Authors: Keenon Werling, Janelle Kaneda, Alan Tan, Rishi Agarwal, Six Skov, Tom Van Wouwe, Scott Uhlrich, Nicholas Bianco, Carmichael Ong, Antoine Falisse, Shardul Sapkota, Aidan Chandra, Joshua Carter, Ezio Preatoni, Benjamin Fregly, Jennifer Hicks, Scott Delp, C. Karen Liu
Abstract: While reconstructing human poses in 3D from inexpensive sensors has advanced significantly in recent years, quantifying the dynamics of human motion, including the muscle-generated joint torques and external forces, remains a challenge. Prior attempts to estimate physics from reconstructed human poses have been hampered by a lack of datasets with high-quality pose and force data for a variety of movements. We present the AddBiomechanics Dataset 1.0, which includes physically accurate human dynamics of 273 human subjects, over 70 hours of motion and force plate data, totaling more than 24 million frames. To construct this dataset, novel analytical methods were required, which are also reported here. We propose a benchmark for estimating human dynamics from motion using this dataset, and present several baseline results. The AddBiomechanics Dataset is publicly available at https://addbiomechanics.org/download_data.html.
Authors: Jiangyuan Guo, Wei Chen, Yuxuan Sun, Jialong Xu, Bo Ai
Abstract: Although semantic communication (SC) has shown its potential in efficiently transmitting multi-modal data such as text, speeches and images, SC for videos has focused primarily on pixel-level reconstruction. However, these SC systems may be suboptimal for downstream intelligent tasks. Moreover, SC systems without pixel-level video reconstruction present advantages by achieving higher bandwidth efficiency and real-time performance of various intelligent tasks. The difficulty in such system design lies in the extraction of task-related compact semantic representations and their accurate delivery over noisy channels. In this paper, we propose an end-to-end SC system for video question answering (VideoQA) tasks called VideoQA-SC. Our goal is to accomplish VideoQA tasks directly based on video semantics over noisy or fading wireless channels, bypassing the need for video reconstruction at the receiver. To this end, we develop a spatiotemporal semantic encoder for effective video semantic extraction, and a learning-based bandwidth-adaptive deep joint source-channel coding (DJSCC) scheme for efficient and robust video semantic transmission. Experiments demonstrate that VideoQA-SC outperforms traditional and advanced DJSCC-based SC systems that rely on video reconstruction at the receiver under a wide range of channel conditions and bandwidth constraints. In particular, when the signal-to-noise ratio is low, VideoQA-SC can improve the answer accuracy by 5.17% while saving almost 99.5% of the bandwidth at the same time, compared with the advanced DJSCC-based SC system. Our results show the great potential of task-oriented SC system design for video applications.
Authors: Hongkun Zhang, Zherong Pan, Congyi Zhang, Lifeng Zhu, Xifeng Gao
Abstract: The recent success of pre-trained diffusion models unlocks the possibility of the automatic generation of textures for arbitrary 3D meshes in the wild. However, these models are trained in the screen space, while converting them to a multi-view consistent texture image poses a major obstacle to the output quality. In this paper, we propose a novel method to enforce multi-view consistency. Our method is based on the observation that latent space in a pre-trained diffusion model is noised separately for each camera view, making it difficult to achieve multi-view consistency by directly manipulating the latent codes. Based on the celebrated Denoising Diffusion Implicit Models (DDIM) scheme, we propose to use an optimization-based color-fusion to enforce consistency and indirectly modify the latent codes by gradient back-propagation. Our method further relaxes the sequential dependency assumption among the camera views. By evaluating on a series of general 3D models, we find our simple approach improves consistency and overall quality of the generated textures as compared to competing state-of-the-arts. Our implementation is available at: https://github.com/Quantuman134/TexPainter
Authors: Yunlong Zhao, Xiaoheng Deng, Yijing Liu, Xinjun Pei, Jiazhi Xia, Wei Chen
Abstract: Model stealing (MS) involves querying and observing the output of a machine learning model to steal its capabilities. The quality of queried data is crucial, yet obtaining a large amount of real data for MS is often challenging. Recent works have reduced reliance on real data by using generative models. However, when high-dimensional query data is required, these methods are impractical due to the high costs of querying and the risk of model collapse. In this work, we propose using sample gradients (SG) to enhance the utility of each real sample, as SG provides crucial guidance on the decision boundaries of the victim model. However, utilizing SG in the model stealing scenario faces two challenges: 1. Pixel-level gradient estimation requires extensive query volume and is susceptible to defenses. 2. The estimation of sample gradients has a significant variance. This paper proposes Superpixel Sample Gradient stealing (SPSG) for model stealing under the constraint of limited real samples. With the basic idea of imitating the victim model's low-variance patch-level gradients instead of pixel-level gradients, SPSG achieves efficient sample gradient estimation through two steps. First, we perform patch-wise perturbations on query images to estimate the average gradient in different regions of the image. Then, we filter the gradients through a threshold strategy to reduce variance. Exhaustive experiments demonstrate that, with the same number of real samples, SPSG achieves accuracy, agreements, and adversarial success rate significantly surpassing the current state-of-the-art MS methods. Codes are available at https://github.com/zyl123456aB/SPSG_attack.
Authors: Jun Zhou, Yaoshun Li, Hongchen Tan, Mingjie Wang, Nannan Li, Xiuping Liu
Abstract: In recent years, point cloud normal estimation, as a classical and foundational algorithm, has garnered extensive attention in the field of 3D geometric processing. Despite the remarkable performance achieved by current Neural Network-based methods, their robustness is still influenced by the quality of training data and the models' performance. In this study, we designed a fundamental framework for normal estimation, enhancing existing model through the incorporation of global information and various constraint mechanisms. Additionally, we employed a confidence-based strategy to select the reasonable samples for fair and robust network training. The introduced sample confidence can be integrated into the loss function to balance the influence of different samples on model training. Finally, we utilized existing orientation methods to correct estimated non-oriented normals, achieving state-of-the-art performance in both oriented and non-oriented tasks. Extensive experimental results demonstrate that our method works well on the widely used benchmarks.
Authors: Mohammad Farzanullah, Han Zhang, Akram Bin Sediq, Ali Afana, Melike Erol-Kantarci
Abstract: Integrated sensing and communications is a key enabler for the 6G wireless communication systems. The multiple sensing modalities will allow the base station to have a more accurate representation of the environment, leading to context-aware communications. Some widely equipped sensors such as cameras and RADAR sensors can provide some environmental perceptions. However, they are not enough to generate precise environmental representations, especially in adverse weather conditions. On the other hand, the LiDAR sensors provide more accurate representations, however, their widespread adoption is hindered by their high cost. This paper proposes a novel approach to enhance the wireless communication systems by synthesizing LiDAR point clouds from images and RADAR data. Specifically, it uses a multimodal transformer architecture and pre-trained encoding models to enable an accurate LiDAR generation. The proposed framework is evaluated on the DeepSense 6G dataset, which is a real-world dataset curated for context-aware wireless applications. Our results demonstrate the efficacy of the proposed approach in accurately generating LiDAR point clouds. We achieve a modified mean squared error of 10.3931. Visual examination of the images indicates that our model can successfully capture the majority of structures present in the LiDAR point cloud for diverse environments. This will enable the base stations to achieve more precise environmental sensing. By integrating LiDAR synthesis with existing sensing modalities, our method can enhance the performance of various wireless applications, including beam and blockage prediction.
Authors: Peng Xu, Qi Gao, Ying-Jie Wu
Abstract: Feature extraction is a critical technology to realize the automatic transmission of feature information throughout product life cycles. As CAD models primarily capture the 3D geometry of products, feature extraction heavily relies on geometric information. However, existing feature extraction methods often yield inaccurate outcomes due to the diverse interpretations of geometric information. This report presents a set-based feature extraction approach to address this uncertainty issue. Unlike existing methods that seek accurate feature results, our approach aims to transform the uncertainty of geometric information into a set of feature subgraphs. First, we define the convexity of basic geometric entities and introduce the concept of two-level attributed adjacency graphs. Second, a feature extraction workflow is designed to determine feature boundaries and identify feature subgraphs from CAD models. This set of feature subgraphs can be used for further feature recognition. A feature extraction system is programmed using C++ and UG/Open to demonstrate the feasibility of our proposed approach.
Authors: Zuo-Liang Zhu, Beibei Wang, Jian Yang
Abstract: 3D Gaussian Splatting (3DGS) has shown a powerful capability for novel view synthesis due to its detailed expressive ability and highly efficient rendering speed. Unfortunately, creating relightable 3D assets with 3DGS is still problematic, particularly for reflective objects, as its discontinuous representation raises difficulties in constraining geometries. Inspired by previous works, the signed distance field (SDF) can serve as an effective way for geometry regularization. However, a direct incorporation between Gaussians and SDF significantly slows training. To this end, we propose GS-ROR for reflective objects relighting with 3DGS aided by SDF priors. At the core of our method is the mutual supervision of the depth and normal between deferred Gaussians and SDF, which avoids the expensive volume rendering of SDF. Thanks to this mutual supervision, the learned deferred Gaussians are well-constrained with a minimal time cost. As the Gaussians are rendered in a deferred shading mode, while the alpha-blended Gaussians are smooth, individual Gaussians may still be outliers, yielding floater artifacts. Therefore, we further introduce an SDF-aware pruning strategy to remove Gaussian outliers, which are located distant from the surface defined by SDF, avoiding the floater issue. Consequently, our method outperforms the existing Gaussian-based inverse rendering methods in terms of relighting quality. Our method also exhibits competitive relighting quality compared to NeRF-based methods with at most 25% of training time and allows rendering at 200+ frames per second on an RTX4090.
Authors: Soumya Dutta, Faheem Nizar, Ahmad Amaan, Ayan Acharya
Abstract: Ubiquitous applications of Deep neural networks (DNNs) in different artificial intelligence systems have led to their adoption in solving challenging visualization problems in recent years. While sophisticated DNNs offer an impressive generalization, it is imperative to comprehend the quality, confidence, robustness, and uncertainty associated with their prediction. A thorough understanding of these quantities produces actionable insights that help application scientists make informed decisions. Unfortunately, the intrinsic design principles of the DNNs cannot beget prediction uncertainty, necessitating separate formulations for robust uncertainty-aware models for diverse visualization applications. To that end, this contribution demonstrates how the prediction uncertainty and sensitivity of DNNs can be estimated efficiently using various methods and then interactively compared and contrasted for deep image synthesis tasks. Our inspection suggests that uncertainty-aware deep visualization models generate illustrations of informative and superior quality and diversity. Furthermore, prediction uncertainty improves the robustness and interpretability of deep visualization models, making them practical and convenient for various scientific domains that thrive on visual analyses.
Authors: Xiaoyi Liu, Hongjie Qiu, Muqing Li, Zhou Yu, Yutian Yang, Yafeng Yan
Abstract: This paper introduces an innovative multi-modal fusion deep learning approach to overcome the drawbacks of traditional single-modal recognition techniques. These drawbacks include incomplete information and limited diagnostic accuracy. During the feature extraction stage, cutting-edge deep learning models including convolutional neural networks (CNN), recurrent neural networks (RNN), and transformers are applied to distill advanced features from image-based, temporal, and structured data sources. The fusion strategy component seeks to determine the optimal fusion mode tailored to the specific disease recognition task. In the experimental section, a comparison is made between the performance of the proposed multi-mode fusion model and existing single-mode recognition methods. The findings demonstrate significant advantages of the multimodal fusion model across multiple evaluation metrics.
Authors: Qian-Wei Wang, Yuqiu Xie, Letian Zhang, Zimo Liu, Shu-Tao Xia
Abstract: Pre-trained vision-language models learn massive data to model unified representations of images and natural languages, which can be widely applied to downstream machine learning tasks. In addition to zero-shot inference, in order to better adapt pre-trained models to the requirements of downstream tasks, people usually use methods such as few-shot or parameter-efficient fine-tuning and knowledge distillation. However, annotating samples is laborious, while a large number of unlabeled samples can be easily obtained. In this paper, we investigate a novel "pre-trained annotating - weakly-supervised learning" paradigm for pre-trained model application and experiment on image classification tasks. Specifically, based on CLIP, we annotate image samples with multiple prompt templates to obtain multiple candidate labels to form the noisy partial label dataset, and design a collaborative consistency regularization algorithm to solve this problem. Our method simultaneously trains two neural networks, which collaboratively purify training labels for each other and obtain pseudo-labels for self-training, while adopting prototypical similarity alignment and noisy supervised contrastive learning to optimize model representation. In experiments, our method achieves performances far beyond zero-shot inference without introducing additional label information, and outperforms other weakly supervised learning and few-shot fine-tuning methods, and obtains smaller deployed models. Our code is available at: \url{https://anonymous.4open.science/r/Co-Reg-8CF9}.
Authors: Songyin Wu, Deepak Vembar, Anton Sochenov, Selvakumar Panneer, Sungye Kim, Anton Kaplanyan, Ling-Qi Yan
Abstract: Real-time rendering has been embracing ever-demanding effects, such as ray tracing. However, rendering such effects in high resolution and high frame rate remains challenging. Frame extrapolation methods, which don't introduce additional latency as opposed to frame interpolation methods such as DLSS 3 and FSR 3, boost the frame rate by generating future frames based on previous frames. However, it is a more challenging task because of the lack of information in the disocclusion regions, and recent methods also have a high engine integration cost due to requiring G-buffers as input. We propose a \emph{G-buffer free} frame extrapolation, GFFE, with a novel heuristic framework and an efficient neural network, to plausibly generate new frames in real-time without introducing additional latency. We analyze the motion of dynamic fragments and different types of disocclusions, and design the corresponding modules of the extrapolation block to handle them. After filling disocclusions, a light-weight shading correction network is used to correct shading and improve overall quality. GFFE achieves comparable or better results compared to previous interpolation as well as G-buffer-dependent extrapolation methods, with more efficient performance and easier game integration.
Authors: Yingying Fang, Zihao Jin, Xiaodan Xing, Simon Walsh, Guang Yang
Abstract: In medical imaging, particularly in early disease detection and prognosis tasks, discerning the rationale behind an AI model's predictions is crucial for evaluating the reliability of its decisions. Conventional explanation methods face challenges in identifying discernible decisive features in medical image classifications, where discriminative features are subtle or not immediately apparent. To bridge this gap, we propose an explainable model that is equipped with both decision reasoning and feature identification capabilities. Our approach not only detects influential image patterns but also uncovers the decisive features that drive the model's final predictions. By implementing our method, we can efficiently identify and visualise class-specific features leveraged by the data-driven model, providing insights into the decision-making processes of deep learning models. We validated our model in the demanding realm of medical prognosis task, demonstrating its efficacy and potential in enhancing the reliability of AI in healthcare and in discovering new knowledge in diseases where prognostic understanding is limited.
