensing images.To address this challenge, this paper analyzes the characteristics of remote sensing images and proposes manifold constraint regularization, a novel approach that tackles overfitting of GANs on remote sensing images for the first time. Our method includes a new measure for evaluating the structure of the data manifold. Leveraging this measure, we propose the manifold constraint regularization term, which not only alleviates the overfitting problem, but also promotes alignment between the generated and real data manifolds, leading to enhanced quality in the generated images. The effectiveness and versatility of this method have been corroborated through extensive validation on various remote sensing datasets and GAN models. The proposed method not only enhances the quality of the generated images, reflected in a 3.13\% improvement in Frechet Inception Distance (FID) score, but also boosts the performance of the GANs on downstream tasks, evidenced by a 3.76\% increase in classification accuracy.
Authors: Tz-Ying Wu, Chih-Hui Ho, Nuno Vasconcelos
Abstract: Visual-language foundation models, like CLIP, learn generalized representations that enable zero-shot open-set classification. Few-shot adaptation methods, based on prompt tuning, have been shown to further improve performance on downstream datasets. However, these methods do not fare well in the taxonomic open set (TOS) setting, where the classifier is asked to make predictions from label sets across different levels of semantic granularity. Frequently, they infer incorrect labels at coarser taxonomic class levels, even when the inference at the leaf level (original class labels) is correct. To address this problem, we propose a prompt tuning technique that calibrates the hierarchical consistency of model predictions. A set of metrics of hierarchical consistency, the Hierarchical Consistent Accuracy (HCA) and the Mean Treecut Accuracy (MTA), are first proposed to evaluate TOS model performance. A new Prompt Tuning for Hierarchical Consistency (ProTeCt) technique is then proposed to calibrate classification across label set granularities. Results show that ProTeCt can be combined with existing prompt tuning methods to significantly improve TOS classification without degrading the leaf level classification performance.
Authors: Aniket Gupta, Yiming Xie, Hanumant Singh, Huaizu Jiang
Abstract: Deep neural networks endow the downsampled superpoints with highly discriminative feature representations. Previous dominant point cloud registration approaches match these feature representations as the first step, e.g., using the Sinkhorn algorithm. A RANSAC-like method is then usually adopted as a post-processing refinement to filter the outliers. Other dominant method is to directly predict the superpoint matchings using learned MLP layers. Both of them have drawbacks: RANSAC-based methods are computationally intensive and prediction-based methods suffer from outputing non-existing points in the point cloud. In this paper, we propose a straightforward and effective baseline to find correspondences of superpoints in a global matching manner. We employ the normalized matching scores as weights for each correspondence, allowing us to reject the outliers and further weigh the rest inliers when fitting the transformation matrix without relying on the cumbersome RANSAC. Moreover, the entire model can be trained in an end-to-end fashion, leading to better accuracy. Our simple yet effective baseline shows comparable or even better results than state-of-the-art methods on three datasets including ModelNet, 3DMatch, and KITTI. We do not advocate our approach to be \emph{the} solution for point cloud registration but use the results to emphasize the role of matching strategy for point cloud registration. The code and models are available at https://github.com/neu-vi/Superpoints_Registration.
Authors: Amrit Diggavi Seshadri, Alessandra Russo
Abstract: In this work, following the intuition that adverbs describing scene-sequences are best identified by reasoning over high-level concepts of object-behavior, we propose the design of a new framework that reasons over object-behaviours extracted from raw-video-clips to recognize the clip's corresponding adverb-types. Importantly, while previous works for general scene adverb-recognition assume knowledge of the clips underlying action-types, our method is directly applicable in the more general problem setting where the action-type of a video-clip is unknown. Specifically, we propose a novel pipeline that extracts human-interpretable object-behaviour-facts from raw video clips and propose novel symbolic and transformer based reasoning methods that operate over these extracted facts to identify adverb-types. Experiment results demonstrate that our proposed methods perform favourably against the previous state-of-the-art. Additionally, to support efforts in symbolic video-processing, we release two new datasets of object-behaviour-facts extracted from raw video clips - the MSR-VTT-ASP and ActivityNet-ASP datasets.
Authors: Xiaoyan Xing, Konrad Groh, Sezer Karaoglu, Theo Gevers
Abstract: The purpose of intrinsic decomposition is to separate an image into its albedo (reflective properties) and shading components (illumination properties). This is challenging because it's an ill-posed problem. Conventional approaches primarily concentrate on 2D imagery and fail to fully exploit the capabilities of 3D data representation. 3D point clouds offer a more comprehensive format for representing scenes, as they combine geometric and color information effectively. To this end, in this paper, we introduce Point Intrinsic Net (PoInt-Net), which leverages 3D point cloud data to concurrently estimate albedo and shading maps. The merits of PoInt-Net include the following aspects. First, the model is efficient, achieving consistent performance across point clouds of any size with training only required on small-scale point clouds. Second, it exhibits remarkable robustness; even when trained exclusively on datasets comprising individual objects, PoInt-Net demonstrates strong generalization to unseen objects and scenes. Third, it delivers superior accuracy over conventional 2D approaches, demonstrating enhanced performance across various metrics on different datasets. (Code Released)
Authors: Yang Qin, Yingke Chen, Dezhong Peng, Xi Peng, Joey Tianyi Zhou, Peng Hu
Abstract: Text-to-image person re-identification (TIReID) is a compelling topic in the cross-modal community, which aims to retrieve the target person based on a textual query. Although numerous TIReID methods have been proposed and achieved promising performance, they implicitly assume the training image-text pairs are correctly aligned, which is not always the case in real-world scenarios. In practice, the image-text pairs inevitably exist under-correlated or even false-correlated, a.k.a noisy correspondence (NC), due to the low quality of the images and annotation errors. To address this problem, we propose a novel Robust Dual Embedding method (RDE) that can learn robust visual-semantic associations even with NC. Specifically, RDE consists of two main components: 1) A Confident Consensus Division (CCD) module that leverages the dual-grained decisions of dual embedding modules to obtain a consensus set of clean training data, which enables the model to learn correct and reliable visual-semantic associations. 2) A Triplet Alignment Loss (TAL) relaxes the conventional Triplet Ranking loss with the hardest negative samples to a log-exponential upper bound over all negative ones, thus preventing the model collapse under NC and can also focus on hard-negative samples for promising performance. We conduct extensive experiments on three public benchmarks, namely CUHK-PEDES, ICFG-PEDES, and RSTPReID, to evaluate the performance and robustness of our RDE. Our method achieves state-of-the-art results both with and without synthetic noisy correspondences on all three datasets. Code is available at https://github.com/QinYang79/RDE.
Authors: Tom Van Wouwe, Seunghwan Lee, Antoine Falisse, Scott Delp, C. Karen Liu
Abstract: Motion capture from a limited number of body-worn sensors, such as inertial measurement units (IMUs) and pressure insoles, has important applications in health, human performance, and entertainment. Recent work has focused on accurately reconstructing whole-body motion from a specific sensor configuration using six IMUs. While a common goal across applications is to use the minimal number of sensors to achieve required accuracy, the optimal arrangement of the sensors might differ from application to application. We propose a single diffusion model, DiffusionPoser, which reconstructs human motion in real-time from an arbitrary combination of sensors, including IMUs placed at specified locations, and, pressure insoles. Unlike existing methods, our model grants users the flexibility to determine the number and arrangement of sensors tailored to the specific activity of interest, without the need for retraining. A novel autoregressive inferencing scheme ensures real-time motion reconstruction that closely aligns with measured sensor signals. The generative nature of DiffusionPoser ensures realistic behavior, even for degrees-of-freedom not directly measured. Qualitative results can be found on our website: https://diffusionposer.github.io/.
Authors: Llukman Cerkezi, Paolo Favaro
Abstract: We propose a novel method for 3D object reconstruction from a sparse set of views captured from a 360-degree calibrated camera rig. We represent the object surface through a hybrid model that uses both an MLP-based neural representation and a triangle mesh. A key contribution in our work is a novel object-centric sampling scheme of the neural representation, where rays are shared among all views. This efficiently concentrates and reduces the number of samples used to update the neural model at each iteration. This sampling scheme relies on the mesh representation to ensure also that samples are well-distributed along its normals. The rendering is then performed efficiently by a differentiable renderer. We demonstrate that this sampling scheme results in a more effective training of the neural representation, does not require the additional supervision of segmentation masks, yields state of the art 3D reconstructions, and works with sparse views on the Google's Scanned Objects, Tank and Temples and MVMC Car datasets. Code available at: https://github.com/llukmancerkezi/ROSTER
Authors: Jicheng Yuan, Anh Le-Tuan, Manh Nguyen-Duc, Trung-Kien Tran, Manfred Hauswirth, Danh Le-Phuoc
Abstract: The availability of vast amounts of visual data with heterogeneous features is a key factor for developing, testing, and benchmarking of new computer vision (CV) algorithms and architectures. Most visual datasets are created and curated for specific tasks or with limited image data distribution for very specific situations, and there is no unified approach to manage and access them across diverse sources, tasks, and taxonomies. This not only creates unnecessary overheads when building robust visual recognition systems, but also introduces biases into learning systems and limits the capabilities of data-centric AI. To address these problems, we propose the Vision Knowledge Graph (VisionKG), a novel resource that interlinks, organizes and manages visual datasets via knowledge graphs and Semantic Web technologies. It can serve as a unified framework facilitating simple access and querying of state-of-the-art visual datasets, regardless of their heterogeneous formats and taxonomies. One of the key differences between our approach and existing methods is that ours is knowledge-based rather than metadatabased. It enhances the enrichment of the semantics at both image and instance levels and offers various data retrieval and exploratory services via SPARQL. VisionKG currently contains 519 million RDF triples that describe approximately 40 million entities, and are accessible at https://vision.semkg.org and through APIs. With the integration of 30 datasets and four popular CV tasks, we demonstrate its usefulness across various scenarios when working with CV pipelines.
URLs: https://vision.semkg.org
Authors: Shan Lin, Albert J. Miao, Ali Alabiad, Fei Liu, Kaiyuan Wang, Jingpei Lu, Florian Richter, Michael C. Yip
Abstract: Manipulation of tissue with surgical tools often results in large deformations that current methods in tracking and reconstructing algorithms have not effectively addressed. A major source of tracking errors during large deformations stems from wrong data association between observed sensor measurements with previously tracked scene. To mitigate this issue, we present a surgical perception framework, SuPerPM, that leverages learning-based non-rigid point cloud matching for data association, thus accommodating larger deformations. The learning models typically require training data with ground truth point cloud correspondences, which is challenging or even impractical to collect in surgical environments. Thus, for tuning the learning model, we gather endoscopic data of soft tissue being manipulated by a surgical robot and then establish correspondences between point clouds at different time points to serve as ground truth. This was achieved by employing a position-based dynamics (PBD) simulation to ensure that the correspondences adhered to physical constraints. The proposed framework is demonstrated on several challenging surgical datasets that are characterized by large deformations, achieving superior performance over state-of-the-art surgical scene tracking algorithms.
Authors: Tom Kelly, John Femiani, Peter Wonka
Abstract: We present WinSyn, a unique dataset and testbed for creating high-quality synthetic data with procedural modeling techniques. The dataset contains high-resolution photographs of windows, selected from locations around the world, with 89,318 individual window crops showcasing diverse geometric and material characteristics. We evaluate a procedural model by training semantic segmentation networks on both synthetic and real images and then comparing their performances on a shared test set of real images. Specifically, we measure the difference in mean Intersection over Union (mIoU) and determine the effective number of real images to match synthetic data's training performance. We design a baseline procedural model as a benchmark and provide 21,290 synthetically generated images. By tuning the procedural model, key factors are identified which significantly influence the model's fidelity in replicating real-world scenarios. Importantly, we highlight the challenge of procedural modeling using current techniques, especially in their ability to replicate the spatial semantics of real-world scenarios. This insight is critical because of the potential of procedural models to bridge to hidden scene aspects such as depth, reflectivity, material properties, and lighting conditions.
Authors: Zhenghan Fang, Sam Buchanan, Jeremias Sulam
Abstract: Proximal operators are ubiquitous in inverse problems, commonly appearing as part of algorithmic strategies to regularize problems that are otherwise ill-posed. Modern deep learning models have been brought to bear for these tasks too, as in the framework of plug-and-play or deep unrolling, where they loosely resemble proximal operators. Yet, something essential is lost in employing these purely data-driven approaches: there is no guarantee that a general deep network represents the proximal operator of any function, nor is there any characterization of the function for which the network might provide some approximate proximal. This not only makes guaranteeing convergence of iterative schemes challenging but, more fundamentally, complicates the analysis of what has been learned by these networks about their training data. Herein we provide a framework to develop learned proximal networks (LPN), prove that they provide exact proximal operators for a data-driven nonconvex regularizer, and show how a new training strategy, dubbed proximal matching, provably promotes the recovery of the log-prior of the true data distribution. Such LPN provide general, unsupervised, expressive proximal operators that can be used for general inverse problems with convergence guarantees. We illustrate our results in a series of cases of increasing complexity, demonstrating that these models not only result in state-of-the-art performance, but provide a window into the resulting priors learned from data.