Authors: Qiang Guo
Abstract: Pedestrian detection has been a hot spot in computer vision over the past decades due to the wide spectrum of promising applications, the major challenge of which is False Positives (FPs) that occur during pedestrian detection. The emergence various Convolutional Neural Network-based detection strategies substantially enhance the pedestrian detection accuracy but still not well solve this problem. This paper deeply analysis the detection framework of the two-stage CNN detection methods and find out false positives in detection results is due to its training strategy miss classify some false proposals, thus weakens the classification capability of following subnetwork and hardly to suppress false ones. To solve this problem, This paper proposes a pedestrian-sensitive training algorithm to effectively help two-stage CNN detection methods learn to distinguish the pedestrian and non-pedestrian samples and suppress the false positives in final detection results. The core of the proposed training algorithm is to redesign the training proposal generating pipeline of the two-stage CNN detection methods, which can avoid a certain number of false ones that mislead its training process. With the help of the proposed algorithm, the detection accuracy of the MetroNext, an smaller and accurate metro passenger detector, is further improved, which further decreases false ones in its metro passengers detection results. Based on various challenging benchmark datasets, experiment results have demonstrated that feasibility of the proposed algorithm to improve pedestrian detection accuracy by removing the false positives. Compared with the competitors, MetroNext-PST demonstrates better overall prediction performance in accuracy, total number of parameters, and inference time, thus it can become a practical solution for hunting pedestrian tailored for mobile and edge devices.
Authors: Luis Miguel Pazos-Out\'on, Cristina Nader Vasconcelos, Anton Raichuk, Anurag Arnab, Dan Morris, Maxim Neumann
Abstract: Protecting and restoring forest ecosystems is critical for biodiversity conservation and carbon sequestration. Forest monitoring on a global scale is essential for prioritizing and assessing conservation efforts. Satellite-based remote sensing is the only viable solution for providing global coverage, but to date, large-scale forest monitoring is limited to single modalities and single time points. In this paper, we present a dataset consisting of data from five public satellites for recognizing forest plantations and planted tree species across the globe. Each satellite modality consists of a multi-year time series. The dataset, named \PlantD, includes over 2M examples of 64 tree label classes (46 genera and 40 species), distributed among 41 countries. This dataset is released to foster research in forest monitoring using multimodal, multi-scale, multi-temporal data sources. Additionally, we present initial baseline results and evaluate modality fusion and data augmentation approaches for this dataset.
Authors: Tengbo Wang, Yu Bai
Abstract: How to extract instance-level masks without instance-level supervision is the main challenge of weakly supervised instance segmentation (WSIS). Popular WSIS methods estimate a displacement field (DF) via learning inter-pixel relations and perform clustering to identify instances. However, the resulting instance centroids are inherently unstable and vary significantly across different clustering algorithms. In this paper, we propose Boundary-Assisted Instance Segmentation (BAISeg), which is a novel paradigm for WSIS that realizes instance segmentation with pixel-level annotations. BAISeg comprises an instance-aware boundary detection (IABD) branch and a semantic segmentation branch. The IABD branch identifies instances by predicting class-agnostic instance boundaries rather than instance centroids, therefore, it is different from previous DF-based approaches. In particular, we proposed the Cascade Fusion Module (CFM) and the Deep Mutual Attention (DMA) in the IABD branch to obtain rich contextual information and capture instance boundaries with weak responses. During the training phase, we employed Pixel-to-Pixel Contrast to enhance the discriminative capacity of the IABD branch. This further strengthens the continuity and closedness of the instance boundaries. Extensive experiments on PASCAL VOC 2012 and MS COCO demonstrate the effectiveness of our approach, and we achieve considerable performance with only pixel-level annotations. The code will be available at https://github.com/wsis-seg/BAISeg.
Authors: Yongmin Lee, Hye Won Chung
Abstract: Dataset distillation aims to synthesize a small number of images per class (IPC) from a large dataset to approximate full dataset training with minimal performance loss. While effective in very small IPC ranges, many distillation methods become less effective, even underperforming random sample selection, as IPC increases. Our examination of state-of-the-art trajectory-matching based distillation methods across various IPC scales reveals that these methods struggle to incorporate the complex, rare features of harder samples into the synthetic dataset even with the increased IPC, resulting in a persistent coverage gap between easy and hard test samples. Motivated by such observations, we introduce SelMatch, a novel distillation method that effectively scales with IPC. SelMatch uses selection-based initialization and partial updates through trajectory matching to manage the synthetic dataset's desired difficulty level tailored to IPC scales. When tested on CIFAR-10/100 and TinyImageNet, SelMatch consistently outperforms leading selection-only and distillation-only methods across subset ratios from 5% to 30%.
Authors: Kimia Hamidieh, Haoran Zhang, Swami Sankaranarayanan, Marzyeh Ghassemi
Abstract: Supervised learning methods have been found to exhibit inductive biases favoring simpler features. When such features are spuriously correlated with the label, this can result in suboptimal performance on minority subgroups. Despite the growing popularity of methods which learn from unlabeled data, the extent to which these representations rely on spurious features for prediction is unclear. In this work, we explore the impact of spurious features on Self-Supervised Learning (SSL) for visual representation learning. We first empirically show that commonly used augmentations in SSL can cause undesired invariances in the image space, and illustrate this with a simple example. We further show that classical approaches in combating spurious correlations, such as dataset re-sampling during SSL, do not consistently lead to invariant representations. Motivated by these findings, we propose LateTVG to remove spurious information from these representations during pre-training, by regularizing later layers of the encoder via pruning. We find that our method produces representations which outperform the baselines on several benchmarks, without the need for group or label information during SSL.
Authors: Gabriel Moreira, Manuel Marques, Jo\~ao Paulo Costeira
Abstract: A cornerstone of geometric reconstruction, rotation averaging seeks the set of absolute rotations that optimally explains a set of measured relative orientations between them. In addition to being an integral part of bundle adjustment and structure-from-motion, the problem of synchronizing rotations also finds applications in visual simultaneous localization and mapping, where it is used as an initialization for iterative solvers, and camera network calibration. Nevertheless, this optimization problem is both non-convex and high-dimensional. In this paper, we address it from a maximum likelihood estimation standpoint and make a twofold contribution. Firstly, we set forth a novel primal-dual method, motivated by the widely accepted spectral initialization. Further, we characterize stationary points of rotation averaging in cycle graphs topologies and contextualize this result within spectral graph theory. We benchmark the proposed method in multiple settings and certify our solution via duality theory, achieving a significant gain in precision and performance.
Authors: Yufei Luo, Zhen Yang, Ru Zhang, Jianyi Liu
Abstract: Currently, most methods for text steganalysis are based on deep neural networks (DNNs). However, in real-life scenarios, obtaining a sufficient amount of labeled stego-text for correctly training networks using a large number of parameters is often challenging and costly. Additionally, due to a phenomenon known as dataset bias or domain shift, recognition models trained on a large dataset exhibit poor generalization performance on novel datasets and tasks. Therefore, to address the issues of missing labeled data and inadequate model generalization in text steganalysis, this paper proposes a cross-domain stego-text analysis method (PDTS) based on pseudo-labeling and domain adaptation (unsupervised learning). Specifically, we propose a model architecture combining pre-trained BERT with a single-layer Bi-LSTM to learn and extract generic features across tasks and generate task-specific representations. Considering the differential contributions of different features to steganalysis, we further design a feature filtering mechanism to achieve selective feature propagation, thereby enhancing classification performance. We train the model using labeled source domain data and adapt it to target domain data distribution using pseudo-labels for unlabeled target domain data through self-training. In the label estimation step, instead of using a static sampling strategy, we propose a progressive sampling strategy to gradually increase the number of selected pseudo-label candidates. Experimental results demonstrate that our method performs well in zero-shot text steganalysis tasks, achieving high detection accuracy even in the absence of labeled data in the target domain, and outperforms current zero-shot text steganalysis methods.
Authors: Ruchika Chavhan, Ondrej Bohdal, Yongshuo Zong, Da Li, Timothy Hospedales
Abstract: Large-scale text-to-image diffusion models excel in generating high-quality images from textual inputs, yet concerns arise as research indicates their tendency to memorize and replicate training data, raising We also addressed the issue of memorization in diffusion models, where models tend to replicate exact training samples raising copyright infringement and privacy issues. Efforts within the text-to-image community to address memorization explore causes such as data duplication, replicated captions, or trigger tokens, proposing per-prompt inference-time or training-time mitigation strategies. In this paper, we focus on the feed-forward layers and begin by contrasting neuron activations of a set of memorized and non-memorized prompts. Experiments reveal a surprising finding: many different sets of memorized prompts significantly activate a common subspace in the model, demonstrating, for the first time, that memorization in the diffusion models lies in a special subspace. Subsequently, we introduce a novel post-hoc method for editing pre-trained models, whereby memorization is mitigated through the straightforward pruning of weights in specialized subspaces, avoiding the need to disrupt the training or inference process as seen in prior research. Finally, we demonstrate the robustness of the pruned model against training data extraction attacks, thereby unveiling new avenues for a practical and one-for-all solution to memorization.
Authors: Nan Dou, Qi Shi, Zhigang Lian
Abstract: For the purpose of achieving a more precise definition and data analysis of images, this study conducted a research on vectorization and rasterization storage of electronic maps, focusing on a large underground parking garage map. During the research, image processing, vectorization and rasterization storage were performed. The paper proposed a method for the vectorization classification storage of indoor two-dimensional map raster data. This method involves converting raster data into vector data and classifying elements such as parking spaces, pathways, and obstacles based on their coordinate positions with the grid indexing method, thereby facilitating efficient storage and rapid querying of indoor maps. Additionally, interpolation algorithms were employed to extract vector data and convert it into raster data. Navigation testing was conducted to validate the accuracy and reliability of the map model under this method, providing effective technical support for the digital storage and navigation of garage maps.
Authors: M. Hosseinzadeh, P. Khoshaght, S. Sadeghi, P. Asghari, Z. Arabi, J. Lansky, P. Budinsky, A. Masoud Rahmani, S. W. Lee
Abstract: Acute lymphoblastic leukemia (ALL) severity is determined by the presence and ratios of blast cells (abnormal white blood cells) in both bone marrow and peripheral blood. Manual diagnosis of this disease is a tedious and time-consuming operation, making it difficult for professionals to accurately examine blast cell characteristics. To address this difficulty, researchers use deep learning and machine learning. In this paper, a ResNet-based feature extractor is utilized to detect ALL, along with a variety of feature selectors and classifiers. To get the best results, a variety of transfer learning models, including the Resnet, VGG, EfficientNet, and DensNet families, are used as deep feature extractors. Following extraction, different feature selectors are used, including Genetic algorithm, PCA, ANOVA, Random Forest, Univariate, Mutual information, Lasso, XGB, Variance, and Binary ant colony. After feature qualification, a variety of classifiers are used, with MLP outperforming the others. The recommended technique is used to categorize ALL and HEM in the selected dataset which is C-NMC 2019. This technique got an impressive 90.71% accuracy and 95.76% sensitivity for the relevant classifications, and its metrics on this dataset outperformed others.
Authors: Qi Qiu, Tao Zhu, Furong Duan, Kevin I-Kai Wang, Liming Chen, Mingxing Nie, Mingxing Nie
Abstract: Inertial Measurement Unit (IMU) sensors are widely employed for Human Activity Recognition (HAR) due to their portability, energy efficiency, and growing research interest. However, a significant challenge for IMU-HAR models is achieving robust generalization performance across diverse users. This limitation stems from substantial variations in data distribution among individual users. One primary reason for this distribution disparity lies in the representation of IMU sensor data in the local coordinate system, which is susceptible to subtle user variations during IMU wearing. To address this issue, we propose a novel approach that extracts a global view representation based on the characteristics of IMU data, effectively alleviating the data distribution discrepancies induced by wearing styles. To validate the efficacy of the global view representation, we fed both global and local view data into model for experiments. The results demonstrate that global view data significantly outperforms local view data in cross-user experiments. Furthermore, we propose a Multi-view Supervised Network (MVFNet) based on Shuffling to effectively fuse local view and global view data. It supervises the feature extraction of each view through view division and view shuffling, so as to avoid the model ignoring important features as much as possible. Extensive experiments conducted on OPPORTUNITY and PAMAP2 datasets demonstrate that the proposed algorithm outperforms the current state-of-the-art methods in cross-user HAR.
Authors: Lucas Mahler, Julius Steiglechner, Benjamin Bender, Tobias Lindig, Dana Ramadan, Jonas Bause, Florian Birk, Rahel Heule, Edyta Charyasz, Michael Erb, Vinod Jangir Kumar, Gisela E Hagberg, Pascal Martin, Gabriele Lohmann, Klaus Scheffler
Abstract: The UltraCortex repository (https://www.ultracortex.org) houses magnetic resonance imaging data of the human brain obtained at an ultra-high field strength of 9.4 T. It contains 86 structural MR images with spatial resolutions ranging from 0.6 to 0.8 mm. Additionally, the repository includes segmentations of 12 brains into gray and white matter compartments. These segmentations have been independently validated by two expert neuroradiologists, thus establishing them as a reliable gold standard. This resource provides researchers with access to high-quality brain imaging data and validated segmentations, facilitating neuroimaging studies and advancing our understanding of brain structure and function. Existing repositories do not accommodate field strengths beyond 7 T, nor do they offer validated segmentations, underscoring the significance of this new resource.
Authors: Ling Li, Yu Ye, Bingchuan Jiang, Wei Zeng
Abstract: This work tackles the problem of geo-localization with a new paradigm using a large vision-language model (LVLM) augmented with human inference knowledge. A primary challenge here is the scarcity of data for training the LVLM - existing street-view datasets often contain numerous low-quality images lacking visual clues, and lack any reasoning inference. To address the data-quality issue, we devise a CLIP-based network to quantify the degree of street-view images being locatable, leading to the creation of a new dataset comprising highly locatable street views. To enhance reasoning inference, we integrate external knowledge obtained from real geo-localization games, tapping into valuable human inference capabilities. The data are utilized to train GeoReasoner, which undergoes fine-tuning through dedicated reasoning and location-tuning stages. Qualitative and quantitative evaluations illustrate that GeoReasoner outperforms counterpart LVLMs by more than 25% at country-level and 38% at city-level geo-localization tasks, and surpasses StreetCLIP performance while requiring fewer training resources. The data and code are available at https://github.com/lingli1996/GeoReasoner.
Authors: Zhiwei Wei, Nai Yang, Wenjia Xu, Su Ding
Abstract: The grid map, often referred to as the tile map, stands as a vital tool in geospatial visualization, possessing unique attributes that differentiate it from more commonly known techniques such as choropleths and cartograms. It transforms geographic regions into grids, which requires the displacement of both region centroids and boundary nodes to establish a coherent grid arrangement. However, existing approaches typically displace region centroids and boundary nodes separately, potentially resulting in self-intersected boundaries and compromised relative orientation relations between regions. In this paper, we introduce a novel approach that leverages the Snake displacement algorithm from cartographic generalization to concurrently displace region centroids and boundary nodes. The revised Constrained Delaunay triangulation (CDT) is employed to represent the relations between regions and serves as a structural foundation for the Snake algorithm. Forces for displacing the region centroids into a grid-like pattern are then computed. These forces are iteratively applied within the Snake model until a satisfactory new boundary is achieved. Subsequently, the grid map is created by aligning the grids with the newly generated boundary, utilizing a one-to-one match algorithm to assign each region to a specific grid. Experimental results demonstrate that the proposed approach excels in maintaining the relative orientation and global shape of regions, albeit with a potential increase in local location deviations. We also present two strategies aligned with existing approaches to generate diverse grid maps for user preferences. Further details and resources are available on our project website: https://github.com/TrentonWei/DorlingMap.git.
Authors: Muhammad Anwar Ma'sum, Mahardhika Pratama, Ramasamy Savitha, Lin Liu, Habibullah, Ryszard Kowalczyk
Abstract: A continual learning (CL) model is desired for remote sensing image analysis because of varying camera parameters, spectral ranges, resolutions, etc. There exist some recent initiatives to develop CL techniques in this domain but they still depend on massive labelled samples which do not fully fit remote sensing applications because ground truths are often obtained via field-based surveys. This paper addresses this problem with a proposal of unsupervised flat-wide learning approach (UNISA) for unsupervised few-shot continual learning approaches of remote sensing image scene classifications which do not depend on any labelled samples for its model updates. UNISA is developed from the idea of prototype scattering and positive sampling for learning representations while the catastrophic forgetting problem is tackled with the flat-wide learning approach combined with a ball generator to address the data scarcity problem. Our numerical study with remote sensing image scene datasets and a hyperspectral dataset confirms the advantages of our solution. Source codes of UNISA are shared publicly in \url{https://github.com/anwarmaxsum/UNISA} to allow convenient future studies and reproductions of our numerical results.
Authors: Chang Zhou, Yang Zhao, Shaobo Liu, Yi Zhao, Xingchen Li, Chiyu Cheng
Abstract: In a society where traffic accidents frequently occur, fatigue driving has emerged as a grave issue. Fatigue driving detection technology, especially those based on the YOLOv8 deep learning model, has seen extensive research and application as an effective preventive measure. This paper discusses in depth the methods and technologies utilized in the YOLOv8 model to detect driver fatigue, elaborates on the current research status both domestically and internationally, and systematically introduces the processing methods and algorithm principles for various datasets. This study aims to provide a robust technical solution for preventing and detecting fatigue driving, thereby contributing significantly to reducing traffic accidents and safeguarding lives.
Authors: Yu Zhang, Chuang Zhu, Guoqing Yang, Siqi Chen
Abstract: Weakly Supervised Object Detection (WSOD) with only image-level annotation has recently attracted wide attention. Many existing methods ignore the inter-image relationship of instances which share similar characteristics while can certainly be determined not to belong to the same category. Therefore, in order to make full use of the weak label, we propose the Negative Prototypes Guided Contrastive learning (NPGC) architecture. Firstly, we define Negative Prototype as the proposal with the highest confidence score misclassified for the category that does not appear in the label. Unlike other methods that only utilize category positive feature, we construct an online updated global feature bank to store both positive prototypes and negative prototypes. Meanwhile, we propose a pseudo label sampling module to mine reliable instances and discard the easily misclassified instances based on the feature similarity with corresponding prototypes in global feature bank. Finally, we follow the contrastive learning paradigm to optimize the proposal's feature representation by attracting same class samples closer and pushing different class samples away in the embedding space. Extensive experiments have been conducted on VOC07, VOC12 datasets, which shows that our proposed method achieves the state-of-the-art performance.
Authors: Xuri Ge, Fuhai Chen, Songpei Xu, Fuxiang Tao, Jie Wang, Joemon M. Jose
Abstract: Image-text matching (ITM) is a fundamental problem in computer vision. The key issue lies in jointly learning the visual and textual representation to estimate their similarity accurately. Most existing methods focus on feature enhancement within modality or feature interaction across modalities, which, however, neglects the contextual information of the object representation based on the inter-object relationships that match the corresponding sentences with rich contextual semantics. In this paper, we propose a Hybrid-modal Interaction with multiple Relational Enhancements (termed \textit{Hire}) for image-text matching, which correlates the intra- and inter-modal semantics between objects and words with implicit and explicit relationship modelling. In particular, the explicit intra-modal spatial-semantic graph-based reasoning network is designed to improve the contextual representation of visual objects with salient spatial and semantic relational connectivities, guided by the explicit relationships of the objects' spatial positions and their scene graph. We use implicit relationship modelling for potential relationship interactions before explicit modelling to improve the fault tolerance of explicit relationship detection. Then the visual and textual semantic representations are refined jointly via inter-modal interactive attention and cross-modal alignment. To correlate the context of objects with the textual context, we further refine the visual semantic representation via cross-level object-sentence and word-image-based interactive attention. Extensive experiments validate that the proposed hybrid-modal interaction with implicit and explicit modelling is more beneficial for image-text matching. And the proposed \textit{Hire} obtains new state-of-the-art results on MS-COCO and Flickr30K benchmarks.
Authors: Lucas Berry, Axel Brando, David Meger
Abstract: Generative diffusion models, notable for their large parameter count (exceeding 100 million) and operation within high-dimensional image spaces, pose significant challenges for traditional uncertainty estimation methods due to computational demands. In this work, we introduce an innovative framework, Diffusion Ensembles for Capturing Uncertainty (DECU), designed for estimating epistemic uncertainty for diffusion models. The DECU framework introduces a novel method that efficiently trains ensembles of conditional diffusion models by incorporating a static set of pre-trained parameters, drastically reducing the computational burden and the number of parameters that require training. Additionally, DECU employs Pairwise-Distance Estimators (PaiDEs) to accurately measure epistemic uncertainty by evaluating the mutual information between model outputs and weights in high-dimensional spaces. The effectiveness of this framework is demonstrated through experiments on the ImageNet dataset, highlighting its capability to capture epistemic uncertainty, specifically in under-sampled image classes.
Authors: Hubert Kompanowski, Binh-Son Hua
Abstract: We present a method to generate 3D objects in styles. Our method takes a text prompt and a style reference image as input and reconstructs a neural radiance field to synthesize a 3D model with the content aligning with the text prompt and the style following the reference image. To simultaneously generate the 3D object and perform style transfer in one go, we propose a stylized score distillation loss to guide a text-to-3D optimization process to output visually plausible geometry and appearance. Our stylized score distillation is based on a combination of an original pretrained text-to-image model and its modified sibling with the key and value features of self-attention layers manipulated to inject styles from the reference image. Comparisons with state-of-the-art methods demonstrated the strong visual performance of our method, further supported by the quantitative results from our user study.
Authors: Miaowei Wang, Changjian Li, Amir Vaxman
Abstract: We present Canonical Consolidation Fields (CanFields): a method for reconstructing a time series of independently-sampled point clouds into a single deforming coherent shape. Such input often comes from motion capture. Existing methods either couple the geometry and the deformation, where by doing so they smooth fine details and lose the ability to track moving points, or they track the deformation explicitly, but introduce topological and geometric artifacts. Our novelty lies in the consolidation of the point clouds into a single canonical shape in a way that reduces the effect of noise and outliers, and enables us to overcome missing regions. We simultaneously reconstruct the velocity fields that guide the deformation. This consolidation allows us to retain the high-frequency details of the geometry, while faithfully reproducing the low-frequency deformation. Our architecture comprises simple components, and fits any single input shape without using datasets. We demonstrate the robustness and accuracy of our methods on a diverse benchmark of dynamic point clouds, including missing regions, sparse frames, and noise.
Authors: Le Zhuo, Ruoyi Du, Han Xiao, Yangguang Li, Dongyang Liu, Rongjie Huang, Wenze Liu, Lirui Zhao, Fu-Yun Wang, Zhanyu Ma, Xu Luo, Zehan Wang, Kaipeng Zhang, Xiangyang Zhu, Si Liu, Xiangyu Yue, Dingning Liu, Wanli Ouyang, Ziwei Liu, Yu Qiao, Hongsheng Li, Peng Gao
Abstract: Lumina-T2X is a nascent family of Flow-based Large Diffusion Transformers that establishes a unified framework for transforming noise into various modalities, such as images and videos, conditioned on text instructions. Despite its promising capabilities, Lumina-T2X still encounters challenges including training instability, slow inference, and extrapolation artifacts. In this paper, we present Lumina-Next, an improved version of Lumina-T2X, showcasing stronger generation performance with increased training and inference efficiency. We begin with a comprehensive analysis of the Flag-DiT architecture and identify several suboptimal components, which we address by introducing the Next-DiT architecture with 3D RoPE and sandwich normalizations. To enable better resolution extrapolation, we thoroughly compare different context extrapolation methods applied to text-to-image generation with 3D RoPE, and propose Frequency- and Time-Aware Scaled RoPE tailored for diffusion transformers. Additionally, we introduced a sigmoid time discretization schedule to reduce sampling steps in solving the Flow ODE and the Context Drop method to merge redundant visual tokens for faster network evaluation, effectively boosting the overall sampling speed. Thanks to these improvements, Lumina-Next not only improves the quality and efficiency of basic text-to-image generation but also demonstrates superior resolution extrapolation capabilities and multilingual generation using decoder-based LLMs as the text encoder, all in a zero-shot manner. To further validate Lumina-Next as a versatile generative framework, we instantiate it on diverse tasks including visual recognition, multi-view, audio, music, and point cloud generation, showcasing strong performance across these domains. By releasing all codes and model weights, we aim to advance the development of next-generation generative AI capable of universal modeling.
Authors: Cristhian Sanchez, Francisco Mena, Marcela Charfuelan, Marlon Nuske, Andreas Dengel
Abstract: Since the launch of the Sentinel-2 (S2) satellites, many ML models have used the data for diverse applications. The scene classification layer (SCL) inside the S2 product provides rich information for training, such as filtering images with high cloud coverage. However, there is more potential in this. We propose a technique to assess the clean optical coverage of a region, expressed by a SITS and calculated with the S2-based SCL data. With a manual threshold and specific labels in the SCL, the proposed technique assigns a percentage of spatial and temporal coverage across the time series and a high/low assessment. By evaluating the AI4EO challenge for Enhanced Agriculture, we show that the assessment is correlated to the predictive results of ML models. The classification results in a region with low spatial and temporal coverage is worse than in a region with high coverage. Finally, we applied the technique across all continents of the global dataset LandCoverNet.
Authors: Lin Zuo, Kunshan Yang, Xianlong Tian, Kunbin He, Yongqi Ding, Mengmeng Jing
Abstract: Existing computer vision methods mainly focus on the recognition of rigid objects, whereas the recognition of flexible objects remains unexplored. Recognizing flexible objects poses significant challenges due to their inherently diverse shapes and sizes, translucent attributes, ambiguous boundaries, and subtle inter-class differences. In this paper, we claim that these problems primarily arise from the lack of object saliency. To this end, we propose the Flexible Vision Graph Neural Network (FViG) to optimize the self-saliency and thereby improve the discrimination of the representations for flexible objects. Specifically, on one hand, we propose to maximize the channel-aware saliency by extracting the weight of neighboring nodes, which adapts to the shape and size variations in flexible objects. On the other hand, we maximize the spatial-aware saliency based on clustering to aggregate neighborhood information for the centroid nodes, which introduces local context information for the representation learning. To verify the performance of flexible objects recognition thoroughly, for the first time we propose the Flexible Dataset (FDA), which consists of various images of flexible objects collected from real-world scenarios or online. Extensive experiments evaluated on our Flexible Dataset demonstrate the effectiveness of our method on enhancing the discrimination of flexible objects.
Authors: Punnawat Siripathitti, Florent Forest, Olga Fink
Abstract: Damage to road pavement can develop into cracks, potholes, spallings, and other issues posing significant challenges to the integrity, safety, and durability of the road structure. Detecting and monitoring the evolution of these damages is crucial for maintaining the condition and structural health of road infrastructure. In recent years, researchers have explored various data-driven methods for image-based damage detection in road monitoring applications. The field gained attention with the introduction of the Road Damage Detection Challenge (RDDC2018), encouraging competition in developing object detectors on street-view images from various countries. Leading teams have demonstrated the effectiveness of ensemble models, mostly based on the YOLO and Faster R-CNN series. Data augmentations have also shown benefits in object detection within the computer vision field, including transformations such as random flipping, cropping, cutting out patches, as well as cut-and-pasting object instances. Applying cut-and-paste augmentation to road damages appears to be a promising approach to increase data diversity. However, the standard cut-and-paste technique, which involves sampling an object instance from a random image and pasting it at a random location onto the target image, has demonstrated limited effectiveness for road damage detection. This method overlooks the location of the road and disregards the difference in perspective between the sampled damage and the target image, resulting in unrealistic augmented images. In this work, we propose an improved Cut-and-Paste augmentation technique that is both content-aware (i.e. considers the true location of the road in the image) and perspective-aware (i.e. takes into account the difference in perspective between the injected damage and the target image).
Authors: Zach Nussbaum, Brandon Duderstadt, Andriy Mulyar
Abstract: This technical report describes the training of nomic-embed-vision, a highly performant, open-code, open-weights image embedding model that shares the same latent space as nomic-embed-text. Together, nomic-embed-vision and nomic-embed-text form the first unified latent space to achieve high performance across vision, language, and multimodal tasks.
Authors: Junhao Chen, Manyi Li, Zherong Pan, Xifeng Gao, Changhe Tu
Abstract: Deep generative models learn the data distribution, which is concentrated on a low-dimensional manifold. The geometric analysis of distribution transformation provides a better understanding of data structure and enables a variety of applications. In this paper, we study the geometric properties of the diffusion model, whose forward diffusion process and reverse generation process construct a series of distributions on manifolds which vary over time. Our key contribution is the introduction of generation rate, which corresponds to the local deformation of manifold over time around an image component. We show that the generation rate is highly correlated with intuitive visual properties, such as visual saliency, of the image component. Further, we propose an efficient and differentiable scheme to estimate the generation rate for a given image component over time, giving rise to a generation curve. The differentiable nature of our scheme allows us to control the shape of the generation curve via optimization. Using different loss functions, our generation curve matching algorithm provides a unified framework for a range of image manipulation tasks, including semantic transfer, object removal, saliency manipulation, image blending, etc. We conduct comprehensive analytical evaluations to support our findings and evaluate our framework on various manipulation tasks. The results show that our method consistently leads to better manipulation results, compared to recent baselines.
Authors: Andreas Stephan, Lukas Miklautz, Collin Leiber, Pedro Henrique Luz de Araujo, Dominik R\'ep\'as, Claudia Plant, Benjamin Roth
Abstract: Traditional image clustering techniques only find a single grouping within visual data. In particular, they do not provide a possibility to explicitly define multiple types of clustering. This work explores the potential of large vision-language models to facilitate alternative image clustering. We propose Text-Guided Alternative Image Consensus Clustering (TGAICC), a novel approach that leverages user-specified interests via prompts to guide the discovery of diverse clusterings. To achieve this, it generates a clustering for each prompt, groups them using hierarchical clustering, and then aggregates them using consensus clustering. TGAICC outperforms image- and text-based baselines on four alternative image clustering benchmark datasets. Furthermore, using count-based word statistics, we are able to obtain text-based explanations of the alternative clusterings. In conclusion, our research illustrates how contemporary large vision-language models can transform explanatory data analysis, enabling the generation of insightful, customizable, and diverse image clusterings.
Authors: Mingxiao Huo, Pengliang Ji, Haotian Lin, Junchen Liu, Yixiao Wang, Yijun Chen
Abstract: We introduce a pioneering unified library that leverages depth anything, segment anything models to augment neural comprehension in language-vision model zero-shot understanding. This library synergizes the capabilities of the Depth Anything Model (DAM), Segment Anything Model (SAM), and GPT-4V, enhancing multimodal tasks such as vision-question-answering (VQA) and composition reasoning. Through the fusion of segmentation and depth analysis at the symbolic instance level, our library provides nuanced inputs for language models, significantly advancing image interpretation. Validated across a spectrum of in-the-wild real-world images, our findings showcase progress in vision-language models through neural-symbolic integration. This novel approach melds visual and language analysis in an unprecedented manner. Overall, our library opens new directions for future research aimed at decoding the complexities of the real world through advanced multimodal technologies and our code is available at \url{https://github.com/AnthonyHuo/SAM-DAM-for-Compositional-Reasoning}.