Authors: Youbo Lei, Feifei He, Chen Chen, Yingbin Mo, Si Jia Li, Defeng Xie, Haonan Lu
Abstract: Due to the success of large-scale visual-language pretraining (VLP) models and the widespread use of image-text retrieval in industry areas, it is now critically necessary to reduce the model size and streamline their mobile-device deployment. Single- and dual-stream model structures are commonly used in image-text retrieval with the goal of closing the semantic gap between textual and visual modalities. While single-stream models use deep feature fusion to achieve more accurate cross-model alignment, dual-stream models are better at offline indexing and fast inference.We propose a Multi-teacher Cross-modality Alignment Distillation (MCAD) technique to integrate the advantages of single- and dual-stream models. By incorporating the fused single-stream features into the image and text features of the dual-stream model, we formulate new modified teacher similarity distributions and features. Then, we conduct both distribution and feature distillation to boost the capability of the student dual-stream model, achieving high retrieval performance without increasing inference complexity.Extensive experiments demonstrate the remarkable performance and high efficiency of MCAD on image-text retrieval tasks. Furthermore, we implement a lightweight CLIP model on Snapdragon/Dimensity chips with only $\sim$100M running memory and $\sim$8.0ms search latency, achieving the mobile-device application of VLP models.
Authors: Yibin Wang, Weizhong Zhang, Jianwei Zheng, Cheng Jin
Abstract: Layout-to-image synthesis is an emerging technique in conditional image generation. It aims to generate complex scenes, where users require fine control over the layout of the objects in a scene. However, it remains challenging to control the object coherence, including semantic coherence (e.g., the cat looks at the flowers or not) and physical coherence (e.g., the hand and the racket should not be misaligned). In this paper, we propose a novel diffusion model with effective global semantic fusion (GSF) and self-similarity feature enhancement modules to guide the object coherence for this task. For semantic coherence, we argue that the image caption contains rich information for defining the semantic relationship within the objects in the images. Instead of simply employing cross-attention between captions and generated images, which addresses the highly relevant layout restriction and semantic coherence separately and thus leads to unsatisfying results shown in our experiments, we develop GSF to fuse the supervision from the layout restriction and semantic coherence requirement and exploit it to guide the image synthesis process. Moreover, to improve the physical coherence, we develop a Self-similarity Coherence Attention (SCA) module to explicitly integrate local contextual physical coherence into each pixel's generation process. Specifically, we adopt a self-similarity map to encode the coherence restrictions and employ it to extract coherent features from text embedding. Through visualization of our self-similarity map, we explore the essence of SCA, revealing that its effectiveness is not only in capturing reliable physical coherence patterns but also in enhancing complex texture generation. Extensive experiments demonstrate the superiority of our proposed method in both image generation quality and controllability.
Authors: Zhiyuan Yan, Yuhao Luo, Siwei Lyu, Qingshan Liu, Baoyuan Wu
Abstract: Deepfake detection faces a critical generalization hurdle, with performance deteriorating when there is a mismatch between the distributions of training and testing data. A broadly received explanation is the tendency of these detectors to be overfitted to forgery-specific artifacts, rather than learning features that are widely applicable across various forgeries. To address this issue, we propose a simple yet effective detector called LSDA (\underline{L}atent \underline{S}pace \underline{D}ata \underline{A}ugmentation), which is based on a heuristic idea: representations with a wider variety of forgeries should be able to learn a more generalizable decision boundary, thereby mitigating the overfitting of method-specific features (see Fig.~\ref{fig:toy}). Following this idea, we propose to enlarge the forgery space by constructing and simulating variations within and across forgery features in the latent space. This approach encompasses the acquisition of enriched, domain-specific features and the facilitation of smoother transitions between different forgery types, effectively bridging domain gaps. Our approach culminates in refining a binary classifier that leverages the distilled knowledge from the enhanced features, striving for a generalizable deepfake detector. Comprehensive experiments show that our proposed method is surprisingly effective and transcends state-of-the-art detectors across several widely used benchmarks.
Authors: Yutao Feng, Yintong Shang, Xuan Li, Tianjia Shao, Chenfanfu Jiang, Yin Yang
Abstract: We show that physics-based simulations can be seamlessly integrated with NeRF to generate high-quality elastodynamics of real-world objects. Unlike existing methods, we discretize nonlinear hyperelasticity in a meshless way, obviating the necessity for intermediate auxiliary shape proxies like a tetrahedral mesh or voxel grid. A quadratic generalized moving least square (Q-GMLS) is employed to capture nonlinear dynamics and large deformation on the implicit model. Such meshless integration enables versatile simulations of complex and codimensional shapes. We adaptively place the least-square kernels according to the NeRF density field to significantly reduce the complexity of the nonlinear simulation. As a result, physically realistic animations can be conveniently synthesized using our method for a wide range of hyperelastic materials at an interactive rate. For more information, please visit our project page at https://fytalon.github.io/pienerf/.
Authors: Zhen Zhao, Jingqun Tang, Chunhui Lin, Binghong Wu, Can Huang, Hao Liu, Xin Tan, Zhizhong Zhang, Yuan Xie
Abstract: Scene text recognition (STR) in the wild frequently encounters challenges when coping with domain variations, font diversity, shape deformations, etc. A straightforward solution is performing model fine-tuning tailored to a specific scenario, but it is computationally intensive and requires multiple model copies for various scenarios. Recent studies indicate that large language models (LLMs) can learn from a few demonstration examples in a training-free manner, termed "In-Context Learning" (ICL). Nevertheless, applying LLMs as a text recognizer is unacceptably resource-consuming. Moreover, our pilot experiments on LLMs show that ICL fails in STR, mainly attributed to the insufficient incorporation of contextual information from diverse samples in the training stage. To this end, we introduce E$^2$STR, a STR model trained with context-rich scene text sequences, where the sequences are generated via our proposed in-context training strategy. E$^2$STR demonstrates that a regular-sized model is sufficient to achieve effective ICL capabilities in STR. Extensive experiments show that E$^2$STR exhibits remarkable training-free adaptation in various scenarios and outperforms even the fine-tuned state-of-the-art approaches on public benchmarks. The code is released at https://github.com/bytedance/E2STR .
Authors: Fei Kong, Jinhao Duan, Lichao Sun, Hao Cheng, Renjing Xu, Hengtao Shen, Xiaofeng Zhu, Xiaoshuang Shi, Kaidi Xu
Abstract: Though diffusion models excel in image generation, their step-by-step denoising leads to slow generation speeds. Consistency training addresses this issue with single-step sampling but often produces lower-quality generations and requires high training costs. In this paper, we show that optimizing consistency training loss minimizes the Wasserstein distance between target and generated distributions. As timestep increases, the upper bound accumulates previous consistency training losses. Therefore, larger batch sizes are needed to reduce both current and accumulated losses. We propose Adversarial Consistency Training (ACT), which directly minimizes the Jensen-Shannon (JS) divergence between distributions at each timestep using a discriminator. Theoretically, ACT enhances generation quality, and convergence. By incorporating a discriminator into the consistency training framework, our method achieves improved FID scores on CIFAR10 and ImageNet 64$\times$64 and LSUN Cat 256$\times$256 datasets, retains zero-shot image inpainting capabilities, and uses less than $1/6$ of the original batch size and fewer than $1/2$ of the model parameters and training steps compared to the baseline method, this leads to a substantial reduction in resource consumption. Our code is available:https://github.com/kong13661/ACT
Authors: Sijie Cheng, Zhicheng Guo, Jingwen Wu, Kechen Fang, Peng Li, Huaping Liu, Yang Liu
Abstract: Vision-language models (VLMs) have recently shown promising results in traditional downstream tasks. Evaluation studies have emerged to assess their abilities, with the majority focusing on the third-person perspective, and only a few addressing specific tasks from the first-person perspective. However, the capability of VLMs to "think" from a first-person perspective, a crucial attribute for advancing autonomous agents and robotics, remains largely unexplored. To bridge this research gap, we introduce EgoThink, a novel visual question-answering benchmark that encompasses six core capabilities with twelve detailed dimensions. The benchmark is constructed using selected clips from egocentric videos, with manually annotated question-answer pairs containing first-person information. To comprehensively assess VLMs, we evaluate eighteen popular VLMs on EgoThink. Moreover, given the open-ended format of the answers, we use GPT-4 as the automatic judge to compute single-answer grading. Experimental results indicate that although GPT-4V leads in numerous dimensions, all evaluated VLMs still possess considerable potential for improvement in first-person perspective tasks. Meanwhile, enlarging the number of trainable parameters has the most significant impact on model performance on EgoThink. In conclusion, EgoThink serves as a valuable addition to existing evaluation benchmarks for VLMs, providing an indispensable resource for future research in the realm of embodied artificial intelligence and robotics.
Authors: Yan Xia, Letian Shi, Zifeng Ding, Jo\~ao F. Henriques, Daniel Cremers
Abstract: We tackle the problem of 3D point cloud localization based on a few natural linguistic descriptions and introduce a novel neural network, Text2Loc, that fully interprets the semantic relationship between points and text. Text2Loc follows a coarse-to-fine localization pipeline: text-submap global place recognition, followed by fine localization. In global place recognition, relational dynamics among each textual hint are captured in a hierarchical transformer with max-pooling (HTM), whereas a balance between positive and negative pairs is maintained using text-submap contrastive learning. Moreover, we propose a novel matching-free fine localization method to further refine the location predictions, which completely removes the need for complicated text-instance matching and is lighter, faster, and more accurate than previous methods. Extensive experiments show that Text2Loc improves the localization accuracy by up to $2\times$ over the state-of-the-art on the KITTI360Pose dataset. Our project page is publicly available at \url{https://yan-xia.github.io/projects/text2loc/}.
Authors: Zhihao Liang, Qi Zhang, Ying Feng, Ying Shan, Kui Jia
Abstract: We propose GS-IR, a novel inverse rendering approach based on 3D Gaussian Splatting (GS) that leverages forward mapping volume rendering to achieve photorealistic novel view synthesis and relighting results. Unlike previous works that use implicit neural representations and volume rendering (e.g. NeRF), which suffer from low expressive power and high computational complexity, we extend GS, a top-performance representation for novel view synthesis, to estimate scene geometry, surface material, and environment illumination from multi-view images captured under unknown lighting conditions. There are two main problems when introducing GS to inverse rendering: 1) GS does not support producing plausible normal natively; 2) forward mapping (e.g. rasterization and splatting) cannot trace the occlusion like backward mapping (e.g. ray tracing). To address these challenges, our GS-IR proposes an efficient optimization scheme that incorporates a depth-derivation-based regularization for normal estimation and a baking-based occlusion to model indirect lighting. The flexible and expressive GS representation allows us to achieve fast and compact geometry reconstruction, photorealistic novel view synthesis, and effective physically-based rendering. We demonstrate the superiority of our method over baseline methods through qualitative and quantitative evaluations on various challenging scenes.
Authors: Wenjie Zhao, Jia Li, Xin Dong, Yu Xiang, Yunhui Guo
Abstract: Semantic segmentation models, while effective for in-distribution categories, face challenges in real-world deployment due to encountering out-of-distribution (OoD) objects. Detecting these OoD objects is crucial for safety-critical applications. Existing methods rely on anomaly scores, but choosing a suitable threshold for generating masks presents difficulties and can lead to fragmentation and inaccuracy. This paper introduces a method to convert anomaly \textbf{S}core \textbf{T}o segmentation \textbf{M}ask, called S2M, a simple and effective framework for OoD detection in semantic segmentation. Unlike assigning anomaly scores to pixels, S2M directly segments the entire OoD object. By transforming anomaly scores into prompts for a promptable segmentation model, S2M eliminates the need for threshold selection. Extensive experiments demonstrate that S2M outperforms the state-of-the-art by approximately 20% in IoU and 40% in mean F1 score, on average, across various benchmarks including Fishyscapes, Segment-Me-If-You-Can, and RoadAnomaly datasets.
Authors: Zeyu Han, Fangrui Zhu, Qianru Lao, Huaizu Jiang
Abstract: Zero-shot referring expression comprehension aims at localizing bounding boxes in an image corresponding to provided textual prompts, which requires: (i) a fine-grained disentanglement of complex visual scene and textual context, and (ii) a capacity to understand relationships among disentangled entities. Unfortunately, existing large vision-language alignment (VLA) models, e.g., CLIP, struggle with both aspects so cannot be directly used for this task. To mitigate this gap, we leverage large foundation models to disentangle both images and texts into triplets in the format of (subject, predicate, object). After that, grounding is accomplished by calculating the structural similarity matrix between visual and textual triplets with a VLA model, and subsequently propagate it to an instance-level similarity matrix. Furthermore, to equip VLA models with the ability of relationship understanding, we design a triplet-matching objective to fine-tune the VLA models on a collection of curated dataset containing abundant entity relationships. Experiments demonstrate that our visual grounding performance increase of up to 19.5% over the SOTA zero-shot model on RefCOCO/+/g. On the more challenging Who's Waldo dataset, our zero-shot approach achieves comparable accuracy to the fully supervised model. Code is available at https://github.com/Show-han/Zeroshot_REC.
Authors: Zelin Peng, Zhengqin Xu, Zhilin Zeng, Lingxi Xie, Qi Tian, Wei Shen
Abstract: Parameter-efficient fine-tuning (PEFT) is an effective methodology to unleash the potential of large foundation models in novel scenarios with limited training data. In the computer vision community, PEFT has shown effectiveness in image classification, but little research has studied its ability for image segmentation. Fine-tuning segmentation models usually require a heavier adjustment of parameters to align the proper projection directions in the parameter space for new scenarios. This raises a challenge to existing PEFT algorithms, as they often inject a limited number of individual parameters into each block, which prevents substantial adjustment of the projection direction of the parameter space due to the limitation of Hidden Markov Chain along blocks. In this paper, we equip PEFT with a cross-block orchestration mechanism to enable the adaptation of the Segment Anything Model (SAM) to various downstream scenarios. We introduce a novel inter-block communication module, which integrates a learnable relation matrix to facilitate communication among different coefficient sets of each PEFT block's parameter space. Moreover, we propose an intra-block enhancement module, which introduces a linear projection head whose weights are generated from a hyper-complex layer, further enhancing the impact of the adjustment of projection directions on the entire parameter space. Extensive experiments on diverse benchmarks demonstrate that our proposed approach consistently improves the segmentation performance significantly on novel scenarios with only around 1K additional parameters.
Authors: Arthur Moreau, Jifei Song, Helisa Dhamo, Richard Shaw, Yiren Zhou, Eduardo P\'erez-Pellitero
Abstract: This work addresses the problem of real-time rendering of photorealistic human body avatars learned from multi-view videos. While the classical approaches to model and render virtual humans generally use a textured mesh, recent research has developed neural body representations that achieve impressive visual quality. However, these models are difficult to render in real-time and their quality degrades when the character is animated with body poses different than the training observations. We propose an animatable human model based on 3D Gaussian Splatting, that has recently emerged as a very efficient alternative to neural radiance fields. The body is represented by a set of gaussian primitives in a canonical space which is deformed with a coarse to fine approach that combines forward skinning and local non-rigid refinement. We describe how to learn our Human Gaussian Splatting (HuGS) model in an end-to-end fashion from multi-view observations, and evaluate it against the state-of-the-art approaches for novel pose synthesis of clothed body. Our method achieves 1.5 dB PSNR improvement over the state-of-the-art on THuman4 dataset while being able to render in real-time (80 fps for 512x512 resolution).
Authors: Hang Li, Chengzhi Shen, Philip Torr, Volker Tresp, Jindong Gu
Abstract: Diffusion-based models have gained significant popularity for text-to-image generation due to their exceptional image-generation capabilities. A risk with these models is the potential generation of inappropriate content, such as biased or harmful images. However, the underlying reasons for generating such undesired content from the perspective of the diffusion model's internal representation remain unclear. Previous work interprets vectors in an interpretable latent space of diffusion models as semantic concepts. However, existing approaches cannot discover directions for arbitrary concepts, such as those related to inappropriate concepts. In this work, we propose a novel self-supervised approach to find interpretable latent directions for a given concept. With the discovered vectors, we further propose a simple approach to mitigate inappropriate generation. Extensive experiments have been conducted to verify the effectiveness of our mitigation approach, namely, for fair generation, safe generation, and responsible text-enhancing generation. Project page: \url{https://interpretdiffusion.github.io}.
Authors: Sumanth Udupa, Prajwal Gurunath, Aniruddh Sikdar, Suresh Sundaram
Abstract: Deep neural networks have shown exemplary performance on semantic scene understanding tasks on source domains, but due to the absence of style diversity during training, enhancing performance on unseen target domains using only single source domain data remains a challenging task. Generation of simulated data is a feasible alternative to retrieving large style-diverse real-world datasets as it is a cumbersome and budget-intensive process. However, the large domain-specfic inconsistencies between simulated and real-world data pose a significant generalization challenge in semantic segmentation. In this work, to alleviate this problem, we propose a novel MultiResolution Feature Perturbation (MRFP) technique to randomize domain-specific fine-grained features and perturb style of coarse features. Our experimental results on various urban-scene segmentation datasets clearly indicate that, along with the perturbation of style-information, perturbation of fine-feature components is paramount to learn domain invariant robust feature maps for semantic segmentation models. MRFP is a simple and computationally efficient, transferable module with no additional learnable parameters or objective functions, that helps state-of-the-art deep neural networks to learn robust domain invariant features for simulation-to-real semantic segmentation.
Authors: Tongjia Chen, Hongshan Yu, Zhengeng Yang, Zechuan Li, Wei Sun, Chen Chen
Abstract: Due to the resource-intensive nature of training vision-language models on expansive video data, a majority of studies have centered on adapting pre-trained image-language models to the video domain. Dominant pipelines propose to tackle the visual discrepancies with additional temporal learners while overlooking the substantial discrepancy for web-scaled descriptive narratives and concise action category names, leading to less distinct semantic space and potential performance limitations. In this work, we prioritize the refinement of text knowledge to facilitate generalizable video recognition. To address the limitations of the less distinct semantic space of category names, we prompt a large language model (LLM) to augment action class names into Spatio-Temporal Descriptors thus bridging the textual discrepancy and serving as a knowledge base for general recognition. Moreover, to assign the best descriptors with different video instances, we propose Optimal Descriptor Solver, forming the video recognition problem as solving the optimal matching flow across frame-level representations and descriptors. Comprehensive evaluations in zero-shot, few-shot, and fully supervised video recognition highlight the effectiveness of our approach. Our best model achieves a state-of-the-art zero-shot accuracy of 75.1% on Kinetics-600.
Authors: Shuhuai Ren, Linli Yao, Shicheng Li, Xu Sun, Lu Hou
Abstract: This work proposes TimeChat, a time-sensitive multimodal large language model specifically designed for long video understanding. Our model incorporates two key architectural contributions: (1) a timestamp-aware frame encoder that binds visual content with the timestamp of each frame, and (2) a sliding video Q-Former that produces a video token sequence of varying lengths to accommodate videos of various durations. Additionally, we construct an instruction-tuning dataset, encompassing 6 tasks and a total of 125K instances, to further enhance TimeChat's instruction-following performance. Experiment results across various video understanding tasks, such as dense captioning, temporal grounding, and highlight detection, demonstrate TimeChat's strong zero-shot temporal localization and reasoning capabilities. For example, it achieves +9.2 F1 score and +2.8 CIDEr on YouCook2, +5.8 HIT@1 on QVHighlights, and +27.5 R@1 (IoU=0.5) on Charades-STA, compared to state-of-the-art video large language models, holding the potential to serve as a versatile video assistant for long-form video comprehension tasks and satisfy realistic user requirements.
Authors: Shenhan Qian, Tobias Kirschstein, Liam Schoneveld, Davide Davoli, Simon Giebenhain, Matthias Nie{\ss}ner
Abstract: We introduce GaussianAvatars, a new method to create photorealistic head avatars that are fully controllable in terms of expression, pose, and viewpoint. The core idea is a dynamic 3D representation based on 3D Gaussian splats that are rigged to a parametric morphable face model. This combination facilitates photorealistic rendering while allowing for precise animation control via the underlying parametric model, e.g., through expression transfer from a driving sequence or by manually changing the morphable model parameters. We parameterize each splat by a local coordinate frame of a triangle and optimize for explicit displacement offset to obtain a more accurate geometric representation. During avatar reconstruction, we jointly optimize for the morphable model parameters and Gaussian splat parameters in an end-to-end fashion. We demonstrate the animation capabilities of our photorealistic avatar in several challenging scenarios. For instance, we show reenactments from a driving video, where our method outperforms existing works by a significant margin.
Authors: Chandradeep Pokhariya, Ishaan N Shah, Angela Xing, Zekun Li, Kefan Chen, Avinash Sharma, Srinath Sridhar
Abstract: Understanding how we grasp objects with our hands has important applications in areas like robotics and mixed reality. However, this challenging problem requires accurate modeling of the contact between hands and objects. To capture grasps, existing methods use skeletons, meshes, or parametric models that does not represent hand shape accurately resulting in inaccurate contacts. We present MANUS, a method for Markerless Hand-Object Grasp Capture using Articulated 3D Gaussians. We build a novel articulated 3D Gaussians representation that extends 3D Gaussian splatting for high-fidelity representation of articulating hands. Since our representation uses Gaussian primitives, it enables us to efficiently and accurately estimate contacts between the hand and the object. For the most accurate results, our method requires tens of camera views that current datasets do not provide. We therefore build MANUS-Grasps, a new dataset that contains hand-object grasps viewed from 50+ cameras across 30+ scenes, 3 subjects, and comprising over 7M frames. In addition to extensive qualitative results, we also show that our method outperforms others on a quantitative contact evaluation method that uses paint transfer from the object to the hand.
Authors: Haithem Turki, Vasu Agrawal, Samuel Rota Bul\`o, Lorenzo Porzi, Peter Kontschieder, Deva Ramanan, Michael Zollh\"ofer, Christian Richardt
Abstract: Neural radiance fields provide state-of-the-art view synthesis quality but tend to be slow to render. One reason is that they make use of volume rendering, thus requiring many samples (and model queries) per ray at render time. Although this representation is flexible and easy to optimize, most real-world objects can be modeled more efficiently with surfaces instead of volumes, requiring far fewer samples per ray. This observation has spurred considerable progress in surface representations such as signed distance functions, but these may struggle to model semi-opaque and thin structures. We propose a method, HybridNeRF, that leverages the strengths of both representations by rendering most objects as surfaces while modeling the (typically) small fraction of challenging regions volumetrically. We evaluate HybridNeRF against the challenging Eyeful Tower dataset along with other commonly used view synthesis datasets. When comparing to state-of-the-art baselines, including recent rasterization-based approaches, we improve error rates by 15-30% while achieving real-time framerates (at least 36 FPS) for virtual-reality resolutions (2Kx2K).
Authors: Jialong Zuo, Hanyu Zhou, Ying Nie, Feng Zhang, Tianyu Guo, Nong Sang, Yunhe Wang, Changxin Gao
Abstract: Existing text-based person retrieval datasets often have relatively coarse-grained text annotations. This hinders the model to comprehend the fine-grained semantics of query texts in real scenarios. To address this problem, we contribute a new benchmark named \textbf{UFineBench} for text-based person retrieval with ultra-fine granularity. Firstly, we construct a new \textbf{dataset} named UFine6926. We collect a large number of person images and manually annotate each image with two detailed textual descriptions, averaging 80.8 words each. The average word count is three to four times that of the previous datasets. In addition of standard in-domain evaluation, we also propose a special \textbf{evaluation paradigm} more representative of real scenarios. It contains a new evaluation set with cross domains, cross textual granularity and cross textual styles, named UFine3C, and a new evaluation metric for accurately measuring retrieval ability, named mean Similarity Distribution (mSD). Moreover, we propose CFAM, a more efficient \textbf{algorithm} especially designed for text-based person retrieval with ultra fine-grained texts. It achieves fine granularity mining by adopting a shared cross-modal granularity decoder and hard negative match mechanism. With standard in-domain evaluation, CFAM establishes competitive performance across various datasets, especially on our ultra fine-grained UFine6926. Furthermore, by evaluating on UFine3C, we demonstrate that training on our UFine6926 significantly improves generalization to real scenarios compared with other coarse-grained datasets. The dataset and code will be made publicly available at \url{https://github.com/Zplusdragon/UFineBench}.