URLs: https://github.com/AnthonyHuo/SAM-DAM-for-Compositional-Reasoning
Authors: Jay Idema, Pieter Peers
Abstract: We propose a material appearance modeling neural network for visualizing plausible, spatially-varying materials under diverse view and lighting conditions, utilizing only a single photograph of a material under co-located light and view as input for appearance estimation. Our neural architecture is composed of two network stages: a network that infers learned per-pixel neural parameters of a material from a single input photograph, and a network that renders the material utilizing these neural parameters, similar to a BRDF. We train our model on a set of 312,165 synthetic spatially-varying exemplars. Since our method infers learned neural parameters rather than analytical BRDF parameters, our method is capable of encoding anisotropic and global illumination (inter-pixel interaction) information into individual pixel parameters. We demonstrate our model's performance compared to prior work and demonstrate the feasibility of the render network as a BRDF by implementing it into the Mitsuba3 rendering engine. Finally, we briefly discuss the capability of neural parameters to encode global illumination information.
Authors: Haoran Li, Xingjian Li, Jiahua Shi, Huaming Chen, Bo Du, Daisuke Kihara, Johan Barthelemy, Jun Shen, Min Xu
Abstract: Cryo-Electron Tomography (cryo-ET) is a 3D imaging technology facilitating the study of macromolecular structures at near-atomic resolution. Recent volumetric segmentation approaches on cryo-ET images have drawn widespread interest in biological sector. However, existing methods heavily rely on manually labeled data, which requires highly professional skills, thereby hindering the adoption of fully-supervised approaches for cryo-ET images. Some unsupervised domain adaptation (UDA) approaches have been designed to enhance the segmentation network performance using unlabeled data. However, applying these methods directly to cryo-ET images segmentation tasks remains challenging due to two main issues: 1) the source data, usually obtained through simulation, contain a certain level of noise, while the target data, directly collected from raw-data from real-world scenario, have unpredictable noise levels. 2) the source data used for training typically consists of known macromoleculars, while the target domain data are often unknown, causing the model's segmenter to be biased towards these known macromolecules, leading to a domain shift problem. To address these challenges, in this work, we introduce the first voxel-wise unsupervised domain adaptation approach, termed Vox-UDA, specifically for cryo-ET subtomogram segmentation. Vox-UDA incorporates a noise generation module to simulate target-like noises in the source dataset for cross-noise level adaptation. Additionally, we propose a denoised pseudo-labeling strategy based on improved Bilateral Filter to alleviate the domain shift problem. Experimental results on both simulated and real cryo-ET subtomogram datasets demonstrate the superiority of our proposed approach compared to state-of-the-art UDA methods.
Authors: Pranjali Singh, Prithwijit Guha
Abstract: Underwater image quality is affected by fluorescence, low illumination, absorption, and scattering. Recent works in underwater image enhancement have proposed different deep network architectures to handle these problems. Most of these works have proposed a single network to handle all the challenges. We believe that deep networks trained for specific conditions deliver better performance than a single network learned from all degradation cases. Accordingly, the first contribution of this work lies in the proposal of an iterative framework where a single dominant degradation condition is identified and resolved. This proposal considers the following eight degradation conditions -- low illumination, low contrast, haziness, blurred image, presence of noise and color imbalance in three different channels. A deep network is designed to identify the dominant degradation condition. Accordingly, an appropriate deep network is selected for degradation condition-specific enhancement. The second contribution of this work is the construction of degradation condition specific datasets from good quality images of two standard datasets (UIEB and EUVP). This dataset is used to learn the condition specific enhancement networks. The proposed approach is found to outperform nine baseline methods on UIEB and EUVP datasets.
Authors: Nabeel Nisar Bhat, Rafael Berkvens Jeroen Famaey
Abstract: In recent years, Joint Communication and Sensing (JC&S), has demonstrated significant success, particularly in utilizing sub-6 GHz frequencies with commercial-off-the-shelf (COTS) Wi-Fi devices for applications such as localization, gesture recognition, and pose classification. Deep learning and the existence of large public datasets has been pivotal in achieving such results. However, at mmWave frequencies (30-300 GHz), which has shown potential for more accurate sensing performance, there is a noticeable lack of research in the domain of COTS Wi-Fi sensing. Challenges such as limited research hardware, the absence of large datasets, limited functionality in COTS hardware, and the complexities of data collection present obstacles to a comprehensive exploration of this field. In this work, we aim to address these challenges by developing a method that can generate synthetic mmWave channel state information (CSI) samples. In particular, we use a generative adversarial network (GAN) on an existing dataset, to generate 30,000 additional CSI samples. The augmented samples exhibit a remarkable degree of consistency with the original data, as indicated by the notably high GAN-train and GAN-test scores. Furthermore, we integrate the augmented samples in training a pose classification model. We observe that the augmented samples complement the real data and improve the generalization of the classification model.
Authors: Manli Zhu, Edmond S. L. Ho, Shuang Chen, Longzhi Yang, Hubert P. H. Shum
Abstract: Cameras are essential vision instruments to capture images for pattern detection and measurement. Human-object interaction (HOI) detection is one of the most popular pattern detection approaches for captured human-centric visual scenes. Recently, Transformer-based models have become the dominant approach for HOI detection due to their advanced network architectures and thus promising results. However, most of them follow the one-stage design of vanilla Transformer, leaving rich geometric priors under-exploited and leading to compromised performance especially when occlusion occurs. Given that geometric features tend to outperform visual ones in occluded scenarios and offer information that complements visual cues, we propose a novel end-to-end Transformer-style HOI detection model, i.e., geometric features enhanced HOI detector (GeoHOI). One key part of the model is a new unified self-supervised keypoint learning method named UniPointNet that bridges the gap of consistent keypoint representation across diverse object categories, including humans. GeoHOI effectively upgrades a Transformer-based HOI detector benefiting from the keypoints similarities measuring the likelihood of human-object interactions as well as local keypoint patches to enhance interaction query representation, so as to boost HOI predictions. Extensive experiments show that the proposed method outperforms the state-of-the-art models on V-COCO and achieves competitive performance on HICO-DET. Case study results on the post-disaster rescue with vision-based instruments showcase the applicability of the proposed GeoHOI in real-world applications.
Authors: Trupti Mahendrakar, Ryan T. White, Madhur Tiwari
Abstract: This paper focuses on autonomously characterizing components such as solar panels, body panels, antennas, and thrusters of an unknown resident space object (RSO) using camera feed to aid autonomous on-orbit servicing (OOS) and active debris removal. Significant research has been conducted in this area using convolutional neural networks (CNNs). While CNNs are powerful at learning patterns and performing object detection, they struggle with missed detections and misclassifications in environments different from the training data, making them unreliable for safety in high-stakes missions like OOS. Additionally, failures exhibited by CNNs are often easily rectifiable by humans using commonsense reasoning and contextual knowledge. Embedding such reasoning in an object detector could improve detection accuracy. To validate this hypothesis, this paper presents an end-to-end object detector called SpaceYOLOv2 (SpY), which leverages the generalizability of CNNs while incorporating contextual knowledge using traditional computer vision techniques. SpY consists of two main components: a shape detector and the SpaceYOLO classifier (SYC). The shape detector uses CNNs to detect primitive shapes of RSOs and SYC associates these shapes with contextual knowledge, such as color and texture, to classify them as spacecraft components or "unknown" if the detected shape is uncertain. SpY's modular architecture allows customizable usage of contextual knowledge to improve detection performance, or SYC as a secondary fail-safe classifier with an existing spacecraft component detector. Performance evaluations on hardware-in-the-loop images of a mock-up spacecraft demonstrate that SpY is accurate and an ensemble of SpY with YOLOv5 trained for satellite component detection improved the performance by 23.4% in recall, demonstrating enhanced safety for vision-based navigation tasks.
Authors: Colton Stearns, Adam Harley, Mikaela Uy, Florian Dubost, Federico Tombari, Gordon Wetzstein, Leonidas Guibas
Abstract: Gaussian splatting has become a popular representation for novel-view synthesis, exhibiting clear strengths in efficiency, photometric quality, and compositional edibility. Following its success, many works have extended Gaussians to 4D, showing that dynamic Gaussians maintain these benefits while also tracking scene geometry far better than alternative representations. Yet, these methods assume dense multi-view videos as supervision, constraining their use to controlled capture settings. In this work, we extend the capability of Gaussian scene representations to casually captured monocular videos. We show that existing 4D Gaussian methods dramatically fail in this setup because the monocular setting is underconstrained. Building off this finding, we propose Dynamic Gaussian Marbles (DGMarbles), consisting of three core modifications that target the difficulties of the monocular setting. First, DGMarbles uses isotropic Gaussian "marbles", reducing the degrees of freedom of each Gaussian, and constraining the optimization to focus on motion and appearance over local shape. Second, DGMarbles employs a hierarchical divide-and-conquer learning strategy to guide the optimization towards solutions with coherent motion. Finally, DGMarbles adds image-level and geometry-level priors into the optimization, including a tracking loss that takes advantage of recent progress in point tracking. By constraining the optimization in these ways, DGMarbles learns Gaussian trajectories that enable novel-view rendering and accurately capture the 3D motion of the scene elements. We evaluate on the (monocular) Nvidia Dynamic Scenes dataset and the Dycheck iPhone dataset, and show that DGMarbles significantly outperforms other Gaussian baselines in quality, and is on-par with non-Gaussian representations, all while maintaining the efficiency, compositionality, editability, and tracking benefits of Gaussians.
Authors: Georgios Tziafas, Yucheng Xu, Zhibin Li, Hamidreza Kasaei
Abstract: Grounding natural language to the physical world is a ubiquitous topic with a wide range of applications in computer vision and robotics. Recently, 2D vision-language models such as CLIP have been widely popularized, due to their impressive capabilities for open-vocabulary grounding in 2D images. Recent works aim to elevate 2D CLIP features to 3D via feature distillation, but either learn neural fields that are scene-specific and hence lack generalization, or focus on indoor room scan data that require access to multiple camera views, which is not practical in robot manipulation scenarios. Additionally, related methods typically fuse features at pixel-level and assume that all camera views are equally informative. In this work, we show that this approach leads to sub-optimal 3D features, both in terms of grounding accuracy, as well as segmentation crispness. To alleviate this, we propose a multi-view feature fusion strategy that employs object-centric priors to eliminate uninformative views based on semantic information, and fuse features at object-level via instance segmentation masks. To distill our object-centric 3D features, we generate a large-scale synthetic multi-view dataset of cluttered tabletop scenes, spawning 15k scenes from over 3300 unique object instances, which we make publicly available. We show that our method reconstructs 3D CLIP features with improved grounding capacity and spatial consistency, while doing so from single-view RGB-D, thus departing from the assumption of multiple camera views at test time. Finally, we show that our approach can generalize to novel tabletop domains and be re-purposed for 3D instance segmentation without fine-tuning, and demonstrate its utility for language-guided robotic grasping in clutter
Authors: William Berman, Alexander Peysakhovich
Abstract: We train a model to generate images from multimodal prompts of interleaved text and images such as "a
Authors: Tao Lian, Jose L. G\'omez, Antonio M. L\'opez
Abstract: The last mile of unsupervised domain adaptation (UDA) for semantic segmentation is the challenge of solving the syn-to-real domain gap. Recent UDA methods have progressed significantly, yet they often rely on strategies customized for synthetic single-source datasets (e.g., GTA5), which limits their generalisation to multi-source datasets. Conversely, synthetic multi-source datasets hold promise for advancing the last mile of UDA but remain underutilized in current research. Thus, we propose DEC, a flexible UDA framework for multi-source datasets. Following a divide-and-conquer strategy, DEC simplifies the task by categorizing semantic classes, training models for each category, and fusing their outputs by an ensemble model trained exclusively on synthetic datasets to obtain the final segmentation mask. DEC can integrate with existing UDA methods, achieving state-of-the-art performance on Cityscapes, BDD100K, and Mapillary Vistas, significantly narrowing the syn-to-real domain gap.
Authors: Mingyang Zhao, Jingen Jiang, Lei Ma, Shiqing Xin, Gaofeng Meng, Dong-Ming Yan
Abstract: This paper presents a novel non-rigid point set registration method that is inspired by unsupervised clustering analysis. Unlike previous approaches that treat the source and target point sets as separate entities, we develop a holistic framework where they are formulated as clustering centroids and clustering members, separately. We then adopt Tikhonov regularization with an $\ell_1$-induced Laplacian kernel instead of the commonly used Gaussian kernel to ensure smooth and more robust displacement fields. Our formulation delivers closed-form solutions, theoretical guarantees, independence from dimensions, and the ability to handle large deformations. Subsequently, we introduce a clustering-improved Nystr\"om method to effectively reduce the computational complexity and storage of the Gram matrix to linear, while providing a rigorous bound for the low-rank approximation. Our method achieves high accuracy results across various scenarios and surpasses competitors by a significant margin, particularly on shapes with substantial deformations. Additionally, we demonstrate the versatility of our method in challenging tasks such as shape transfer and medical registration.
Authors: Huaying Zhang, Rintaro Yanagi, Ren Togo, Takahiro Ogawa, Miki Haseyama
Abstract: This paper proposes a novel zero-shot composed image retrieval (CIR) method considering the query-target relationship by masked image-text pairs. The objective of CIR is to retrieve the target image using a query image and a query text. Existing methods use a textual inversion network to convert the query image into a pseudo word to compose the image and text and use a pre-trained visual-language model to realize the retrieval. However, they do not consider the query-target relationship to train the textual inversion network to acquire information for retrieval. In this paper, we propose a novel zero-shot CIR method that is trained end-to-end using masked image-text pairs. By exploiting the abundant image-text pairs that are convenient to obtain with a masking strategy for learning the query-target relationship, it is expected that accurate zero-shot CIR using a retrieval-focused textual inversion network can be realized. Experimental results show the effectiveness of the proposed method.
Authors: Yuxiang Huang, Yuhao Chen, John Zelek
Abstract: Motion segmentation from a single moving camera presents a significant challenge in the field of computer vision. This challenge is compounded by the unknown camera movements and the lack of depth information of the scene. While deep learning has shown impressive capabilities in addressing these issues, supervised models require extensive training on massive annotated datasets, and unsupervised models also require training on large volumes of unannotated data, presenting significant barriers for both. In contrast, traditional methods based on optical flow do not require training data, however, they often fail to capture object-level information, leading to over-segmentation or under-segmentation. In addition, they also struggle in complex scenes with substantial depth variations and non-rigid motion, due to the overreliance of optical flow. To overcome these challenges, we propose an innovative hybrid approach that leverages the advantages of both deep learning methods and traditional optical flow based methods to perform dense motion segmentation without requiring any training. Our method initiates by automatically generating object proposals for each frame using foundation models. These proposals are then clustered into distinct motion groups using both optical flow and relative depth maps as motion cues. The integration of depth maps derived from state-of-the-art monocular depth estimation models significantly enhances the motion cues provided by optical flow, particularly in handling motion parallax issues. Our method is evaluated on the DAVIS-Moving and YTVOS-Moving datasets, and the results demonstrate that our method outperforms the best unsupervised method and closely matches with the state-of-theart supervised methods.