Authors: Ekkasit Pinyoanuntapong, Pu Wang, Minwoo Lee, Chen Chen
Abstract: Recent advances in text-to-motion generation using diffusion and autoregressive models have shown promising results. However, these models often suffer from a trade-off between real-time performance, high fidelity, and motion editability. To address this gap, we introduce MMM, a novel yet simple motion generation paradigm based on Masked Motion Model. MMM consists of two key components: (1) a motion tokenizer that transforms 3D human motion into a sequence of discrete tokens in latent space, and (2) a conditional masked motion transformer that learns to predict randomly masked motion tokens, conditioned on the pre-computed text tokens. By attending to motion and text tokens in all directions, MMM explicitly captures inherent dependency among motion tokens and semantic mapping between motion and text tokens. During inference, this allows parallel and iterative decoding of multiple motion tokens that are highly consistent with fine-grained text descriptions, therefore simultaneously achieving high-fidelity and high-speed motion generation. In addition, MMM has innate motion editability. By simply placing mask tokens in the place that needs editing, MMM automatically fills the gaps while guaranteeing smooth transitions between editing and non-editing parts. Extensive experiments on the HumanML3D and KIT-ML datasets demonstrate that MMM surpasses current leading methods in generating high-quality motion (evidenced by superior FID scores of 0.08 and 0.429), while offering advanced editing features such as body-part modification, motion in-betweening, and the synthesis of long motion sequences. In addition, MMM is two orders of magnitude faster on a single mid-range GPU than editable motion diffusion models. Our project page is available at \url{https://exitudio.github.io/MMM-page}.
Authors: Xinzhou Wang, Yikai Wang, Junliang Ye, Zhengyi Wang, Fuchun Sun, Pengkun Liu, Ling Wang, Kai Sun, Xintong Wang, Bin He
Abstract: Advances in 3D generation have facilitated sequential 3D model generation (a.k.a 4D generation), yet its application for animatable objects with large motion remains scarce. Our work proposes AnimatableDreamer, a text-to-4D generation framework capable of generating diverse categories of non-rigid objects on skeletons extracted from a monocular video. At its core, AnimatableDreamer is equipped with our novel optimization design dubbed Canonical Score Distillation (CSD), which lifts 2D diffusion for temporal consistent 4D generation. CSD, designed from a score gradient perspective, generates a canonical model with warp-robustness across different articulations. Notably, it also enhances the authenticity of bones and skinning by integrating inductive priors from a diffusion model. Furthermore, with multi-view distillation, CSD infers invisible regions, thereby improving the fidelity of monocular non-rigid reconstruction. Extensive experiments demonstrate the capability of our method in generating high-flexibility text-guided 3D models from the monocular video, while also showing improved reconstruction performance over existing non-rigid reconstruction methods.
Authors: Javier Tirado-Gar\'in, Javier Civera
Abstract: Estimating the relative camera pose from $n \geq 5$ correspondences between two calibrated views is a fundamental task in computer vision. This process typically involves two stages: 1) estimating the essential matrix between the views, and 2) disambiguating among the four candidate relative poses that satisfy the epipolar geometry. In this paper, we demonstrate a novel approach that, for the first time, bypasses the second stage. Specifically, we show that it is possible to directly estimate the correct relative camera pose from correspondences without needing a post-processing step to enforce the cheirality constraint on the correspondences. Building on recent advances in certifiable non-minimal optimization, we frame the relative pose estimation as a Quadratically Constrained Quadratic Program (QCQP). By applying the appropriate constraints, we ensure the estimation of a camera pose that corresponds to a valid 3D geometry and that is globally optimal when certified. We validate our method through exhaustive synthetic and real-world experiments, confirming the efficacy, efficiency and accuracy of the proposed approach. Code is available at https://github.com/javrtg/C2P.
Authors: Alexandros Graikos, Srikar Yellapragada, Minh-Quan Le, Saarthak Kapse, Prateek Prasanna, Joel Saltz, Dimitris Samaras
Abstract: To synthesize high-fidelity samples, diffusion models typically require auxiliary data to guide the generation process. However, it is impractical to procure the painstaking patch-level annotation effort required in specialized domains like histopathology and satellite imagery; it is often performed by domain experts and involves hundreds of millions of patches. Modern-day self-supervised learning (SSL) representations encode rich semantic and visual information. In this paper, we posit that such representations are expressive enough to act as proxies to fine-grained human labels. We introduce a novel approach that trains diffusion models conditioned on embeddings from SSL. Our diffusion models successfully project these features back to high-quality histopathology and remote sensing images. In addition, we construct larger images by assembling spatially consistent patches inferred from SSL embeddings, preserving long-range dependencies. Augmenting real data by generating variations of real images improves downstream classifier accuracy for patch-level and larger, image-scale classification tasks. Our models are effective even on datasets not encountered during training, demonstrating their robustness and generalizability. Generating images from learned embeddings is agnostic to the source of the embeddings. The SSL embeddings used to generate a large image can either be extracted from a reference image, or sampled from an auxiliary model conditioned on any related modality (e.g. class labels, text, genomic data). As proof of concept, we introduce the text-to-large image synthesis paradigm where we successfully synthesize large pathology and satellite images out of text descriptions.
Authors: Johannes S. Fischer, Ming Gui, Pingchuan Ma, Nick Stracke, Stefan A. Baumann, Bj\"orn Ommer
Abstract: Recently, there has been tremendous progress in visual synthesis and the underlying generative models. Here, diffusion models (DMs) stand out particularly, but lately, flow matching (FM) has also garnered considerable interest. While DMs excel in providing diverse images, they suffer from long training and slow generation. With latent diffusion, these issues are only partially alleviated. Conversely, FM offers faster training and inference but exhibits less diversity in synthesis. We demonstrate that introducing FM between the Diffusion model and the convolutional decoder offers high-resolution image synthesis with reduced computational cost and model size. Diffusion can then efficiently provide the necessary generation diversity. FM compensates for the lower resolution, mapping the small latent space to a high-dimensional one. Subsequently, the convolutional decoder of the LDM maps these latents to high-resolution images. By combining the diversity of DMs, the efficiency of FMs, and the effectiveness of convolutional decoders, we achieve state-of-the-art high-resolution image synthesis at $1024^2$ with minimal computational cost. Importantly, our approach is orthogonal to recent approximation and speed-up strategies for the underlying DMs, making it easily integrable into various DM frameworks.
Authors: Li Ma, Vasu Agrawal, Haithem Turki, Changil Kim, Chen Gao, Pedro Sander, Michael Zollh\"ofer, Christian Richardt
Abstract: Neural radiance fields have achieved remarkable performance in modeling the appearance of 3D scenes. However, existing approaches still struggle with the view-dependent appearance of glossy surfaces, especially under complex lighting of indoor environments. Unlike existing methods, which typically assume distant lighting like an environment map, we propose a learnable Gaussian directional encoding to better model the view-dependent effects under near-field lighting conditions. Importantly, our new directional encoding captures the spatially-varying nature of near-field lighting and emulates the behavior of prefiltered environment maps. As a result, it enables the efficient evaluation of preconvolved specular color at any 3D location with varying roughness coefficients. We further introduce a data-driven geometry prior that helps alleviate the shape radiance ambiguity in reflection modeling. We show that our Gaussian directional encoding and geometry prior significantly improve the modeling of challenging specular reflections in neural radiance fields, which helps decompose appearance into more physically meaningful components.
Authors: Geunhyuk Youk, Jihyong Oh, Munchurl Kim
Abstract: We present a joint learning scheme of video super-resolution and deblurring, called VSRDB, to restore clean high-resolution (HR) videos from blurry low-resolution (LR) ones. This joint restoration problem has drawn much less attention compared to single restoration problems. In this paper, we propose a novel flow-guided dynamic filtering (FGDF) and iterative feature refinement with multi-attention (FRMA), which constitutes our VSRDB framework, denoted as FMA-Net. Specifically, our proposed FGDF enables precise estimation of both spatio-temporally-variant degradation and restoration kernels that are aware of motion trajectories through sophisticated motion representation learning. Compared to conventional dynamic filtering, the FGDF enables the FMA-Net to effectively handle large motions into the VSRDB. Additionally, the stacked FRMA blocks trained with our novel temporal anchor (TA) loss, which temporally anchors and sharpens features, refine features in a course-to-fine manner through iterative updates. Extensive experiments demonstrate the superiority of the proposed FMA-Net over state-of-the-art methods in terms of both quantitative and qualitative quality. Codes and pre-trained models are available at: https://kaist-viclab.github.io/fmanet-site
Authors: Jiawei Wang, Kai Hu, Zhuoyao Zhong, Lei Sun, Qiang Huo
Abstract: Document structure analysis (aka document layout analysis) is crucial for understanding the physical layout and logical structure of documents, with applications in information retrieval, document summarization, knowledge extraction, etc. In this paper, we concentrate on Hierarchical Document Structure Analysis (HDSA) to explore hierarchical relationships within structured documents created using authoring software employing hierarchical schemas, such as LaTeX, Microsoft Word, and HTML. To comprehensively analyze hierarchical document structures, we propose a tree construction based approach that addresses multiple subtasks concurrently, including page object detection (Detect), reading order prediction of identified objects (Order), and the construction of intended hierarchical structure (Construct). We present an effective end-to-end solution based on this framework to demonstrate its performance. To assess our approach, we develop a comprehensive benchmark called Comp-HRDoc, which evaluates the above subtasks simultaneously. Our end-to-end system achieves state-of-the-art performance on two large-scale document layout analysis datasets (PubLayNet and DocLayNet), a high-quality hierarchical document structure reconstruction dataset (HRDoc), and our Comp-HRDoc benchmark. The Comp-HRDoc benchmark will be released to facilitate further research in this field.
Authors: Seokju Yun, Youngmin Ro
Abstract: Recently, efficient Vision Transformers have shown great performance with low latency on resource-constrained devices. Conventionally, they use 4x4 patch embeddings and a 4-stage structure at the macro level, while utilizing sophisticated attention with multi-head configuration at the micro level. This paper aims to address computational redundancy at all design levels in a memory-efficient manner. We discover that using larger-stride patchify stem not only reduces memory access costs but also achieves competitive performance by leveraging token representations with reduced spatial redundancy from the early stages. Furthermore, our preliminary analyses suggest that attention layers in the early stages can be substituted with convolutions, and several attention heads in the latter stages are computationally redundant. To handle this, we introduce a single-head attention module that inherently prevents head redundancy and simultaneously boosts accuracy by parallelly combining global and local information. Building upon our solutions, we introduce SHViT, a Single-Head Vision Transformer that obtains the state-of-the-art speed-accuracy tradeoff. For example, on ImageNet-1k, our SHViT-S4 is 3.3x, 8.1x, and 2.4x faster than MobileViTv2 x1.0 on GPU, CPU, and iPhone12 mobile device, respectively, while being 1.3% more accurate. For object detection and instance segmentation on MS COCO using Mask-RCNN head, our model achieves performance comparable to FastViT-SA12 while exhibiting 3.8x and 2.0x lower backbone latency on GPU and mobile device, respectively.
Authors: Zhengcong Fei, Mingyuan Fan, Changqian Yu, Junshi Huang
Abstract: This paper presents a new exploration into a category of diffusion models built upon state space architecture. We endeavor to train diffusion models for image data, wherein the traditional U-Net backbone is supplanted by a state space backbone, functioning on raw patches or latent space. Given its notable efficacy in accommodating long-range dependencies, Diffusion State Space Models (DiS) are distinguished by treating all inputs including time, condition, and noisy image patches as tokens. Our assessment of DiS encompasses both unconditional and class-conditional image generation scenarios, revealing that DiS exhibits comparable, if not superior, performance to CNN-based or Transformer-based U-Net architectures of commensurate size. Furthermore, we analyze the scalability of DiS, gauged by the forward pass complexity quantified in Gflops. DiS models with higher Gflops, achieved through augmentation of depth/width or augmentation of input tokens, consistently demonstrate lower FID. In addition to demonstrating commendable scalability characteristics, DiS-H/2 models in latent space achieve performance levels akin to prior diffusion models on class-conditional ImageNet benchmarks at the resolution of 256$\times$256 and 512$\times$512, while significantly reducing the computational burden. The code and models are available at: https://github.com/feizc/DiS.
Authors: Yanghao Zhang, Tianle Zhang, Ronghui Mu, Xiaowei Huang, Wenjie Ruan
Abstract: Although adversarial training (AT) has proven effective in enhancing the model's robustness, the recently revealed issue of fairness in robustness has not been well addressed, i.e. the robust accuracy varies significantly among different categories. In this paper, instead of uniformly evaluating the model's average class performance, we delve into the issue of robust fairness, by considering the worst-case distribution across various classes. We propose a novel learning paradigm, named Fairness-Aware Adversarial Learning (FAAL). As a generalization of conventional AT, we re-define the problem of adversarial training as a min-max-max framework, to ensure both robustness and fairness of the trained model. Specifically, by taking advantage of distributional robust optimization, our method aims to find the worst distribution among different categories, and the solution is guaranteed to obtain the upper bound performance with high probability. In particular, FAAL can fine-tune an unfair robust model to be fair within only two epochs, without compromising the overall clean and robust accuracies. Extensive experiments on various image datasets validate the superior performance and efficiency of the proposed FAAL compared to other state-of-the-art methods.
Authors: Hongxin Li, Zeyu Wang, Xu Yang, Yuran Yang, Shuqi Mei, Zhaoxiang Zhang
Abstract: Image-goal navigation is a challenging task that requires an agent to navigate to a goal indicated by an image in unfamiliar environments. Existing methods utilizing diverse scene memories suffer from inefficient exploration since they use all historical observations for decision-making without considering the goal-relevant fraction. To address this limitation, we present MemoNav, a novel memory model for image-goal navigation, which utilizes a working memory-inspired pipeline to improve navigation performance. Specifically, we employ three types of navigation memory. The node features on a map are stored in the short-term memory (STM), as these features are dynamically updated. A forgetting module then retains the informative STM fraction to increase efficiency. We also introduce long-term memory (LTM) to learn global scene representations by progressively aggregating STM features. Subsequently, a graph attention module encodes the retained STM and the LTM to generate working memory (WM) which contains the scene features essential for efficient navigation. The synergy among these three memory types boosts navigation performance by enabling the agent to learn and leverage goal-relevant scene features within a topological map. Our evaluation on multi-goal tasks demonstrates that MemoNav significantly outperforms previous methods across all difficulty levels in both Gibson and Matterport3D scenes. Qualitative results further illustrate that MemoNav plans more efficient routes.
Authors: Linwei Chen, Lin Gu, Ying Fu
Abstract: Dilated convolution, which expands the receptive field by inserting gaps between its consecutive elements, is widely employed in computer vision. In this study, we propose three strategies to improve individual phases of dilated convolution from the view of spectrum analysis. Departing from the conventional practice of fixing a global dilation rate as a hyperparameter, we introduce Frequency-Adaptive Dilated Convolution (FADC), which dynamically adjusts dilation rates spatially based on local frequency components. Subsequently, we design two plug-in modules to directly enhance effective bandwidth and receptive field size. The Adaptive Kernel (AdaKern) module decomposes convolution weights into low-frequency and high-frequency components, dynamically adjusting the ratio between these components on a per-channel basis. By increasing the high-frequency part of convolution weights, AdaKern captures more high-frequency components, thereby improving effective bandwidth. The Frequency Selection (FreqSelect) module optimally balances high- and low-frequency components in feature representations through spatially variant reweighting. It suppresses high frequencies in the background to encourage FADC to learn a larger dilation, thereby increasing the receptive field for an expanded scope. Extensive experiments on segmentation and object detection consistently validate the efficacy of our approach. The code is publicly available at \url{https://github.com/Linwei-Chen/FADC}.
Authors: Ramin Mousa, Mitra Khezli, Mohamadreza Azadi, Vahid Nikoofard, Saba Hesaraki
Abstract: Accurate classification of objects in 3D point clouds is a significant problem in several applications, such as autonomous navigation and augmented/virtual reality scenarios, which has become a research hot spot. In this paper, we presented a deep learning strategy for 3D object classification in augmented reality. The proposed approach is a combination of the GRU and LSTM. LSTM networks learn longer dependencies well, but due to the number of gates, it takes longer to train; on the other hand, GRU networks have a weaker performance than LSTM, but their training speed is much higher than GRU, which is The speed is due to its fewer gates. The proposed approach used the combination of speed and accuracy of these two networks. The proposed approach achieved an accuracy of 0.99 in the 4,499,0641 points dataset, which includes eight classes (unlabeled, man-made terrain, natural terrain, high vegetation, low vegetation, buildings, hardscape, scanning artifacts, cars). Meanwhile, the traditional machine learning approaches could achieve a maximum accuracy of 0.9489 in the best case. Keywords: Point Cloud Classification, Virtual Reality, Hybrid Model, GRULSTM, GRU, LSTM
Authors: Benjamin D. Killeen, Liam J. Wang, Han Zhang, Mehran Armand, Russell H. Taylor, Dave Dreizin, Greg Osgood, Mathias Unberath
Abstract: Automated X-ray image segmentation would accelerate research and development in diagnostic and interventional precision medicine. Prior efforts have contributed task-specific models capable of solving specific image analysis problems, but the utility of these models is restricted to their particular task domain, and expanding to broader use requires additional data, labels, and retraining efforts. Recently, foundation models (FMs) -- machine learning models trained on large amounts of highly variable data thus enabling broad applicability -- have emerged as promising tools for automated image analysis. Existing FMs for medical image analysis focus on scenarios and modalities where objects are clearly defined by visually apparent boundaries, such as surgical tool segmentation in endoscopy. X-ray imaging, by contrast, does not generally offer such clearly delineated boundaries or structure priors. During X-ray image formation, complex 3D structures are projected in transmission onto the imaging plane, resulting in overlapping features of varying opacity and shape. To pave the way toward an FM for comprehensive and automated analysis of arbitrary medical X-ray images, we develop FluoroSAM, a language-aligned variant of the Segment-Anything Model, trained from scratch on 1.6M synthetic X-ray images. FluoroSAM is trained on data including masks for 128 organ types and 464 non-anatomical objects, such as tools and implants. In real X-ray images of cadaveric specimens, FluoroSAM is able to segment bony anatomical structures based on text-only prompting with 0.51 and 0.79 DICE with point-based refinement, outperforming competing SAM variants for all structures. FluoroSAM is also capable of zero-shot generalization to segmenting classes beyond the training set thanks to its language alignment, which we demonstrate for full lung segmentation on real chest X-rays.
Authors: Yinan Deng, Jiahui Wang, Jingyu Zhao, Xinyu Tian, Guangyan Chen, Yi Yang, Yufeng Yue
Abstract: Environment representations endowed with sophisticated semantics are pivotal for facilitating seamless interaction between robots and humans, enabling them to effectively carry out various tasks. Open-vocabulary maps, powered by Visual-Language models (VLMs), possess inherent advantages, including zero-shot learning and support for open-set classes. However, existing open-vocabulary maps are primarily designed for small-scale environments, such as desktops or rooms, and are typically geared towards limited-area tasks involving robotic indoor navigation or in-place manipulation. They face challenges in direct generalization to outdoor environments characterized by numerous objects and complex tasks, owing to limitations in both understanding level and map structure. In this work, we propose OpenGraph, the first open-vocabulary hierarchical graph representation designed for large-scale outdoor environments. OpenGraph initially extracts instances and their captions from visual images, enhancing textual reasoning by encoding them. Subsequently, it achieves 3D incremental object-centric mapping with feature embedding by projecting images onto LiDAR point clouds. Finally, the environment is segmented based on lane graph connectivity to construct a hierarchical graph. Validation results from public dataset SemanticKITTI demonstrate that OpenGraph achieves the highest segmentation and query accuracy. The source code of OpenGraph is publicly available at https://github.com/BIT-DYN/OpenGraph.
Authors: Tengchuan Kou, Xiaohong Liu, Zicheng Zhang, Chunyi Li, Haoning Wu, Xiongkuo Min, Guangtao Zhai, Ning Liu
Abstract: With the rapid development of generative models, Artificial Intelligence-Generated Contents (AIGC) have exponentially increased in daily lives. Among them, Text-to-Video (T2V) generation has received widespread attention. Though many T2V models have been released for generating high perceptual quality videos, there is still lack of a method to evaluate the quality of these videos quantitatively. To solve this issue, we establish the largest-scale Text-to-Video Quality Assessment DataBase (T2VQA-DB) to date. The dataset is composed of 10,000 videos generated by 9 different T2V models. We also conduct a subjective study to obtain each video's corresponding mean opinion score. Based on T2VQA-DB, we propose a novel transformer-based model for subjective-aligned Text-to-Video Quality Assessment (T2VQA). The model extracts features from text-video alignment and video fidelity perspectives, then it leverages the ability of a large language model to give the prediction score. Experimental results show that T2VQA outperforms existing T2V metrics and SOTA video quality assessment models. Quantitative analysis indicates that T2VQA is capable of giving subjective-align predictions, validating its effectiveness. The dataset and code will be released at https://github.com/QMME/T2VQA.
Authors: Jing Zhang, Irving Fang, Juexiao Zhang, Hao Wu, Akshat Kaushik, Alice Rodriguez, Hanwen Zhao, Zhuo Zheng, Radu Iovita, Chen Feng
Abstract: Lithic Use-Wear Analysis (LUWA) using microscopic images is an underexplored vision-for-science research area. It seeks to distinguish the worked material, which is critical for understanding archaeological artifacts, material interactions, tool functionalities, and dental records. However, this challenging task goes beyond the well-studied image classification problem for common objects. It is affected by many confounders owing to the complex wear mechanism and microscopic imaging, which makes it difficult even for human experts to identify the worked material successfully. In this paper, we investigate the following three questions on this unique vision task for the first time:(i) How well can state-of-the-art pre-trained models (like DINOv2) generalize to the rarely seen domain? (ii) How can few-shot learning be exploited for scarce microscopic images? (iii) How do the ambiguous magnification and sensing modality influence the classification accuracy? To study these, we collaborated with archaeologists and built the first open-source and the largest LUWA dataset containing 23,130 microscopic images with different magnifications and sensing modalities. Extensive experiments show that existing pre-trained models notably outperform human experts but still leave a large gap for improvements. Most importantly, the LUWA dataset provides an underexplored opportunity for vision and learning communities and complements existing image classification problems on common objects.
Authors: Weipeng Deng, Runyu Ding, Jihan Yang, Jiahui Liu, Yijiang Li, Xiaojuan Qi, Edith Ngai
Abstract: Rapid advancements in 3D vision-language (3D-VL) tasks have opened up new avenues for human interaction with embodied agents or robots using natural language. Despite this progress, we find a notable limitation: existing 3D-VL models exhibit sensitivity to the styles of language input, struggling to understand sentences with the same semantic meaning but written in different variants. This observation raises a critical question: Can 3D vision-language models truly understand natural language? To test the language understandability of 3D-VL models, we first propose a language robustness task for systematically assessing 3D-VL models across various tasks, benchmarking their performance when presented with different language style variants. Importantly, these variants are commonly encountered in applications requiring direct interaction with humans, such as embodied robotics, given the diversity and unpredictability of human language. We propose a 3D Language Robustness Dataset, designed based on the characteristics of human language, to facilitate the systematic study of robustness. Our comprehensive evaluation uncovers a significant drop in the performance of all existing models across various 3D-VL tasks. Even the state-of-the-art 3D-LLM fails to understand some variants of the same sentences. Further in-depth analysis suggests that the existing models have a fragile and biased fusion module, which stems from the low diversity of the existing dataset. Finally, we propose a training-free module driven by LLM, which improves language robustness. Datasets and code will be available at github.
Authors: You Xie, Hongyi Xu, Guoxian Song, Chao Wang, Yichun Shi, Linjie Luo
Abstract: We propose X-Portrait, an innovative conditional diffusion model tailored for generating expressive and temporally coherent portrait animation. Specifically, given a single portrait as appearance reference, we aim to animate it with motion derived from a driving video, capturing both highly dynamic and subtle facial expressions along with wide-range head movements. As its core, we leverage the generative prior of a pre-trained diffusion model as the rendering backbone, while achieve fine-grained head pose and expression control with novel controlling signals within the framework of ControlNet. In contrast to conventional coarse explicit controls such as facial landmarks, our motion control module is learned to interpret the dynamics directly from the original driving RGB inputs. The motion accuracy is further enhanced with a patch-based local control module that effectively enhance the motion attention to small-scale nuances like eyeball positions. Notably, to mitigate the identity leakage from the driving signals, we train our motion control modules with scaling-augmented cross-identity images, ensuring maximized disentanglement from the appearance reference modules. Experimental results demonstrate the universal effectiveness of X-Portrait across a diverse range of facial portraits and expressive driving sequences, and showcase its proficiency in generating captivating portrait animations with consistently maintained identity characteristics.