Authors: Siyuan Liang, Jiawei Liang, Tianyu Pang, Chao Du, Aishan Liu, Ee-Chien Chang, Xiaochun Cao
Abstract: Instruction tuning enhances large vision-language models (LVLMs) but raises security risks through potential backdoor attacks due to their openness. Previous backdoor studies focus on enclosed scenarios with consistent training and testing instructions, neglecting the practical domain gaps that could affect attack effectiveness. This paper empirically examines the generalizability of backdoor attacks during the instruction tuning of LVLMs for the first time, revealing certain limitations of most backdoor strategies in practical scenarios. We quantitatively evaluate the generalizability of six typical backdoor attacks on image caption benchmarks across multiple LVLMs, considering both visual and textual domain offsets. Our findings indicate that attack generalizability is positively correlated with the backdoor trigger's irrelevance to specific images/models and the preferential correlation of the trigger pattern. Additionally, we modify existing backdoor attacks based on the above key observations, demonstrating significant improvements in cross-domain scenario generalizability (+86% attack success rate). Notably, even without access to the instruction datasets, a multimodal instruction set can be successfully poisoned with a very low poisoning rate (0.2%), achieving an attack success rate of over 97%. This paper underscores that even simple traditional backdoor strategies pose a serious threat to LVLMs, necessitating more attention and in-depth research.
Authors: Lan Chen, Dong Li, Xiao Wang, Pengpeng Shao, Wei Zhang, Yaowei Wang, Yonghong Tian, Jin Tang
Abstract: Existing event stream-based pattern recognition models usually represent the event stream as the point cloud, voxel, image, etc., and design various deep neural networks to learn their features. Although considerable results can be achieved in simple cases, however, the model performance may be limited by monotonous modality expressions, sub-optimal fusion, and readout mechanisms. In this paper, we propose a novel dual-stream framework for event stream-based pattern recognition via differentiated fusion, termed EFV++. It models two common event representations simultaneously, i.e., event images and event voxels. The spatial and three-dimensional stereo information can be learned separately by utilizing Transformer and Graph Neural Network (GNN). We believe the features of each representation still contain both efficient and redundant features and a sub-optimal solution may be obtained if we directly fuse them without differentiation. Thus, we divide each feature into three levels and retain high-quality features, blend medium-quality features, and exchange low-quality features. The enhanced dual features will be fed into the fusion Transformer together with bottleneck features. In addition, we introduce a novel hybrid interaction readout mechanism to enhance the diversity of features as final representations. Extensive experiments demonstrate that our proposed framework achieves state-of-the-art performance on multiple widely used event stream-based classification datasets. Specifically, we achieve new state-of-the-art performance on the Bullying10k dataset, i.e., $90.51\%$, which exceeds the second place by $+2.21\%$. The source code of this paper has been released on \url{https://github.com/Event-AHU/EFV_event_classification/tree/EFVpp}.
URLs: https://github.com/Event-AHU/EFV_event_classification/tree/EFVpp
Authors: Jie Zhang, Zhongqi Wang, Mengqi Lei, Zheng Yuan, Bei Yan, Shiguang Shan, Xilin Chen
Abstract: Currently many benchmarks have been proposed to evaluate the perception ability of the Large Vision-Language Models (LVLMs). However, most benchmarks conduct questions by selecting images from existing datasets, resulting in the potential data leakage. Besides, these benchmarks merely focus on evaluating LVLMs on the realistic style images and clean scenarios, leaving the multi-stylized images and noisy scenarios unexplored. In response to these challenges, we propose a dynamic and scalable benchmark named Dysca for evaluating LVLMs by leveraging synthesis images. Specifically, we leverage Stable Diffusion and design a rule-based method to dynamically generate novel images, questions and the corresponding answers. We consider 51 kinds of image styles and evaluate the perception capability in 20 subtasks. Moreover, we conduct evaluations under 4 scenarios (i.e., Clean, Corruption, Print Attacking and Adversarial Attacking) and 3 question types (i.e., Multi-choices, True-or-false and Free-form). Thanks to the generative paradigm, Dysca serves as a scalable benchmark for easily adding new subtasks and scenarios. A total of 8 advanced open-source LVLMs with 10 checkpoints are evaluated on Dysca, revealing the drawbacks of current LVLMs. The benchmark is released in \url{https://github.com/Benchmark-Dysca/Dysca}.
Authors: Wenxuan Ma, Shuang Li, Lincan Cai, Jingxuan Kang
Abstract: Cross-modality transfer aims to leverage large pretrained models to complete tasks that may not belong to the modality of pretraining data. Existing works achieve certain success in extending classical finetuning to cross-modal scenarios, yet we still lack understanding about the influence of modality gap on the transfer. In this work, a series of experiments focusing on the source representation quality during transfer are conducted, revealing the connection between larger modality gap and lesser knowledge reuse which means ineffective transfer. We then formalize the gap as the knowledge misalignment between modalities using conditional distribution P(Y|X). Towards this problem, we present Modality kNowledge Alignment (MoNA), a meta-learning approach that learns target data transformation to reduce the modality knowledge discrepancy ahead of the transfer. Experiments show that out method enables better reuse of source modality knowledge in cross-modality transfer, which leads to improvements upon existing finetuning methods.
Authors: Yicheng Xu, Yuxin Chen, Jiahao Nie, Yusong Wang, Huiping Zhuang, Manabu Okumura
Abstract: Continual learning (CL) with Vision-Language Models (VLMs) has overcome the constraints of traditional CL, which only focuses on previously encountered classes. During the CL of VLMs, we need not only to prevent the catastrophic forgetting on incrementally learned knowledge but also to preserve the zero-shot ability of VLMs. However, existing methods require additional reference datasets to maintain such zero-shot ability and rely on domain-identity hints to classify images across different domains. In this study, we propose Regression-based Analytic Incremental Learning (RAIL), which utilizes a recursive ridge regression-based adapter to learn from a sequence of domains in a non-forgetting manner and decouple the cross-domain correlations by projecting features to a higher-dimensional space. Cooperating with a training-free fusion module, RAIL absolutely preserves the VLM's zero-shot ability on unseen domains without any reference data. Additionally, we introduce Cross-domain Task-Agnostic Incremental Learning (X-TAIL) setting. In this setting, a CL learner is required to incrementally learn from multiple domains and classify test images from both seen and unseen domains without any domain-identity hint. We theoretically prove RAIL's absolute memorization on incrementally learned domains. Experiment results affirm RAIL's state-of-the-art performance in both X-TAIL and existing Multi-domain Task-Incremental Learning settings. The code will be released upon acceptance.
Authors: Aishwarya Agarwal, Srikrishna Karanam, Balaji Vasan Srinivasan
Abstract: We consider the problem of customizing text-to-image diffusion models with user-supplied reference images. Given new prompts, the existing methods can capture the key concept from the reference images but fail to align the generated image with the prompt. In this work, we seek to address this key issue by proposing new methods that can easily be used in conjunction with existing customization methods that optimize the embeddings/weights at various intermediate stages of the text encoding process. The first contribution of this paper is a dissection of the various stages of the text encoding process leading up to the conditioning vector for text-to-image models. We take a holistic view of existing customization methods and notice that key and value outputs from this process differs substantially from their corresponding baseline (non-customized) models (e.g., baseline stable diffusion). While this difference does not impact the concept being customized, it leads to other parts of the generated image not being aligned with the prompt (see first row in Fig 1). Further, we also observe that these keys and values allow independent control various aspects of the final generation, enabling semantic manipulation of the output. Taken together, the features spanning these keys and values, serve as the basis for our next contribution where we fix the aforementioned issues with existing methods. We propose a new post-processing algorithm, \textbf{AlignIT}, that infuses the keys and values for the concept of interest while ensuring the keys and values for all other tokens in the input prompt are unchanged. Our proposed method can be plugged in directly to existing customization methods, leading to a substantial performance improvement in the alignment of the final result with the input prompt while retaining the customization quality.
Authors: Kibaek Park, Francois Rameau, Jaesik Park, In So Kweon
Abstract: The large abundance of perspective camera datasets facilitated the emergence of novel learning-based strategies for various tasks, such as camera localization, single image depth estimation, or view synthesis. However, panoramic or omnidirectional image datasets, including essential information, such as pose and depth, are mostly made with synthetic scenes. In this work, we introduce a large scale 360$^{\circ}$ videos dataset in the wild. This dataset has been carefully scraped from the Internet and has been captured from various locations worldwide. Hence, this dataset exhibits very diversified environments (e.g., indoor and outdoor) and contexts (e.g., with and without moving objects). Each of the 25K images constituting our dataset is provided with its respective camera's pose and depth map. We illustrate the relevance of our dataset for two main tasks, namely, single image depth estimation and view synthesis.
Authors: Srinitish Srinivasan, Karthik Seemakurthy
Abstract: Deep neural networks (DNNs) have exhibited remarkable performance across various tasks, yet their susceptibility to spurious correlations poses a significant challenge for out-of-distribution (OOD) generalization. Spurious correlations refer to erroneous associations in data that do not reflect true underlying relationships but are instead artifacts of dataset characteristics or biases. These correlations can lead DNNs to learn patterns that are not robust across diverse datasets or real-world scenarios, hampering their ability to generalize beyond training data. In this paper, we propose an autoencoder-based approach to analyze the nature of spurious correlations that exist in the Global Wheat Head Detection (GWHD) 2021 dataset. We then use inpainting followed by Weighted Boxes Fusion (WBF) to achieve a 2% increase in the Average Domain Accuracy (ADA) over the YOLOv5 baseline and consistently show that our approach has the ability to suppress some of the spurious correlations in the GWHD 2021 dataset. The key advantage of our approach is that it is more suitable in scenarios where there is limited scope to adapt or fine-tune the trained model in unseen test environments.
Authors: Qiushi Guo
Abstract: Detecting obstacles in railway scenarios is both crucial and challenging due to the wide range of obstacle categories and varying ambient conditions such as weather and light. Given the impossibility of encompassing all obstacle categories during the training stage, we address this out-of-distribution (OOD) issue with a semi-supervised segmentation approach guided by optical flow clues. We reformulate the task as a binary segmentation problem instead of the traditional object detection approach. To mitigate data shortages, we generate highly realistic synthetic images using Segment Anything (SAM) and YOLO, eliminating the need for manual annotation to produce abundant pixel-level annotations. Additionally, we leverage optical flow as prior knowledge to train the model effectively. Several experiments are conducted, demonstrating the feasibility and effectiveness of our approach.
Authors: Zhaokang Liao, Hao Feng, Shaokai Liu, Wengang Zhou, Houqiang Li
Abstract: Fisheye images are categorized fisheye into central and deviated based on the optical center position. Existing rectification methods are limited to central fisheye images, while this paper proposes a novel method that extends to deviated fisheye image rectification. The challenge lies in the variant global distortion distribution pattern caused by the random optical center position. To address this challenge, we propose a distortion vector map (DVM) that measures the degree and direction of local distortion. By learning the DVM, the model can independently identify local distortions at each pixel without relying on global distortion patterns. The model adopts a pre-training and fine-tuning training paradigm. In the pre-training stage, it predicts the distortion vector map and perceives the local distortion features of each pixel. In the fine-tuning stage, it predicts a pixel-wise flow map for deviated fisheye image rectification. We also propose a data augmentation method mixing central, deviated, and distorted-free images. Such data augmentation promotes the model performance in rectifying both central and deviated fisheye images, compared with models trained on single-type fisheye images. Extensive experiments demonstrate the effectiveness and superiority of the proposed method.
Authors: Zuo Zuo, Jiahao Dong, Yao Wu, Yanyun Qu, Zongze Wu
Abstract: Few-shot anomaly detection methods can effectively address data collecting difficulty in industrial scenarios. Compared to 2D few-shot anomaly detection (2D-FSAD), 3D few-shot anomaly detection (3D-FSAD) is still an unexplored but essential task. In this paper, we propose CLIP3D-AD, an efficient 3D-FSAD method extended on CLIP. We successfully transfer strong generalization ability of CLIP into 3D-FSAD. Specifically, we synthesize anomalous images on given normal images as sample pairs to adapt CLIP for 3D anomaly classification and segmentation. For classification, we introduce an image adapter and a text adapter to fine-tune global visual features and text features. Meanwhile, we propose a coarse-to-fine decoder to fuse and facilitate intermediate multi-layer visual representations of CLIP. To benefit from geometry information of point cloud and eliminate modality and data discrepancy when processed by CLIP, we project and render point cloud to multi-view normal and anomalous images. Then we design multi-view fusion module to fuse features of multi-view images extracted by CLIP which are used to facilitate visual representations for further enhancing vision-language correlation. Extensive experiments demonstrate that our method has a competitive performance of 3D few-shot anomaly classification and segmentation on MVTec-3D AD dataset.
Authors: Yixin Liu, Ruoxi Chen, Lichao Sun
Abstract: Personalized diffusion models have gained popularity for adapting pre-trained text-to-image models to generate images of specific topics with only a few images. However, recent studies find that these models are vulnerable to minor adversarial perturbation, and the fine-tuning performance is largely degraded on corrupted datasets. Such characteristics are further exploited to craft protective perturbation on sensitive images like portraits that prevent unauthorized generation. In response, diffusion-based purification methods have been proposed to remove these perturbations and retain generation performance. However, existing works lack detailed analysis of the fundamental shortcut learning vulnerability of personalized diffusion models and also turn to over-purifying the images cause information loss. In this paper, we take a closer look at the fine-tuning process of personalized diffusion models through the lens of shortcut learning and propose a hypothesis that could explain the underlying manipulation mechanisms of existing perturbation methods. Specifically, we find that the perturbed images are greatly shifted from their original paired prompt in the CLIP-based latent space. As a result, training with this mismatched image-prompt pair creates a construction that causes the models to dump their out-of-distribution noisy patterns to the identifier, thus causing serious performance degradation. Based on this observation, we propose a systematic approach to retain the training performance with purification that realigns the latent image and its semantic meaning and also introduces contrastive learning with a negative token to decouple the learning of wanted clean identity and the unwanted noisy pattern, that shows strong potential capacity against further adaptive perturbation.
Authors: Yanan Sun, Yanchen Liu, Yinhao Tang, Wenjie Pei, Kai Chen
Abstract: The field of text-to-image (T2I) generation has made significant progress in recent years, largely driven by advancements in diffusion models. Linguistic control enables effective content creation, but struggles with fine-grained control over image generation. This challenge has been explored, to a great extent, by incorporating additional user-supplied spatial conditions, such as depth maps and edge maps, into pre-trained T2I models through extra encoding. However, multi-control image synthesis still faces several challenges. Specifically, current approaches are limited in handling free combinations of diverse input control signals, overlook the complex relationships among multiple spatial conditions, and often fail to maintain semantic alignment with provided textual prompts. This can lead to suboptimal user experiences. To address these challenges, we propose AnyControl, a multi-control image synthesis framework that supports arbitrary combinations of diverse control signals. AnyControl develops a novel Multi-Control Encoder that extracts a unified multi-modal embedding to guide the generation process. This approach enables a holistic understanding of user inputs, and produces high-quality, faithful results under versatile control signals, as demonstrated by extensive quantitative and qualitative evaluations. Our project page is available in \url{https://any-control.github.io}.
Authors: Vu Minh Hieu Phan, Yutong Xie, Bowen Zhang, Yuankai Qi, Zhibin Liao, Antonios Perperidis, Son Lam Phung, Johan W. Verjans, Minh-Son To
Abstract: Unpaired medical image synthesis aims to provide complementary information for an accurate clinical diagnostics, and address challenges in obtaining aligned multi-modal medical scans. Transformer-based models excel in imaging translation tasks thanks to their ability to capture long-range dependencies. Although effective in supervised training settings, their performance falters in unpaired image synthesis, particularly in synthesizing structural details. This paper empirically demonstrates that, lacking strong inductive biases, Transformer can converge to non-optimal solutions in the absence of paired data. To address this, we introduce UNet Structured Transformer (UNest), a novel architecture incorporating structural inductive biases for unpaired medical image synthesis. We leverage the foundational Segment-Anything Model to precisely extract the foreground structure and perform structural attention within the main anatomy. This guides the model to learn key anatomical regions, thus improving structural synthesis under the lack of supervision in unpaired training. Evaluated on two public datasets, spanning three modalities, i.e., MR, CT, and PET, UNest improves recent methods by up to 19.30% across six medical image synthesis tasks. Our code is released at https://github.com/HieuPhan33/MICCAI2024-UNest.