Authors: Minzhou Pan, Zhengting Wang, Xin Dong, Vikash Sehwag, Lingjuan Lyu, Xue Lin
Abstract: In this paper, we propose WaterMark Detection (WMD), the first invisible watermark detection method under a black-box and annotation-free setting. WMD is capable of detecting arbitrary watermarks within a given reference dataset using a clean non-watermarked dataset as a reference, without relying on specific decoding methods or prior knowledge of the watermarking techniques. We develop WMD using foundations of offset learning, where a clean non-watermarked dataset enables us to isolate the influence of only watermarked samples in the reference dataset. Our comprehensive evaluations demonstrate the effectiveness of WMD, significantly outperforming naive detection methods, which only yield AUC scores around 0.5. In contrast, WMD consistently achieves impressive detection AUC scores, surpassing 0.9 in most single-watermark datasets and exceeding 0.7 in more challenging multi-watermark scenarios across diverse datasets and watermarking methods. As invisible watermarks become increasingly prevalent, while specific decoding techniques remain undisclosed, our approach provides a versatile solution and establishes a path toward increasing accountability, transparency, and trust in our digital visual content.
Authors: Junhong Zhao, Wei Ying, Yaoqiang Pan, Zhenfeng Yi, Chao Chen, Kewei Hu, Hanwen Kang
Abstract: Accurate collection of plant phenotyping is critical to optimising sustainable farming practices in precision agriculture. Traditional phenotyping in controlled laboratory environments, while valuable, falls short in understanding plant growth under real-world conditions. Emerging sensor and digital technologies offer a promising approach for direct phenotyping of plants in farm environments. This study investigates a learning-based phenotyping method using the Neural Radiance Field to achieve accurate in-situ phenotyping of pepper plants in greenhouse environments. To quantitatively evaluate the performance of this method, traditional point cloud registration on 3D scanning data is implemented for comparison. Experimental result shows that NeRF(Neural Radiance Fields) achieves competitive accuracy compared to the 3D scanning methods. The mean distance error between the scanner-based method and the NeRF-based method is 0.865mm. This study shows that the learning-based NeRF method achieves similar accuracy to 3D scanning-based methods but with improved scalability and robustness.
Authors: Xiaojun Hou, Jiazheng Xing, Yijie Qian, Yaowei Guo, Shuo Xin, Junhao Chen, Kai Tang, Mengmeng Wang, Zhengkai Jiang, Liang Liu, Yong Liu
Abstract: Multimodal Visual Object Tracking (VOT) has recently gained significant attention due to its robustness. Early research focused on fully fine-tuning RGB-based trackers, which was inefficient and lacked generalized representation due to the scarcity of multimodal data. Therefore, recent studies have utilized prompt tuning to transfer pre-trained RGB-based trackers to multimodal data. However, the modality gap limits pre-trained knowledge recall, and the dominance of the RGB modality persists, preventing the full utilization of information from other modalities. To address these issues, we propose a novel symmetric multimodal tracking framework called SDSTrack. We introduce lightweight adaptation for efficient fine-tuning, which directly transfers the feature extraction ability from RGB to other domains with a small number of trainable parameters and integrates multimodal features in a balanced, symmetric manner. Furthermore, we design a complementary masked patch distillation strategy to enhance the robustness of trackers in complex environments, such as extreme weather, poor imaging, and sensor failure. Extensive experiments demonstrate that SDSTrack outperforms state-of-the-art methods in various multimodal tracking scenarios, including RGB+Depth, RGB+Thermal, and RGB+Event tracking, and exhibits impressive results in extreme conditions. Our source code is available at https://github.com/hoqolo/SDSTrack.
Authors: Jie Tian, Lingxiao Yang, Ran Ji, Yuexin Ma, Lan Xu, Jingyi Yu, Ye Shi, Jingya Wang
Abstract: Gaze plays a crucial role in revealing human attention and intention, shedding light on the cognitive processes behind human actions. The integration of gaze guidance with the dynamics of hand-object interactions boosts the accuracy of human motion prediction. However, the lack of datasets that capture the intricate relationship and consistency among gaze, hand, and object movements remains a substantial hurdle. In this paper, we introduce the first Gaze-guided Hand-Object Interaction dataset, GazeHOI, and present a novel task for synthesizing gaze-guided hand-object interactions. Our dataset, GazeHOI, features simultaneous 3D modeling of gaze, hand, and object interactions, comprising 479 sequences with an average duration of 19.1 seconds, 812 sub-sequences, and 33 objects of various sizes. We propose a hierarchical framework centered on a gaze-guided hand-object interaction diffusion model, named GHO-Diffusion. In the pre-diffusion phase, we separate gaze conditions into spatial-temporal features and goal pose conditions at different levels of information granularity. During the diffusion phase, two gaze-conditioned diffusion models are stacked to simplify the complex synthesis of hand-object motions. Here, the object motion diffusion model generates sequences of object motions based on gaze conditions, while the hand motion diffusion model produces hand motions based on the generated object motion. To improve fine-grained goal pose alignment, we introduce a Spherical Gaussian constraint to guide the denoising step. In the subsequent post-diffusion phase, we optimize the generated hand motions using contact consistency. Our extensive experiments highlight the uniqueness of our dataset and the effectiveness of our approach.
Authors: Zhuowan Li, Bhavan Jasani, Peng Tang, Shabnam Ghadar
Abstract: Understanding data visualizations like charts and plots requires reasoning about both visual elements and numerics. Although strong in extractive questions, current chart visual question answering (chart VQA) models suffer on complex reasoning questions. In this work, we address the lack of reasoning ability by data augmentation. We leverage Large Language Models (LLMs), which have shown to have strong reasoning ability, as an automatic data annotator that generates question-answer annotations for chart images. The key innovation in our method lies in the Synthesize Step-by-Step strategy: our LLM-based data generator learns to decompose the complex question into step-by-step sub-questions (rationales), which are then used to derive the final answer using external tools, i.e. Python. This step-wise generation procedure is trained on synthetic data generated using a template-based QA generation pipeline. Experimental results highlight the significance of the proposed step-by-step generation. By training with the LLM-augmented data (LAMENDA), we significantly enhance the chart VQA models, achieving the state-of-the-art accuracy on the ChartQA and PlotQA datasets. In particular, our approach improves the accuracy of the previous state-of-the-art approach from 38% to 54% on the human-written questions in the ChartQA dataset, which needs strong reasoning. We hope our work underscores the potential of synthetic data and encourages further exploration of data augmentation using LLMs for reasoning-heavy tasks.
Authors: Patrick Grommelt, Louis Weiss, Franz-Josef Pfreundt, Janis Keuper
Abstract: The widespread adoption of generative image models has highlighted the urgent need to detect artificial content, which is a crucial step in combating widespread manipulation and misinformation. Consequently, numerous detectors and associated datasets have emerged. However, many of these datasets inadvertently introduce undesirable biases, thereby impacting the effectiveness and evaluation of detectors. In this paper, we emphasize that many datasets for AI-generated image detection contain biases related to JPEG compression and image size. Using the GenImage dataset, we demonstrate that detectors indeed learn from these undesired factors. Furthermore, we show that removing the named biases substantially increases robustness to JPEG compression and significantly alters the cross-generator performance of evaluated detectors. Specifically, it leads to more than 11 percentage points increase in cross-generator performance for ResNet50 and Swin-T detectors on the GenImage dataset, achieving state-of-the-art results. We provide the dataset and source codes of this paper on the anonymous website: https://www.unbiased-genimage.org
Authors: Maciej K Wozniak, Mattias Hansson, Marko Thiel, Patric Jensfelt
Abstract: In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains. Our code is open-source and will be available soon.
Authors: Qihang Fan, Quanzeng You, Xiaotian Han, Yongfei Liu, Yunzhe Tao, Huaibo Huang, Ran He, Hongxia Yang
Abstract: This paper tackles a significant challenge faced by Vision Transformers (ViTs): their constrained scalability across different image resolutions. Typically, ViTs experience a performance decline when processing resolutions different from those seen during training. Our work introduces two key innovations to address this issue. Firstly, we propose a novel module for dynamic resolution adjustment, designed with a single Transformer block, specifically to achieve highly efficient incremental token integration. Secondly, we introduce fuzzy positional encoding in the Vision Transformer to provide consistent positional awareness across multiple resolutions, thereby preventing overfitting to any single training resolution. Our resulting model, ViTAR (Vision Transformer with Any Resolution), demonstrates impressive adaptability, achieving 83.3\% top-1 accuracy at a 1120x1120 resolution and 80.4\% accuracy at a 4032x4032 resolution, all while reducing computational costs. ViTAR also shows strong performance in downstream tasks such as instance and semantic segmentation and can easily combined with self-supervised learning techniques like Masked AutoEncoder. Our work provides a cost-effective solution for enhancing the resolution scalability of ViTs, paving the way for more versatile and efficient high-resolution image processing.
Authors: Trong-Tung Nguyen, Duc-Anh Nguyen, Anh Tran, Cuong Pham
Abstract: Our work addresses limitations seen in previous approaches for object-centric editing problems, such as unrealistic results due to shape discrepancies and limited control in object replacement or insertion. To this end, we introduce FlexEdit, a flexible and controllable editing framework for objects where we iteratively adjust latents at each denoising step using our FlexEdit block. Initially, we optimize latents at test time to align with specified object constraints. Then, our framework employs an adaptive mask, automatically extracted during denoising, to protect the background while seamlessly blending new content into the target image. We demonstrate the versatility of FlexEdit in various object editing tasks and curate an evaluation test suite with samples from both real and synthetic images, along with novel evaluation metrics designed for object-centric editing. We conduct extensive experiments on different editing scenarios, demonstrating the superiority of our editing framework over recent advanced text-guided image editing methods. Our project page is published at https://flex-edit.github.io/.
Authors: Suraj Patni, Aradhye Agarwal, Chetan Arora
Abstract: In the absence of parallax cues, a learning-based single image depth estimation (SIDE) model relies heavily on shading and contextual cues in the image. While this simplicity is attractive, it is necessary to train such models on large and varied datasets, which are difficult to capture. It has been shown that using embeddings from pre-trained foundational models, such as CLIP, improves zero shot transfer in several applications. Taking inspiration from this, in our paper we explore the use of global image priors generated from a pre-trained ViT model to provide more detailed contextual information. We argue that the embedding vector from a ViT model, pre-trained on a large dataset, captures greater relevant information for SIDE than the usual route of generating pseudo image captions, followed by CLIP based text embeddings. Based on this idea, we propose a new SIDE model using a diffusion backbone which is conditioned on ViT embeddings. Our proposed design establishes a new state-of-the-art (SOTA) for SIDE on NYUv2 dataset, achieving Abs Rel error of 0.059(14% improvement) compared to 0.069 by the current SOTA (VPD). And on KITTI dataset, achieving Sq Rel error of 0.139 (2% improvement) compared to 0.142 by the current SOTA (GEDepth). For zero-shot transfer with a model trained on NYUv2, we report mean relative improvement of (20%, 23%, 81%, 25%) over NeWCRFs on (Sun-RGBD, iBims1, DIODE, HyperSim) datasets, compared to (16%, 18%, 45%, 9%) by ZoeDepth. The code is available at https://ecodepth-iitd.github.io
Authors: Taha Samavati, Mahdi Farvardin, Aboozar Ghaffari
Abstract: With the increasing use of smartphones in our daily lives, these devices have become capable of performing many complex tasks. Concerning the need for continuous monitoring of vital signs, especially for the elderly or those with certain types of diseases, the development of algorithms that can estimate vital signs using smartphones has attracted researchers worldwide. In particular, researchers have been exploring ways to estimate vital signs, such as heart rate, oxygen saturation levels, and respiratory rate, using algorithms that can be run on smartphones. However, many of these algorithms require multiple pre-processing steps that might introduce some implementation overheads or require the design of a couple of hand-crafted stages to obtain an optimal result. To address this issue, this research proposes a novel end-to-end solution to mobile-based vital sign estimation using deep learning that eliminates the need for pre-processing. By using a fully convolutional architecture, the proposed model has much fewer parameters and less computational complexity compared to the architectures that use fully-connected layers as the prediction heads. This also reduces the risk of overfitting. Additionally, a public dataset for vital sign estimation, which includes 62 videos collected from 35 men and 27 women, is provided. Overall, the proposed end-to-end approach promises significantly improved efficiency and performance for on-device health monitoring on readily available consumer electronics.
Authors: Tobias Schlosser, Frederik Beuth, Trixy Meyer, Arunodhayan Sampath Kumar, Gabriel Stolze, Olga Furashova, Katrin Engelmann, Danny Kowerko
Abstract: In ophthalmology, intravitreal operative medication therapy (IVOM) is a widespread treatment for diseases related to the age-related macular degeneration (AMD), the diabetic macular edema (DME), as well as the retinal vein occlusion (RVO). However, in real-world settings, patients often suffer from loss of vision on time scales of years despite therapy, whereas the prediction of the visual acuity (VA) and the earliest possible detection of deterioration under real-life conditions is challenging due to heterogeneous and incomplete data. In this contribution, we present a workflow for the development of a research-compatible data corpus fusing different IT systems of the department of ophthalmology of a German maximum care hospital. The extensive data corpus allows predictive statements of the expected progression of a patient and his or her VA in each of the three diseases. For the disease AMD, we found out a significant deterioration of the visual acuity over time. Within our proposed multistage system, we subsequently classify the VA progression into the three groups of therapy "winners", "stabilizers", and "losers" (WSL classification scheme). Our OCT biomarker classification using an ensemble of deep neural networks results in a classification accuracy (F1-score) of over 98 %, enabling us to complete incomplete OCT documentations while allowing us to exploit them for a more precise VA modelling process. Our VA prediction requires at least four VA examinations and optionally OCT biomarkers from the same time period to predict the VA progression within a forecasted time frame, whereas our prediction is currently restricted to IVOM / no therapy. We achieve a final prediction accuracy of 69 % in macro average F1-score, while being in the same range as the ophthalmologists with 57.8 and 50 +- 10.7 % F1-score.
Authors: Gaurav Kumar Nayak, Inder Khatri, Ruchit Rawal, Anirban Chakraborty
Abstract: Several companies often safeguard their trained deep models (i.e., details of architecture, learnt weights, training details etc.) from third-party users by exposing them only as black boxes through APIs. Moreover, they may not even provide access to the training data due to proprietary reasons or sensitivity concerns. In this work, we propose a novel defense mechanism for black box models against adversarial attacks in a data-free set up. We construct synthetic data via generative model and train surrogate network using model stealing techniques. To minimize adversarial contamination on perturbed samples, we propose 'wavelet noise remover' (WNR) that performs discrete wavelet decomposition on input images and carefully select only a few important coefficients determined by our 'wavelet coefficient selection module' (WCSM). To recover the high-frequency content of the image after noise removal via WNR, we further train a 'regenerator' network with an objective to retrieve the coefficients such that the reconstructed image yields similar to original predictions on the surrogate model. At test time, WNR combined with trained regenerator network is prepended to the black box network, resulting in a high boost in adversarial accuracy. Our method improves the adversarial accuracy on CIFAR-10 by 38.98% and 32.01% on state-of-the-art Auto Attack compared to baseline, even when the attacker uses surrogate architecture (Alexnet-half and Alexnet) similar to the black box architecture (Alexnet) with same model stealing strategy as defender. The code is available at https://github.com/vcl-iisc/data-free-black-box-defense
URLs: https://github.com/vcl-iisc/data-free-black-box-defense
Authors: Arsenii Senokosov, Alexandr Sedykh, Asel Sagingalieva, Basil Kyriacou, Alexey Melnikov
Abstract: Image classification, a pivotal task in multiple industries, faces computational challenges due to the burgeoning volume of visual data. This research addresses these challenges by introducing two quantum machine learning models that leverage the principles of quantum mechanics for effective computations. Our first model, a hybrid quantum neural network with parallel quantum circuits, enables the execution of computations even in the noisy intermediate-scale quantum era, where circuits with a large number of qubits are currently infeasible. This model demonstrated a record-breaking classification accuracy of 99.21% on the full MNIST dataset, surpassing the performance of known quantum-classical models, while having eight times fewer parameters than its classical counterpart. Also, the results of testing this hybrid model on a Medical MNIST (classification accuracy over 99%), and on CIFAR-10 (classification accuracy over 82%), can serve as evidence of the generalizability of the model and highlights the efficiency of quantum layers in distinguishing common features of input data. Our second model introduces a hybrid quantum neural network with a Quanvolutional layer, reducing image resolution via a convolution process. The model matches the performance of its classical counterpart, having four times fewer trainable parameters, and outperforms a classical model with equal weight parameters. These models represent advancements in quantum machine learning research and illuminate the path towards more accurate image classification systems.
Authors: Amin Ghafourian, Huanyi Shui, Devesh Upadhyay, Rajesh Gupta, Dimitar Filev, Iman Soltani Bozchalooi
Abstract: Autoencoders have been extensively used in the development of recent anomaly detection techniques. The premise of their application is based on the notion that after training the autoencoder on normal training data, anomalous inputs will exhibit a significant reconstruction error. Consequently, this enables a clear differentiation between normal and anomalous samples. In practice, however, it is observed that autoencoders can generalize beyond the normal class and achieve a small reconstruction error on some of the anomalous samples. To improve the performance, various techniques propose additional components and more sophisticated training procedures. In this work, we propose a remarkably straightforward alternative: instead of adding neural network components, involved computations, and cumbersome training, we complement the reconstruction loss with a computationally light term that regulates the norm of representations in the latent space. The simplicity of our approach minimizes the requirement for hyperparameter tuning and customization for new applications which, paired with its permissive data modality constraint, enhances the potential for successful adoption across a broad range of applications. We test the method on various visual and tabular benchmarks and demonstrate that the technique matches and frequently outperforms more complex alternatives. We further demonstrate that implementing this idea in the context of state-of-the-art methods can further improve their performance. We also provide a theoretical analysis and numerical simulations that help demonstrate the underlying process that unfolds during training and how it helps with anomaly detection. This mitigates the black-box nature of autoencoder-based anomaly detection algorithms and offers an avenue for further investigation of advantages, fail cases, and potential new directions.
Authors: Yiwen Zhang, Chuanpu Li, Liming Zhong, Zeli Chen, Wei Yang, Xuetao Wang
Abstract: Treatment planning, which is a critical component of the radiotherapy workflow, is typically carried out by a medical physicist in a time-consuming trial-and-error manner. Previous studies have proposed knowledge-based or deep-learning-based methods for predicting dose distribution maps to assist medical physicists in improving the efficiency of treatment planning. However, these dose prediction methods usually fail to effectively utilize distance information between surrounding tissues and targets or organs-at-risk (OARs). Moreover, they are poor at maintaining the distribution characteristics of ray paths in the predicted dose distribution maps, resulting in a loss of valuable information. In this paper, we propose a distance-aware diffusion model (DoseDiff) for precise prediction of dose distribution. We define dose prediction as a sequence of denoising steps, wherein the predicted dose distribution map is generated with the conditions of the computed tomography (CT) image and signed distance maps (SDMs). The SDMs are obtained by distance transformation from the masks of targets or OARs, which provide the distance from each pixel in the image to the outline of the targets or OARs. We further propose a multi-encoder and multi-scale fusion network (MMFNet) that incorporates multi-scale and transformer-based fusion modules to enhance information fusion between the CT image and SDMs at the feature level. We evaluate our model on two in-house datasets and a public dataset, respectively. The results demonstrate that our DoseDiff method outperforms state-of-the-art dose prediction methods in terms of both quantitative performance and visual quality.
Authors: Nishant Kumar, Lukas Krause, Thomas Wondrak, Sven Eckert, Kerstin Eckert, Stefan Gumhold
Abstract: Electrolysis is crucial for eco-friendly hydrogen production, but gas bubbles generated during the process hinder reactions, reduce cell efficiency, and increase energy consumption. Additionally, these gas bubbles cause changes in the conductivity inside the cell, resulting in corresponding variations in the induced magnetic field around the cell. Therefore, measuring these gas bubble-induced magnetic field fluctuations using external magnetic sensors and solving the inverse problem of Biot-Savart Law allows for estimating the conductivity in the cell and, thus, bubble size and location. However, determining high-resolution conductivity maps from only a few induced magnetic field measurements is an ill-posed inverse problem. To overcome this, we exploit Invertible Neural Networks (INNs) to reconstruct the conductivity field. Our qualitative results and quantitative evaluation using random error diffusion show that INN achieves far superior performance compared to Tikhonov regularization.
Authors: Hao Cheng, Jiahang Cao, Erjia Xiao, Mengshu Sun, Renjing Xu
Abstract: Deploying energy-efficient deep learning algorithms on computational-limited devices, such as robots, is still a pressing issue for real-world applications. Spiking Neural Networks (SNNs), a novel brain-inspired algorithm, offer a promising solution due to their low-latency and low-energy properties over traditional Artificial Neural Networks (ANNs). Despite their advantages, the dense structure of deep SNNs can still result in extra energy consumption. The Lottery Ticket Hypothesis (LTH) posits that within dense neural networks, there exist winning Lottery Tickets (LTs), namely sub-networks, that can be obtained without compromising performance. Inspired by this, this paper delves into the spiking-based LTs (SLTs), examining their unique properties and potential for extreme efficiency. Then, two significant sparse \textbf{\textit{Rewards}} are gained through comprehensive explorations and meticulous experiments on SLTs across various dense structures. Moreover, a sparse algorithm tailored for spiking transformer structure, which incorporates convolution operations into the Patch Embedding Projection (ConvPEP) module, has been proposed to achieve Multi-level Sparsity (MultiSp). MultiSp refers to (1) Patch number sparsity; (2) ConvPEP weights sparsity and binarization; and (3) ConvPEP activation layer binarization. Extensive experiments demonstrate that our method achieves extreme sparsity with only a slight performance decrease, paving the way for deploying energy-efficient neural networks in robotics and beyond.
Authors: Miao Zhu, Qiming Fu, Bo Liu, Mengxi Zhang, Bojian Li, Xiaoyan Luo, Fugen Zhou
Abstract: Radiotherapy is one of the primary treatment methods for tumors, but the organ movement caused by respiration limits its accuracy. Recently, 3D imaging from a single X-ray projection has received extensive attention as a promising approach to address this issue. However, current methods can only reconstruct 3D images without directly locating the tumor and are only validated for fixed-angle imaging, which fails to fully meet the requirements of motion control in radiotherapy. In this study, a novel imaging method RT-SRTS is proposed which integrates 3D imaging and tumor segmentation into one network based on multi-task learning (MTL) and achieves real-time simultaneous 3D reconstruction and tumor segmentation from a single X-ray projection at any angle. Furthermore, the attention enhanced calibrator (AEC) and uncertain-region elaboration (URE) modules have been proposed to aid feature extraction and improve segmentation accuracy. The proposed method was evaluated on fifteen patient cases and compared with three state-of-the-art methods. It not only delivers superior 3D reconstruction but also demonstrates commendable tumor segmentation results. Simultaneous reconstruction and segmentation can be completed in approximately 70 ms, significantly faster than the required time threshold for real-time tumor tracking. The efficacies of both AEC and URE have also been validated in ablation studies. The code of work is available at https://github.com/ZywooSimple/RT-SRTS.
Authors: Eli Verwimp, Rahaf Aljundi, Shai Ben-David, Matthias Bethge, Andrea Cossu, Alexander Gepperth, Tyler L. Hayes, Eyke H\"ullermeier, Christopher Kanan, Dhireesha Kudithipudi, Christoph H. Lampert, Martin Mundt, Razvan Pascanu, Adrian Popescu, Andreas S. Tolias, Joost van de Weijer, Bing Liu, Vincenzo Lomonaco, Tinne Tuytelaars, Gido M. van de Ven
Abstract: Continual learning is a subfield of machine learning, which aims to allow machine learning models to continuously learn on new data, by accumulating knowledge without forgetting what was learned in the past. In this work, we take a step back, and ask: "Why should one care about continual learning in the first place?". We set the stage by examining recent continual learning papers published at four major machine learning conferences, and show that memory-constrained settings dominate the field. Then, we discuss five open problems in machine learning, and even though they might seem unrelated to continual learning at first sight, we show that continual learning will inevitably be part of their solution. These problems are model editing, personalization and specialization, on-device learning, faster (re-)training and reinforcement learning. Finally, by comparing the desiderata from these unsolved problems and the current assumptions in continual learning, we highlight and discuss four future directions for continual learning research. We hope that this work offers an interesting perspective on the future of continual learning, while displaying its potential value and the paths we have to pursue in order to make it successful. This work is the result of the many discussions the authors had at the Dagstuhl seminar on Deep Continual Learning, in March 2023.
Authors: Zhixuan Liang, Yao Mu, Hengbo Ma, Masayoshi Tomizuka, Mingyu Ding, Ping Luo
Abstract: Diffusion models have demonstrated strong potential for robotic trajectory planning. However, generating coherent trajectories from high-level instructions remains challenging, especially for long-range composition tasks requiring multiple sequential skills. We propose SkillDiffuser, an end-to-end hierarchical planning framework integrating interpretable skill learning with conditional diffusion planning to address this problem. At the higher level, the skill abstraction module learns discrete, human-understandable skill representations from visual observations and language instructions. These learned skill embeddings are then used to condition the diffusion model to generate customized latent trajectories aligned with the skills. This allows generating diverse state trajectories that adhere to the learnable skills. By integrating skill learning with conditional trajectory generation, SkillDiffuser produces coherent behavior following abstract instructions across diverse tasks. Experiments on multi-task robotic manipulation benchmarks like Meta-World and LOReL demonstrate state-of-the-art performance and human-interpretable skill representations from SkillDiffuser. More visualization results and information could be found on our website.