Authors: Lijie Hu, Tianhao Huang, Huanyi Xie, Chenyang Ren, Zhengyu Hu, Lu Yu, Di Wang
Abstract: Concept Bottleneck Models (CBMs) have garnered increasing attention due to their ability to provide concept-based explanations for black-box deep learning models while achieving high final prediction accuracy using human-like concepts. However, the training of current CBMs heavily relies on the accuracy and richness of annotated concepts in the dataset. These concept labels are typically provided by experts, which can be costly and require significant resources and effort. Additionally, concept saliency maps frequently misalign with input saliency maps, causing concept predictions to correspond to irrelevant input features - an issue related to annotation alignment. To address these limitations, we propose a new framework called SSCBM (Semi-supervised Concept Bottleneck Model). Our SSCBM is suitable for practical situations where annotated data is scarce. By leveraging joint training on both labeled and unlabeled data and aligning the unlabeled data at the concept level, we effectively solve these issues. We proposed a strategy to generate pseudo labels and an alignment loss. Experiments demonstrate that our SSCBM is both effective and efficient. With only 20% labeled data, we achieved 93.19% (96.39% in a fully supervised setting) concept accuracy and 75.51% (79.82% in a fully supervised setting) prediction accuracy.
Authors: Yu Zhe, Jun Sakuma
Abstract: Zero-shot domain adaptation (ZSDA) is a domain adaptation problem in the situation that labeled samples for a target task (task of interest) are only available from the source domain at training time, but for a task different from the task of interest (irrelevant task), labeled samples are available from both source and target domains. In this situation, classical domain adaptation techniques can only learn domain-invariant features in the irrelevant task. However, due to the difference in sample distribution between the two tasks, domain-invariant features learned in the irrelevant task are biased and not necessarily domain-invariant in the task of interest. To solve this problem, this paper proposes a new ZSDA method to learn domain-invariant features with low task bias. To this end, we propose (1) data augmentation with dual-level mixups in both task and domain to fill the absence of target task-of-interest data, (2) an extension of domain adversarial learning to learn domain-invariant features with less task bias, and (3) a new dual-level contrastive learning method that enhances domain-invariance and less task biasedness of features. Experimental results show that our proposal achieves good performance on several benchmarks.
Authors: Mikko Impi\"o, Jenni Raitoharju
Abstract: Image-based species identification could help scaling biodiversity monitoring to a global scale. Many challenges still need to be solved in order to implement these systems in real-world applications. A reliable image-based monitoring system must detect out-of-distribution (OOD) classes it has not been presented before. This is challenging especially with fine-grained classes. Emerging environmental monitoring techniques, DNA metabarcoding and eDNA, can help by providing information on OOD classes that are present in a sample. In this paper, we study if DNA barcodes can also support in finding the outlier images based on the outlier DNA sequence's similarity to the seen classes. We propose a re-ordering approach that can be easily applied on any pre-trained models and existing OOD detection methods. We experimentally show that the proposed approach improves taxonomic OOD detection compared to all common baselines. We also show that the method works thanks to a correlation between visual similarity and DNA barcode proximity. The code and data are available at https://github.com/mikkoim/dnaimg-ood.
Authors: Hui Lu, Albert Ali Salah, Ronald Poppe
Abstract: Video understanding requires the extraction of rich spatio-temporal representations, which transformer models achieve through self-attention. Unfortunately, self-attention poses a computational burden. In NLP, Mamba has surfaced as an efficient alternative for transformers. However, Mamba's successes do not trivially extend to computer vision tasks, including those in video analysis. In this paper, we theoretically analyze the differences between self-attention and Mamba. We identify two limitations in Mamba's token processing: historical decay and element contradiction. We propose VideoMambaPro (VMP) that solves the identified limitations by adding masked backward computation and elemental residual connections to a VideoMamba backbone. VideoMambaPro shows state-of-the-art video action recognition performance compared to transformer models, and surpasses VideoMamba by clear margins: 7.9% and 8.1% top-1 on Kinetics-400 and Something-Something V2, respectively. Our VideoMambaPro-M model achieves 91.9% top-1 on Kinetics-400, only 0.2% below InternVideo2-6B but with only 1.2% of its parameters. The combination of high performance and efficiency makes VideoMambaPro an interesting alternative for transformer models.
Authors: Jiangtong Tan, Feng Zhao
Abstract: Image restoration has made marvelous progress with the advent of deep learning. Previous methods usually rely on designing powerful network architecture to elevate performance, however, the natural visual effect of the restored results is limited by color and texture distortions. Besides the visual perceptual quality, the semantic perception recovery is an important but often overlooked perspective of restored image, which is crucial for the deployment in high-level tasks. In this paper, we propose a new perspective to resort these issues by introducing a naturalness-oriented and semantic-aware optimization mechanism, dubbed DiffLoss. Specifically, inspired by the powerful distribution coverage capability of the diffusion model for natural image generation, we exploit the Markov chain sampling property of diffusion model and project the restored results of existing networks into the sampling space. Besides, we reveal that the bottleneck feature of diffusion models, also dubbed h-space feature, is a natural high-level semantic space. We delve into this property and propose a semantic-aware loss to further unlock its potential of semantic perception recovery, which paves the way to connect image restoration task and downstream high-level recognition task. With these two strategies, the DiffLoss can endow existing restoration methods with both more natural and semantic-aware results. We verify the effectiveness of our method on substantial common image restoration tasks and benchmarks. Code will be available at https://github.com/JosephTiTan/DiffLoss.
Authors: Yang Song, Lin Wang
Abstract: 3D object detection is an important task that has been widely applied in autonomous driving. Recently, fusing multi-modal inputs, i.e., LiDAR and camera data, to perform this task has become a new trend. Existing methods, however, either ignore the sparsity of Lidar features or fail to preserve the original spatial structure of LiDAR and the semantic density of camera features simultaneously due to the modality gap. To address issues, this letter proposes a novel bidirectional complementary Lidar-camera fusion framework, called BiCo-Fusion that can achieve robust semantic- and spatial-aware 3D object detection. The key insight is to mutually fuse the multi-modal features to enhance the semantics of LiDAR features and the spatial awareness of the camera features and adaptatively select features from both modalities to build a unified 3D representation. Specifically, we introduce Pre-Fusion consisting of a Voxel Enhancement Module (VEM) to enhance the semantics of voxel features from 2D camera features and Image Enhancement Module (IEM) to enhance the spatial characteristics of camera features from 3D voxel features. Both VEM and IEM are bidirectionally updated to effectively reduce the modality gap. We then introduce Unified Fusion to adaptively weight to select features from the enchanted Lidar and camera features to build a unified 3D representation. Extensive experiments demonstrate the superiority of our BiCo-Fusion against the prior arts. Project page: https://t-ys.github.io/BiCo-Fusion/.
Authors: Ming Chen, Yuxuan Cheng, Xinwei He, Xinyue Wang, Yan Aze, Jinhai Xiang
Abstract: Integrating visible and infrared images into one high-quality image, also known as visible and infrared image fusion, is a challenging yet critical task for many downstream vision tasks. Most existing works utilize pretrained deep neural networks or design sophisticated frameworks with strong priors for this task, which may be unsuitable or lack flexibility. This paper presents SimpleFusion, a simple yet effective framework for visible and infrared image fusion. Our framework follows the decompose-and-fusion paradigm, where the visible and the infrared images are decomposed into reflectance and illumination components via Retinex theory and followed by the fusion of these corresponding elements. The whole framework is designed with two plain convolutional neural networks without downsampling, which can perform image decomposition and fusion efficiently. Moreover, we introduce decomposition loss and a detail-to-semantic loss to preserve the complementary information between the two modalities for fusion. We conduct extensive experiments on the challenging benchmarks, verifying the superiority of our method over previous state-of-the-arts. Code is available at \href{https://github.com/hxwxss/SimpleFusion-A-Simple-Fusion-Framework-for-Infrared-and-Visible-Images}{https://github.com/hxwxss/SimpleFusion-A-Simple-Fusion-Framework-for-Infrared-and-Visible-Images}
URLs: https://github.com/hxwxss/SimpleFusion-A-Simple-Fusion-Framework-for-Infrared-and-Visible-Images, https://github.com/hxwxss/SimpleFusion-A-Simple-Fusion-Framework-for-Infrared-and-Visible-Images
Authors: Fuseini Mumuni, Alhassan Mumuni
Abstract: Grounding DINO and the Segment Anything Model (SAM) have achieved impressive performance in zero-shot object detection and image segmentation, respectively. Together, they have a great potential in revolutionizing zero-shot semantic segmentation or data annotation. Yet, in specialized domains like medical image segmentation, objects of interest (e.g., organs, tissues, and tumors) may not fall in existing class names. To address this problem, the referring expression comprehension (REC) ability of Grounding DINO is leveraged to detect arbitrary targets by their language descriptions. However, recent studies have highlighted severe limitation of the REC framework in this application setting owing to its tendency to make false positive predictions when the target is absent in the given image. And, while this bottleneck is central to the prospect of open-set semantic segmentation, it is still largely unknown how much improvement can be achieved by studying the prediction errors. To this end, we perform empirical studies on eight publicly available datasets and reveal that these errors consistently follow a predictable pattern and can, thus, be mitigated by a simple strategy. Specifically, we show that these false positive detections with appreciable confidence scores generally occupy large image areas and can usually be filtered by their relative sizes. More importantly, we expect these observations to inspire future research in improving REC-based detection and automated segmentation. Using this technique, we evaluate the performance of SAM on multiple datasets from various specialized domains and report significant improvement in segmentation performance and annotation time savings over manual approaches.
Authors: Yixin Xuan, Xinyang Li, Gongxin Yao, Shiwei Zhou, Donghui Sun, Xiaoxin Chen, Yu Pan
Abstract: High-fidelity reconstruction of 3D human avatars has a wild application in visual reality. In this paper, we introduce FAGhead, a method that enables fully controllable human portraits from monocular videos. We explicit the traditional 3D morphable meshes (3DMM) and optimize the neutral 3D Gaussians to reconstruct with complex expressions. Furthermore, we employ a novel Point-based Learnable Representation Field (PLRF) with learnable Gaussian point positions to enhance reconstruction performance. Meanwhile, to effectively manage the edges of avatars, we introduced the alpha rendering to supervise the alpha value of each pixel. Extensive experimental results on the open-source datasets and our capturing datasets demonstrate that our approach is able to generate high-fidelity 3D head avatars and fully control the expression and pose of the virtual avatars, which is outperforming than existing works.
Authors: Florian P. Mahner, Lukas Muttenthaler, Umut G\"u\c{c}l\"u, Martin N. Hebart
Abstract: Determining the similarities and differences between humans and artificial intelligence is an important goal both in machine learning and cognitive neuroscience. However, similarities in representations only inform us about the degree of alignment, not the factors that determine it. Drawing upon recent developments in cognitive science, we propose a generic framework for yielding comparable representations in humans and deep neural networks (DNN). Applying this framework to humans and a DNN model of natural images revealed a low-dimensional DNN embedding of both visual and semantic dimensions. In contrast to humans, DNNs exhibited a clear dominance of visual over semantic features, indicating divergent strategies for representing images. While in-silico experiments showed seemingly-consistent interpretability of DNN dimensions, a direct comparison between human and DNN representations revealed substantial differences in how they process images. By making representations directly comparable, our results reveal important challenges for representational alignment, offering a means for improving their comparability.
Authors: Jiaxin Zhang, Wentao Yang, Songxuan Lai, Zecheng Xie, Lianwen Jin
Abstract: Current multimodal large language models (MLLMs) face significant challenges in visual document understanding (VDU) tasks due to the high resolution, dense text, and complex layouts typical of document images. These characteristics demand a high level of detail perception ability from MLLMs. While increasing input resolution improves detail perception, it also leads to longer sequences of visual tokens, increasing computational costs and straining the models' ability to handle long contexts. To address these challenges, we introduce DocKylin, a document-centric MLLM that performs visual content slimming at both the pixel and token levels, thereby reducing token sequence length in VDU scenarios. DocKylin utilizes an Adaptive Pixel Slimming (APS) preprocessing module to perform pixel-level slimming, increasing the proportion of informative pixels. Moreover, DocKylin incorporates a novel Dynamic Token Slimming (DTS) module to conduct token-level slimming, filtering essential tokens and removing others to create a compressed, adaptive visual sequence. Experiments demonstrate DocKylin's promising performance across various VDU benchmarks. Notably, both the proposed APS and DTS are parameter-free, facilitating easy integration into existing MLLMs, and our experiments indicate their potential for broader applications.
Authors: Yogesh Aggarwal, Prithwijit Guha
Abstract: Face detection is frequently attempted by using heavy pre-trained backbone networks like ResNet-50/101/152 and VGG16/19. Few recent works have also proposed lightweight detectors with customized backbones, novel loss functions and efficient training strategies. The novelty of this work lies in the design of a lightweight detector while training with only the commonly used loss functions and learning strategies. The proposed face detector grossly follows the established RetinaFace architecture. The first contribution of this work is the design of a customized lightweight backbone network (BLite) having 0.167M parameters with 0.52 GFLOPs. The second contribution is the use of two independent multi-task losses. The proposed lightweight face detector (FDLite) has 0.26M parameters with 0.94 GFLOPs. The network is trained on the WIDER FACE dataset. FDLite is observed to achieve 92.3\%, 89.8\%, and 82.2\% Average Precision (AP) on the easy, medium, and hard subsets of the WIDER FACE validation dataset, respectively.
Authors: Yibo Gao, Zheyao Gao, Xin Gao, Yuanye Liu, Bomin Wang, Xiahai Zhuang
Abstract: Due to the high stakes in medical decision-making, there is a compelling demand for interpretable deep learning methods in medical image analysis. Concept Bottleneck Models (CBM) have emerged as an active interpretable framework incorporating human-interpretable concepts into decision-making. However, their concept predictions may lack reliability when applied to clinical diagnosis, impeding concept explanations' quality. To address this, we propose an evidential Concept Embedding Model (evi-CEM), which employs evidential learning to model the concept uncertainty. Additionally, we offer to leverage the concept uncertainty to rectify concept misalignments that arise when training CBMs using vision-language models without complete concept supervision. With the proposed methods, we can enhance concept explanations' reliability for both supervised and label-efficient settings. Furthermore, we introduce concept uncertainty for effective test-time intervention. Our evaluation demonstrates that evi-CEM achieves superior performance in terms of concept prediction, and the proposed concept rectification effectively mitigates concept misalignments for label-efficient training. Our code is available at https://github.com/obiyoag/evi-CEM.
Authors: Meiqi Chen, Bo Peng, Yan Zhang, Chaochao Lu
Abstract: Causal reasoning is fundamental to human intelligence and crucial for effective decision-making in real-world environments. Despite recent advancements in large vision-language models (LVLMs), their ability to comprehend causality remains unclear. Previous work typically focuses on commonsense causality between events and/or actions, which is insufficient for applications like embodied agents and lacks the explicitly defined causal graphs required for formal causal reasoning. To overcome these limitations, we introduce a fine-grained and unified definition of causality involving interactions between humans and/or objects. Building on the definition, we construct a novel dataset, CELLO, consisting of 14,094 causal questions across all four levels of causality: discovery, association, intervention, and counterfactual. This dataset surpasses traditional commonsense causality by including explicit causal graphs that detail the interactions between humans and objects. Extensive experiments on CELLO reveal that current LVLMs still struggle with causal reasoning tasks, but they can benefit significantly from our proposed CELLO-CoT, a causally inspired chain-of-thought prompting strategy. Both quantitative and qualitative analyses from this study provide valuable insights for future research. Our project page is at https://github.com/OpenCausaLab/CELLO.
Authors: Kit Mills Bransby, Arian Beqiri, Woo-Jin Cho Kim, Jorge Oliveira, Agisilaos Chartsias, Alberto Gomez
Abstract: Neural networks can learn spurious correlations that lead to the correct prediction in a validation set, but generalise poorly because the predictions are right for the wrong reason. This undesired learning of naive shortcuts (Clever Hans effect) can happen for example in echocardiogram view classification when background cues (e.g. metadata) are biased towards a class and the model learns to focus on those background features instead of on the image content. We propose a simple, yet effective random background augmentation method called BackMix, which samples random backgrounds from other examples in the training set. By enforcing the background to be uncorrelated with the outcome, the model learns to focus on the data within the ultrasound sector and becomes invariant to the regions outside this. We extend our method in a semi-supervised setting, finding that the positive effects of BackMix are maintained with as few as 5% of segmentation labels. A loss weighting mechanism, wBackMix, is also proposed to increase the contribution of the augmented examples. We validate our method on both in-distribution and out-of-distribution datasets, demonstrating significant improvements in classification accuracy, region focus and generalisability. Our source code is available at: https://github.com/kitbransby/BackMix
Authors: Varun Nagaraj Rao, Siddharth Choudhary, Aditya Deshpande, Ravi Kumar Satzoda, Srikar Appalaraju
Abstract: The scaling of large language models to encode all the world's knowledge in model parameters is unsustainable and has exacerbated resource barriers. Retrieval-Augmented Generation (RAG) presents a potential solution, yet its application to vision-language models (VLMs) is under explored. Existing methods focus on models designed for single tasks. Furthermore, they're limited by the need for resource intensive pre training, additional parameter requirements, unaddressed modality prioritization and lack of clear benefit over non-retrieval baselines. This paper introduces RAVEN, a multitask retrieval augmented VLM framework that enhances base VLMs through efficient, task specific fine-tuning. By integrating retrieval augmented samples without the need for additional retrieval-specific parameters, we show that the model acquires retrieval properties that are effective across multiple tasks. Our results and extensive ablations across retrieved modalities for the image captioning and VQA tasks indicate significant performance improvements compared to non retrieved baselines +1 CIDEr on MSCOCO, +4 CIDEr on NoCaps and nearly a +3\% accuracy on specific VQA question types. This underscores the efficacy of applying RAG approaches to VLMs, marking a stride toward more efficient and accessible multimodal learning.
Authors: Lennart Bruns, Lucas Lamparter, Milos Galic, Xiaoyi Jiang
Abstract: In this paper we study the problem of estimating the migration direction of cells based on a single image. To the best of our knowledge, there is only one related work that uses a classification CNN for four classes (quadrants). This approach does not allow detailed directional resolution. We solve the single image estimation problem using deep circular regression with special attention to cycle-sensitive methods. On two databases we achieve an average accuracy of $\sim$17 degrees, which is a significant improvement over the previous work.
Authors: Zhimin Shao, Jialang Xu, Danail Stoyanov, Evangelos B. Mazomenos, Yueming Jin
Abstract: Despite significant advancements in robotic systems and surgical data science, ensuring safe and optimal execution in robot-assisted minimally invasive surgery (RMIS) remains a complex challenge. Current surgical error detection methods involve two parts: identifying surgical gestures and then detecting errors within each gesture clip. These methods seldom consider the rich contextual and semantic information inherent in surgical videos, limiting their performance due to reliance on accurate gesture identification. Motivated by the chain-of-thought prompting in natural language processing, this letter presents a novel and real-time end-to-end error detection framework, Chain-of-Thought (COG) prompting, leveraging contextual information from surgical videos. This encompasses two reasoning modules designed to mimic the decision-making processes of expert surgeons. Concretely, we first design a Gestural-Visual Reasoning module, which utilizes transformer and attention architectures for gesture prompting, while the second, a Multi-Scale Temporal Reasoning module, employs a multi-stage temporal convolutional network with both slow and fast paths for temporal information extraction. We extensively validate our method on the public benchmark RMIS dataset JIGSAWS. Our method encapsulates the reasoning processes inherent to surgical activities enabling it to outperform the state-of-the-art by 4.6% in F1 score, 4.6% in Accuracy, and 5.9% in Jaccard index while processing each frame in 6.69 milliseconds on average, demonstrating the great potential of our approach in enhancing the safety and efficacy of RMIS procedures and surgical education. The code will be available.
Authors: Nazanin Moradinasab, Laura S. Shankman, Rebecca A. Deaton, Gary K. Owens, Donald E. Brown
Abstract: Domain adaptive semantic segmentation aims to generate accurate and dense predictions for an unlabeled target domain by leveraging a supervised model trained on a labeled source domain. The prevalent self-training approach involves retraining the dense discriminative classifier of $p(class|pixel feature)$ using the pseudo-labels from the target domain. While many methods focus on mitigating the issue of noisy pseudo-labels, they often overlook the underlying data distribution p(pixel feature|class) in both the source and target domains. To address this limitation, we propose the multi-prototype Gaussian-Mixture-based (ProtoGMM) model, which incorporates the GMM into contrastive losses to perform guided contrastive learning. Contrastive losses are commonly executed in the literature using memory banks, which can lead to class biases due to underrepresented classes. Furthermore, memory banks often have fixed capacities, potentially restricting the model's ability to capture diverse representations of the target/source domains. An alternative approach is to use global class prototypes (i.e. averaged features per category). However, the global prototypes are based on the unimodal distribution assumption per class, disregarding within-class variation. To address these challenges, we propose the ProtoGMM model. This novel approach involves estimating the underlying multi-prototype source distribution by utilizing the GMM on the feature space of the source samples. The components of the GMM model act as representative prototypes. To achieve increased intra-class semantic similarity, decreased inter-class similarity, and domain alignment between the source and target domains, we employ multi-prototype contrastive learning between source distribution and target samples. The experiments show the effectiveness of our method on UDA benchmarks.
Authors: Timin Gao, Wensheng Pan, Yan Zhang, Sicheng Zhao, Shengchuan Zhang, Xiawu Zheng, Ke Li, Liujuan Cao, Rongrong Ji
Abstract: Contrastive learning has considerably advanced the field of Image Quality Assessment (IQA), emerging as a widely adopted technique. The core mechanism of contrastive learning involves minimizing the distance between quality-similar (positive) examples while maximizing the distance between quality-dissimilar (negative) examples. Despite its successes, current contrastive learning methods often neglect the importance of preserving the local manifold structure. This oversight can result in a high degree of similarity among hard examples within the feature space, thereby impeding effective differentiation and assessment. To address this issue, we propose an innovative framework that integrates local manifold learning with contrastive learning for No-Reference Image Quality Assessment (NR-IQA). Our method begins by sampling multiple crops from a given image, identifying the most visually salient crop. This crop is then used to cluster other crops from the same image as the positive class, while crops from different images are treated as negative classes to increase inter-class distance. Uniquely, our approach also considers non-saliency crops from the same image as intra-class negative classes to preserve their distinctiveness. Additionally, we employ a mutual learning framework, which further enhances the model's ability to adaptively learn and identify visual saliency regions. Our approach demonstrates a better performance compared to state-of-the-art methods in 7 standard datasets, achieving PLCC values of 0.942 (compared to 0.908 in TID2013) and 0.914 (compared to 0.894 in LIVEC).
Authors: Hao Fei, Shengqiong Wu, Meishan Zhang, Min Zhang, Tat-Seng Chua, Shuicheng Yan
Abstract: While pre-training large-scale video-language models (VLMs) has shown remarkable potential for various downstream video-language tasks, existing VLMs can still suffer from certain commonly seen limitations, e.g., coarse-grained cross-modal aligning , under-modeling of temporal dynamics, detached video-language view. In this work, we target enhancing VLMs with a fine-grained structural spatio-temporal alignment learning method (namely Finsta). First of all, we represent the input texts and videos with fine-grained scene graph (SG) structures, both of which are further unified into a holistic SG (HSG) for bridging two modalities. Then, an SG-based framework is built, where the textual SG (TSG) is encoded with a graph Transformer, while the video dynamic SG (DSG) and the HSG are modeled with a novel recurrent graph Transformer for spatial and temporal feature propagation. A spatial-temporal Gaussian differential graph Transformer is further devised to strengthen the sense of the changes in objects across spatial and temporal dimensions. Next, based on the fine-grained structural features of TSG and DSG, we perform object-centered spatial alignment and predicate-centered temporal alignment respectively, enhancing the video-language grounding in both the spatiality and temporality. We design our method as a plug&play system, which can be integrated into existing well-trained VLMs for further representation augmentation, without training from scratch or relying on SG annotations in downstream applications. On 6 representative VL modeling tasks over 12 datasets in both standard and long-form video scenarios, Finsta consistently improves the existing 13 strong-performing VLMs persistently, and refreshes the current state-of-the-art end task performance significantly in both the fine-tuning and zero-shot settings.
Authors: Junying Chen, Ruyi Ouyang, Anningzhe Gao, Shunian Chen, Guiming Hardy Chen, Xidong Wang, Ruifei Zhang, Zhenyang Cai, Ke Ji, Guangjun Yu, Xiang Wan, Benyou Wang
Abstract: The rapid development of multimodal large language models (MLLMs), such as GPT-4V, has led to significant advancements. However, these models still face challenges in medical multimodal capabilities due to limitations in the quantity and quality of medical vision-text data, stemming from data privacy concerns and high annotation costs. While pioneering approaches utilize PubMed's large-scale, de-identified medical image-text pairs to address these limitations, they still fall short due to inherent data noise. To tackle this, we refined medical image-text pairs from PubMed and employed MLLMs (GPT-4V) in an 'unblinded' capacity to denoise and reformat the data, resulting in the creation of the PubMedVision dataset with 1.3 million medical VQA samples. Our validation demonstrates that: (1) PubMedVision can significantly enhance the medical multimodal capabilities of current MLLMs, showing significant improvement in benchmarks including the MMMU Health & Medicine track; (2) manual checks by medical experts and empirical results validate the superior data quality of our dataset compared to other data construction methods. Using PubMedVision, we train a 34B medical MLLM HuatuoGPT-Vision, which shows superior performance in medical multimodal scenarios among open-source MLLMs.
Authors: Pawel Knap
Abstract: Human modelling and pose estimation stands at the crossroads of Computer Vision, Computer Graphics, and Machine Learning. This paper presents a thorough investigation of this interdisciplinary field, examining various algorithms, methodologies, and practical applications. It explores the diverse range of sensor technologies relevant to this domain and delves into a wide array of application areas. Additionally, we discuss the challenges and advancements in 2D and 3D human modelling methodologies, along with popular datasets, metrics, and future research directions. The main contribution of this paper lies in its up-to-date comparison of state-of-the-art (SOTA) human pose estimation algorithms in both 2D and 3D domains. By providing this comprehensive overview, the paper aims to enhance understanding of 3D human modelling and pose estimation, offering insights into current SOTA achievements, challenges, and future prospects within the field.
Authors: Malvina Nikandrou, Georgios Pantazopoulos, Ioannis Konstas, Alessandro Suglia
Abstract: Continual learning focuses on incrementally training a model on a sequence of tasks with the aim of learning new tasks while minimizing performance drop on previous tasks. Existing approaches at the intersection of Continual Learning and Visual Question Answering (VQA) do not study how the multimodal nature of the input affects the learning dynamics of a model. In this paper, we demonstrate that each modality evolves at different rates across a continuum of tasks and that this behavior occurs in established encoder-only models as well as modern recipes for developing Vision & Language (VL) models. Motivated by this observation, we propose a modality-aware feature distillation (MAFED) approach which outperforms existing baselines across models of varying scale in three multimodal continual learning settings. Furthermore, we provide ablations showcasing that modality-aware distillation complements experience replay. Overall, our results emphasize the importance of addressing modality-specific dynamics to prevent forgetting in multimodal continual learning.
Authors: Jocelin Su, Nan Liu, Yanbo Wang, Joshua B. Tenenbaum, Yilun Du
Abstract: Given an image of a natural scene, we are able to quickly decompose it into a set of components such as objects, lighting, shadows, and foreground. We can then envision a scene where we combine certain components with those from other images, for instance a set of objects from our bedroom and animals from a zoo under the lighting conditions of a forest, even if we have never encountered such a scene before. In this paper, we present a method to decompose an image into such compositional components. Our approach, Decomp Diffusion, is an unsupervised method which, when given a single image, infers a set of different components in the image, each represented by a diffusion model. We demonstrate how components can capture different factors of the scene, ranging from global scene descriptors like shadows or facial expression to local scene descriptors like constituent objects. We further illustrate how inferred factors can be flexibly composed, even with factors inferred from other models, to generate a variety of scenes sharply different than those seen in training time. Website and code at https://energy-based-model.github.io/decomp-diffusion.
URLs: https://energy-based-model.github.io/decomp-diffusion.
Authors: Sonam Gupta, Snehal Singh Tomar, Grigorios G Chrysos, Sukhendu Das, A. N. Rajagopalan
Abstract: Extracting Implicit Neural Representations (INRs) on video data poses unique challenges due to the additional temporal dimension. In the context of videos, INRs have predominantly relied on a frame-only parameterization, which sacrifices the spatiotemporal continuity observed in pixel-level (spatial) representations. To mitigate this, we introduce Polynomial Neural Representation for Videos (PNeRV), a parameter-wise efficient, patch-wise INR for videos that preserves spatiotemporal continuity. PNeRV leverages the modeling capabilities of Polynomial Neural Networks to perform the modulation of a continuous spatial (patch) signal with a continuous time (frame) signal. We further propose a custom Hierarchical Patch-wise Spatial Sampling Scheme that ensures spatial continuity while retaining parameter efficiency. We also employ a carefully designed Positional Embedding methodology to further enhance PNeRV's performance. Our extensive experimentation demonstrates that PNeRV outperforms the baselines in conventional Implicit Neural Representation tasks like compression along with downstream applications that require spatiotemporal continuity in the underlying representation. PNeRV not only addresses the challenges posed by video data in the realm of INRs but also opens new avenues for advanced video processing and analysis.
Authors: Burak Ekim, Michael Schmitt
Abstract: In recent decades, the causes and consequences of climate change have accelerated, affecting our planet on an unprecedented scale. This change is closely tied to the ways in which humans alter their surroundings. As our actions continue to impact natural areas, using satellite images to observe and measure these effects has become crucial for understanding and combating climate change. Aiming to map land naturalness on the continuum of modern human pressure, we have developed a multi-modal supervised deep learning framework that addresses the unique challenges of satellite data and the task at hand. We incorporate contextual and geographical priors, represented by corresponding coordinate information and broader contextual information, including and surrounding the immediate patch to be predicted. Our framework improves the model's predictive performance in mapping land naturalness from Sentinel-2 data, a type of multi-spectral optical satellite imagery. Recognizing that our protective measures are only as effective as our understanding of the ecosystem, quantifying naturalness serves as a crucial step toward enhancing our environmental stewardship.