Authors: Yuhta Takida, Yukara Ikemiya, Takashi Shibuya, Kazuki Shimada, Woosung Choi, Chieh-Hsin Lai, Naoki Murata, Toshimitsu Uesaka, Kengo Uchida, Wei-Hsiang Liao, Yuki Mitsufuji
Abstract: Vector quantization (VQ) is a technique to deterministically learn features with discrete codebook representations. It is commonly performed with a variational autoencoding model, VQ-VAE, which can be further extended to hierarchical structures for making high-fidelity reconstructions. However, such hierarchical extensions of VQ-VAE often suffer from the codebook/layer collapse issue, where the codebook is not efficiently used to express the data, and hence degrades reconstruction accuracy. To mitigate this problem, we propose a novel unified framework to stochastically learn hierarchical discrete representation on the basis of the variational Bayes framework, called hierarchically quantized variational autoencoder (HQ-VAE). HQ-VAE naturally generalizes the hierarchical variants of VQ-VAE, such as VQ-VAE-2 and residual-quantized VAE (RQ-VAE), and provides them with a Bayesian training scheme. Our comprehensive experiments on image datasets show that HQ-VAE enhances codebook usage and improves reconstruction performance. We also validated HQ-VAE in terms of its applicability to a different modality with an audio dataset.
Authors: Ningyu Zhang, Yunzhi Yao, Bozhong Tian, Peng Wang, Shumin Deng, Mengru Wang, Zekun Xi, Shengyu Mao, Jintian Zhang, Yuansheng Ni, Siyuan Cheng, Ziwen Xu, Xin Xu, Jia-Chen Gu, Yong Jiang, Pengjun Xie, Fei Huang, Lei Liang, Zhiqiang Zhang, Xiaowei Zhu, Jun Zhou, Huajun Chen
Abstract: Large Language Models (LLMs) have shown extraordinary capabilities in understanding and generating text that closely mirrors human communication. However, a primary limitation lies in the significant computational demands during training, arising from their extensive parameterization. This challenge is further intensified by the dynamic nature of the world, necessitating frequent updates to LLMs to correct outdated information or integrate new knowledge, thereby ensuring their continued relevance. Note that many applications demand continual model adjustments post-training to address deficiencies or undesirable behaviors. There is an increasing interest in efficient, lightweight methods for on-the-fly model modifications. To this end, recent years have seen a burgeoning in the techniques of knowledge editing for LLMs, which aim to efficiently modify LLMs' behaviors within specific domains while preserving overall performance across various inputs. In this paper, we first define the knowledge editing problem and then provide a comprehensive review of cutting-edge approaches. Drawing inspiration from educational and cognitive research theories, we propose a unified categorization criterion that classifies knowledge editing methods into three groups: resorting to external knowledge, merging knowledge into the model, and editing intrinsic knowledge. Furthermore, we introduce a new benchmark, KnowEdit, for a comprehensive empirical evaluation of representative knowledge editing approaches. Additionally, we provide an in-depth analysis of knowledge location, which can give a deeper understanding of the knowledge structures inherent within LLMs. Finally, we discuss several potential applications of knowledge editing, outlining its broad and impactful implications.
Authors: Qi Chen, Xiaoxi Chen, Haorui Song, Zhiwei Xiong, Alan Yuille, Chen Wei, Zongwei Zhou
Abstract: Tumor synthesis enables the creation of artificial tumors in medical images, facilitating the training of AI models for tumor detection and segmentation. However, success in tumor synthesis hinges on creating visually realistic tumors that are generalizable across multiple organs and, furthermore, the resulting AI models being capable of detecting real tumors in images sourced from different domains (e.g., hospitals). This paper made a progressive stride toward generalizable tumor synthesis by leveraging a critical observation: early-stage tumors (< 2cm) tend to have similar imaging characteristics in computed tomography (CT), whether they originate in the liver, pancreas, or kidneys. We have ascertained that generative AI models, e.g., Diffusion Models, can create realistic tumors generalized to a range of organs even when trained on a limited number of tumor examples from only one organ. Moreover, we have shown that AI models trained on these synthetic tumors can be generalized to detect and segment real tumors from CT volumes, encompassing a broad spectrum of patient demographics, imaging protocols, and healthcare facilities.
Authors: Ramy Farag, Parth Upadhyay, Yixiang Gao, Jacket Demby, Katherin Garces Montoya, Seyed Mohamad Ali Tousi, Gbenga Omotara, Guilherme DeSouza
Abstract: Manual analysis and diagnosis of COVID-19 through the examination of Computed Tomography (CT) images of the lungs can be time-consuming and result in errors, especially given high volume of patients and numerous images per patient. So, we address the need for automation of this task by developing a new deep learning model-based pipeline. Our motivation was sparked by the CVPR Workshop on "Domain Adaptation, Explainability and Fairness in AI for Medical Image Analysis", more specifically, the "COVID-19 Diagnosis Competition (DEF-AI-MIA COV19D)" under the same Workshop. This challenge provides an opportunity to assess our proposed pipeline for COVID-19 detection from CT scan images. The same pipeline incorporates the original EfficientNet, but with an added Attention Mechanism: EfficientNet-AM. Also, unlike the traditional/past pipelines, which relied on a pre-processing step, our pipeline takes the raw selected input images without any such step, except for an image-selection step to simply reduce the number of CT images required for training and/or testing. Moreover, our pipeline is computationally efficient, as, for example, it does not incorporate a decoder for segmenting the lungs. It also does not combine different backbones nor combine RNN with a backbone, as other pipelines in the past did. Nevertheless, our pipeline still outperforms all approaches presented by other teams in last year's instance of the same challenge, at least based on the validation subset of the competition dataset.
Authors: Xinyu Shi, Zecheng Hao, Zhaofei Yu
Abstract: The remarkable success of Vision Transformers in Artificial Neural Networks (ANNs) has led to a growing interest in incorporating the self-attention mechanism and transformer-based architecture into Spiking Neural Networks (SNNs). While existing methods propose spiking self-attention mechanisms that are compatible with SNNs, they lack reasonable scaling methods, and the overall architectures proposed by these methods suffer from a bottleneck in effectively extracting local features. To address these challenges, we propose a novel spiking self-attention mechanism named Dual Spike Self-Attention (DSSA) with a reasonable scaling method. Based on DSSA, we propose a novel spiking Vision Transformer architecture called SpikingResformer, which combines the ResNet-based multi-stage architecture with our proposed DSSA to improve both performance and energy efficiency while reducing parameters. Experimental results show that SpikingResformer achieves higher accuracy with fewer parameters and lower energy consumption than other spiking Vision Transformer counterparts. Notably, our SpikingResformer-L achieves 79.40% top-1 accuracy on ImageNet with 4 time-steps, which is the state-of-the-art result in the SNN field.
Authors: Mengru Wang, Ningyu Zhang, Ziwen Xu, Zekun Xi, Shumin Deng, Yunzhi Yao, Qishen Zhang, Linyi Yang, Jindong Wang, Huajun Chen
Abstract: This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories with various powerful attack prompts and equips comprehensive metrics for systematic evaluation. We conduct experiments with several knowledge editing approaches, indicating that knowledge editing has the potential to efficiently detoxify LLMs with limited impact on general performance. Then, we propose a simple yet effective baseline, dubbed Detoxifying with Intraoperative Neural Monitoring (DINM), to diminish the toxicity of LLMs within a few tuning steps via only one instance. We further provide an in-depth analysis of the internal mechanism for various detoxify approaches, demonstrating that previous methods like SFT and DPO may merely suppress the activations of toxic parameters, while DINM mitigates the toxicity of the toxic parameters to a certain extent, making permanent adjustments. We hope that these insights could shed light on future work of developing detoxifying approaches and the underlying knowledge mechanisms of LLMs. Code and benchmark are available at https://github.com/zjunlp/EasyEdit.
Authors: Yushan Huang, Josh Millar, Yuxuan Long, Yuchen Zhao, Hamed Hadaddi
Abstract: The personalization of machine learning (ML) models to address data drift is a significant challenge in the context of Internet of Things (IoT) applications. Presently, most approaches focus on fine-tuning either the full base model or its last few layers to adapt to new data, while often neglecting energy costs. However, various types of data drift exist, and fine-tuning the full base model or the last few layers may not result in optimal performance in certain scenarios. We propose Target Block Fine-Tuning (TBFT), a low-energy adaptive personalization framework designed for resource-constrained devices. We categorize data drift and personalization into three types: input-level, feature-level, and output-level. For each type, we fine-tune different blocks of the model to achieve optimal performance with reduced energy costs. Specifically, input-, feature-, and output-level correspond to fine-tuning the front, middle, and rear blocks of the model. We evaluate TBFT on a ResNet model, three datasets, three different training sizes, and a Raspberry Pi. Compared with the $Block Avg$, where each block is fine-tuned individually and their performance improvements are averaged, TBFT exhibits an improvement in model accuracy by an average of 15.30% whilst saving 41.57% energy consumption on average compared with full fine-tuning.
Authors: Hager Radi Abdelwahed, M\'elisande Teng, David Rolnick
Abstract: To address the interlinked biodiversity and climate crises, we need an understanding of where species occur and how these patterns are changing. However, observational data on most species remains very limited, and the amount of data available varies greatly between taxonomic groups. We introduce the problem of predicting species occurrence patterns given (a) satellite imagery, and (b) known information on the occurrence of other species. To evaluate algorithms on this task, we introduce SatButterfly, a dataset of satellite images, environmental data and observational data for butterflies, which is designed to pair with the existing SatBird dataset of bird observational data. To address this task, we propose a general model, R-Tran, for predicting species occurrence patterns that enables the use of partial observational data wherever found. We find that R-Tran outperforms other methods in predicting species encounter rates with partial information both within a taxon (birds) and across taxa (birds and butterflies). Our approach opens new perspectives to leveraging insights from species with abundant data to other species with scarce data, by modelling the ecosystems in which they co-occur.
Authors: Jinpeng Lu, Jingyun Chen, Linghan Cai, Songhan Jiang, Yongbing Zhang
Abstract: Positron emission tomography (PET) combined with computed tomography (CT) imaging is routinely used in cancer diagnosis and prognosis by providing complementary information. Automatically segmenting tumors in PET/CT images can significantly improve examination efficiency. Traditional multi-modal segmentation solutions mainly rely on concatenation operations for modality fusion, which fail to effectively model the non-linear dependencies between PET and CT modalities. Recent studies have investigated various approaches to optimize the fusion of modality-specific features for enhancing joint representations. However, modality-specific encoders used in these methods operate independently, inadequately leveraging the synergistic relationships inherent in PET and CT modalities, for example, the complementarity between semantics and structure. To address these issues, we propose a Hierarchical Adaptive Interaction and Weighting Network termed H2ASeg to explore the intrinsic cross-modal correlations and transfer potential complementary information. Specifically, we design a Modality-Cooperative Spatial Attention (MCSA) module that performs intra- and inter-modal interactions globally and locally. Additionally, a Target-Aware Modality Weighting (TAMW) module is developed to highlight tumor-related features within multi-modal features, thereby refining tumor segmentation. By embedding these modules across different layers, H2ASeg can hierarchically model cross-modal correlations, enabling a nuanced understanding of both semantic and structural tumor features. Extensive experiments demonstrate the superiority of H2ASeg, outperforming state-of-the-art methods on AutoPet-II and Hecktor2022 benchmarks. The code is released at https://github.com/JinPLu/H2ASeg.
Authors: Meiqi Chen, Yixin Cao, Yan Zhang, Chaochao Lu
Abstract: Recent advancements in Large Language Models (LLMs) have facilitated the development of Multimodal LLMs (MLLMs). Despite their impressive capabilities, MLLMs often suffer from an over-reliance on unimodal biases (e.g., language bias and vision bias), leading to incorrect answers in complex multimodal tasks. To investigate this issue, we propose a causal framework to interpret the biases in Visual Question Answering (VQA) problems. Within our framework, we devise a causal graph to elucidate the predictions of MLLMs on VQA problems, and assess the causal effect of biases through an in-depth causal analysis. Motivated by the causal graph, we introduce a novel MORE dataset, consisting of 12,000 VQA instances. This dataset is designed to challenge MLLMs' abilities, necessitating multi-hop reasoning and the surmounting of unimodal biases. Furthermore, we propose two strategies to mitigate unimodal biases and enhance MLLMs' reasoning capabilities, including a Decompose-Verify-Answer (DeVA) framework for limited-access MLLMs and the refinement of open-source MLLMs through fine-tuning. Extensive quantitative and qualitative experiments offer valuable insights for future research. Our project page is at https://opencausalab.github.io/MORE.