Authors: KuanChao Chu, Satoshi Yamazaki, Hideki Nakayama
Abstract: This work focuses on training dataset enhancement of informative relational triplets for Scene Graph Generation (SGG). Due to the lack of effective supervision, the current SGG model predictions perform poorly for informative relational triplets with inadequate training samples. Therefore, we propose two novel training dataset enhancement modules: Feature Space Triplet Augmentation (FSTA) and Soft Transfer. FSTA leverages a feature generator trained to generate representations of an object in relational triplets. The biased prediction based sampling in FSTA efficiently augments artificial triplets focusing on the challenging ones. In addition, we introduce Soft Transfer, which assigns soft predicate labels to general relational triplets to make more supervisions for informative predicate classes effectively. Experimental results show that integrating FSTA and Soft Transfer achieve high levels of both Recall and mean Recall in Visual Genome dataset. The mean of Recall and mean Recall is the highest among all the existing model-agnostic methods.
Authors: Yushun Tang, Shuoshuo Chen, Zhehan Kan, Yi Zhang, Qinghai Guo, Zhihai He
Abstract: Fully test-time adaptation aims to adapt the network model based on sequential analysis of input samples during the inference stage to address the cross-domain performance degradation problem of deep neural networks. This work is based on the following interesting finding: in transformer-based image classification, the class token at the first transformer encoder layer can be learned to capture the domain-specific characteristics of target samples during test-time adaptation. This learned token, when combined with input image patch embeddings, is able to gradually remove the domain-specific information from the feature representations of input samples during the transformer encoding process, thereby significantly improving the test-time adaptation performance of the source model across different domains. We refer to this class token as visual conditioning token (VCT). To successfully learn the VCT, we propose a bi-level learning approach to capture the long-term variations of domain-specific characteristics while accommodating local variations of instance-specific characteristics. Experimental results on the benchmark datasets demonstrate that our proposed bi-level visual conditioning token learning method is able to achieve significantly improved test-time adaptation performance by up to 1.9%.
Authors: Chengwen Zhang, Yun Liu, Ruofan Xing, Bingda Tang, Li Yi
Abstract: Understanding how humans cooperatively rearrange household objects is critical for VR/AR and human-robot interaction. However, in-depth studies on modeling these behaviors are under-researched due to the lack of relevant datasets. We fill this gap by presenting CORE4D, a novel large-scale 4D human-object-human interaction dataset focusing on collaborative object rearrangement, which encompasses diverse compositions of various object geometries, collaboration modes, and 3D scenes. With 1K human-object-human motion sequences captured in the real world, we enrich CORE4D by contributing an iterative collaboration retargeting strategy to augment motions to a variety of novel objects. Leveraging this approach, CORE4D comprises a total of 11K collaboration sequences spanning 3K real and virtual object shapes. Benefiting from extensive motion patterns provided by CORE4D, we benchmark two tasks aiming at generating human-object interaction: human-object motion forecasting and interaction synthesis. Extensive experiments demonstrate the effectiveness of our collaboration retargeting strategy and indicate that CORE4D has posed new challenges to existing human-object interaction generation methodologies. Our dataset and code are available at https://github.com/leolyliu/CORE4D-Instructions.
Authors: Yanan Zhang, Chao Zhou, Di Huang
Abstract: Existing 3D object detection suffers from expensive annotation costs and poor transferability to unknown data due to the domain gap, Unsupervised Domain Adaptation (UDA) aims to generalize detection models trained in labeled source domains to perform robustly on unexplored target domains, providing a promising solution for cross-domain 3D object detection. Although Self-Training (ST) based cross-domain 3D detection methods with the assistance of pseudo-labeling techniques have achieved remarkable progress, they still face the issue of low-quality pseudo-labels when there are significant domain disparities due to the absence of a process for feature distribution alignment. While Adversarial Learning (AL) based methods can effectively align the feature distributions of the source and target domains, the inability to obtain labels in the target domain forces the adoption of asymmetric optimization losses, resulting in a challenging issue of source domain bias. To overcome these limitations, we propose a novel unsupervised domain adaptation framework for 3D object detection via collaborating ST and AL, dubbed as STAL3D, unleashing the complementary advantages of pseudo labels and feature distribution alignment. Additionally, a Background Suppression Adversarial Learning (BS-AL) module and a Scale Filtering Module (SFM) are designed tailored for 3D cross-domain scenes, effectively alleviating the issues of the large proportion of background interference and source domain size bias. Our STAL3D achieves state-of-the-art performance on multiple cross-domain tasks and even surpasses the Oracle results on Waymo $\rightarrow$ KITTI and Waymo $\rightarrow$ KITTI-rain.
Authors: Yuxin Xie, Tao Zhou, Yi Zhou, Geng Chen
Abstract: Weakly-supervised medical image segmentation is a challenging task that aims to reduce the annotation cost while keep the segmentation performance. In this paper, we present a novel framework, SimTxtSeg, that leverages simple text cues to generate high-quality pseudo-labels and study the cross-modal fusion in training segmentation models, simultaneously. Our contribution consists of two key components: an effective Textual-to-Visual Cue Converter that produces visual prompts from text prompts on medical images, and a text-guided segmentation model with Text-Vision Hybrid Attention that fuses text and image features. We evaluate our framework on two medical image segmentation tasks: colonic polyp segmentation and MRI brain tumor segmentation, and achieve consistent state-of-the-art performance.
Authors: Haobo Yuan, Xiangtai Li, Lu Qi, Tao Zhang, Ming-Hsuan Yang, Shuicheng Yan, Chen Change Loy
Abstract: Transformer-based segmentation methods face the challenge of efficient inference when dealing with high-resolution images. Recently, several linear attention architectures, such as Mamba and RWKV, have attracted much attention as they can process long sequences efficiently. In this work, we focus on designing an efficient segment-anything model by exploring these different architectures. Specifically, we design a mixed backbone that contains convolution and RWKV operation, which achieves the best for both accuracy and efficiency. In addition, we design an efficient decoder to utilize the multiscale tokens to obtain high-quality masks. We denote our method as RWKV-SAM, a simple, effective, fast baseline for SAM-like models. Moreover, we build a benchmark containing various high-quality segmentation datasets and jointly train one efficient yet high-quality segmentation model using this benchmark. Based on the benchmark results, our RWKV-SAM achieves outstanding performance in efficiency and segmentation quality compared to transformers and other linear attention models. For example, compared with the same-scale transformer model, RWKV-SAM achieves more than 2x speedup and can achieve better segmentation performance on various datasets. In addition, RWKV-SAM outperforms recent vision Mamba models with better classification and semantic segmentation results. Code and models will be publicly available.
Authors: Tao Zhang, Xiangtai Li, Hao Fei, Haobo Yuan, Shengqiong Wu, Shunping Ji, Chen Change Loy, Shuicheng Yan
Abstract: Current universal segmentation methods demonstrate strong capabilities in pixel-level image and video understanding. However, they lack reasoning abilities and cannot be controlled via text instructions. In contrast, large vision-language multimodal models exhibit powerful vision-based conversation and reasoning capabilities but lack pixel-level understanding and have difficulty accepting visual prompts for flexible user interaction. This paper proposes OMG-LLaVA, a new and elegant framework combining powerful pixel-level vision understanding with reasoning abilities. It can accept various visual and text prompts for flexible user interaction. Specifically, we use a universal segmentation method as the visual encoder, integrating image information, perception priors, and visual prompts into visual tokens provided to the LLM. The LLM is responsible for understanding the user's text instructions and providing text responses and pixel-level segmentation results based on the visual information. We propose perception prior embedding to better integrate perception priors with image features. OMG-LLaVA achieves image-level, object-level, and pixel-level reasoning and understanding in a single model, matching or surpassing the performance of specialized methods on multiple benchmarks. Rather than using LLM to connect each specialist, our work aims at end-to-end training on one encoder, one decoder, and one LLM. The code and model have been released for further research.
Authors: John Lambert, Yuguang Li, Ivaylo Boyadzhiev, Lambert Wixson, Manjunath Narayana, Will Hutchcroft, James Hays, Frank Dellaert, Sing Bing Kang
Abstract: We propose a new system for automatic 2D floorplan reconstruction that is enabled by SALVe, our novel pairwise learned alignment verifier. The inputs to our system are sparsely located 360$^\circ$ panoramas, whose semantic features (windows, doors, and openings) are inferred and used to hypothesize pairwise room adjacency or overlap. SALVe initializes a pose graph, which is subsequently optimized using GTSAM. Once the room poses are computed, room layouts are inferred using HorizonNet, and the floorplan is constructed by stitching the most confident layout boundaries. We validate our system qualitatively and quantitatively as well as through ablation studies, showing that it outperforms state-of-the-art SfM systems in completeness by over 200%, without sacrificing accuracy. Our results point to the significance of our work: poses of 81% of panoramas are localized in the first 2 connected components (CCs), and 89% in the first 3 CCs. Code and models are publicly available at https://github.com/zillow/salve.
Authors: Ali Khaleghi Rahimian, Manish Kumar Govind, Subhajit Maity, Dominick Reilly, Christian K\"ummerle, Srijan Das, Aritra Dutta
Abstract: Visual perception tasks are predominantly solved by Vision Transformer (ViT) architectures, which, despite their effectiveness, encounter a computational bottleneck due to the quadratic complexity of computing self-attention. This inefficiency is largely due to the self-attention heads capturing redundant token interactions, reflecting inherent redundancy within visual data. Many works have aimed to reduce the computational complexity of self-attention in ViTs, leading to the development of efficient and sparse transformer architectures. In this paper, viewing through the efficiency lens, we realized that introducing any sparse self-attention strategy in ViTs can keep the computational overhead low. However, these strategies are sub-optimal as they often fail to capture fine-grained visual details. This observation leads us to propose a general, efficient, sparse architecture, named Fibottention, for approximating self-attention with superlinear complexity that is built upon Fibonacci sequences. The key strategies in Fibottention include: it excludes proximate tokens to reduce redundancy, employs structured sparsity by design to decrease computational demands, and incorporates inception-like diversity across attention heads. This diversity ensures the capture of complementary information through non-overlapping token interactions, optimizing both performance and resource utilization in ViTs for visual representation learning. We embed our Fibottention mechanism into multiple state-of-the-art transformer architectures dedicated to visual tasks. Leveraging only 2-6% of the elements in the self-attention heads, Fibottention in conjunction with ViT and its variants, consistently achieves significant performance boosts compared to standard ViTs in nine datasets across three domains $\unicode{x2013}$ image classification, video understanding, and robot learning tasks.
Authors: Jr-Jen Chen, Yu-Chien Liao, Hsi-Che Lin, Yu-Chu Yu, Yen-Chun Chen, Yu-Chiang Frank Wang
Abstract: We introduce ReXTime, a benchmark designed to rigorously test AI models' ability to perform temporal reasoning within video events. Specifically, ReXTime focuses on reasoning across time, i.e. human-like understanding when the question and its corresponding answer occur in different video segments. This form of reasoning, requiring advanced understanding of cause-and-effect relationships across video segments, poses significant challenges to even the frontier multimodal large language models. To facilitate this evaluation, we develop an automated pipeline for generating temporal reasoning question-answer pairs, significantly reducing the need for labor-intensive manual annotations. Our benchmark includes 921 carefully vetted validation samples and 2,143 test samples, each manually curated for accuracy and relevance. Evaluation results show that while frontier large language models outperform academic models, they still lag behind human performance by a significant 14.3% accuracy gap. Additionally, our pipeline creates a training dataset of 9,695 machine generated samples without manual effort, which empirical studies suggest can enhance the across-time reasoning via fine-tuning.
Authors: Ankan Bhunia, Changjian Li, Hakan Bilen
Abstract: Automatic anomaly detection based on visual cues holds practical significance in various domains, such as manufacturing and product quality assessment. This paper introduces a new conditional anomaly detection problem, which involves identifying anomalies in a query image by comparing it to a reference shape. To address this challenge, we have created a large dataset, BrokenChairs-180K, consisting of around 180K images, with diverse anomalies, geometries, and textures paired with 8,143 reference 3D shapes. To tackle this task, we have proposed a novel transformer-based approach that explicitly learns the correspondence between the query image and reference 3D shape via feature alignment and leverages a customized attention mechanism for anomaly detection. Our approach has been rigorously evaluated through comprehensive experiments, serving as a benchmark for future research in this domain.
Authors: Liujuan Cao, Jianghang Lin, Zebo Hong, Yunhang Shen, Shaohui Lin, Chao Chen, Rongrong Ji
Abstract: Most WSOD methods rely on traditional object proposals to generate candidate regions and are confronted with unstable training, which easily gets stuck in a poor local optimum. In this paper, we introduce a unified, high-capacity weakly supervised object detection (WSOD) network called HUWSOD, which utilizes a comprehensive self-training framework without needing external modules or additional supervision. HUWSOD innovatively incorporates a self-supervised proposal generator and an autoencoder proposal generator with a multi-rate resampling pyramid to replace traditional object proposals, enabling end-to-end WSOD training and inference. Additionally, we implement a holistic self-training scheme that refines detection scores and coordinates through step-wise entropy minimization and consistency-constraint regularization, ensuring consistent predictions across stochastic augmentations of the same image. Extensive experiments on PASCAL VOC and MS COCO demonstrate that HUWSOD competes with state-of-the-art WSOD methods, eliminating the need for offline proposals and additional data. The peak performance of HUWSOD approaches that of fully-supervised Faster R-CNN. Our findings also indicate that randomly initialized boxes, although significantly different from well-designed offline object proposals, are effective for WSOD training.
Authors: Mohammad Salama, Jonathan Kahana, Eliahu Horwitz, Yedid Hoshen
Abstract: Model inversion and membership inference attacks aim to reconstruct and verify the data which a model was trained on. However, they are not guaranteed to find all training samples as they do not know the size of the training set. In this paper, we introduce a new task: dataset size recovery, that aims to determine the number of samples used to train a model, directly from its weights. We then propose DSiRe, a method for recovering the number of images used to fine-tune a model, in the common case where fine-tuning uses LoRA. We discover that both the norm and the spectrum of the LoRA matrices are closely linked to the fine-tuning dataset size; we leverage this finding to propose a simple yet effective prediction algorithm. To evaluate dataset size recovery of LoRA weights, we develop and release a new benchmark, LoRA-WiSE, consisting of over 25000 weight snapshots from more than 2000 diverse LoRA fine-tuned models. Our best classifier can predict the number of fine-tuning images with a mean absolute error of 0.36 images, establishing the feasibility of this attack.
Authors: Yinqiu Feng, Bo Zhang, Lingxi Xiao, Yutian Yang, Tana Gegen, Zexi Chen
Abstract: In this research, we introduce an innovative method for synthesizing medical images using generative adversarial networks (GANs). Our proposed GANs method demonstrates the capability to produce realistic synthetic images even when trained on a limited quantity of real medical image data, showcasing commendable generalization prowess. To achieve this, we devised a generator and discriminator network architecture founded on deep convolutional neural networks (CNNs), leveraging the adversarial training paradigm for model optimization. Through extensive experimentation across diverse medical image datasets, our method exhibits robust performance, consistently generating synthetic images that closely emulate the structural and textural attributes of authentic medical images.
Authors: Yuxiang Hu, Haowei Yang, Ting Xu, Shuyao He, Jiajie Yuan, Haozhang Deng
Abstract: The diagnosis of brain cancer relies heavily on medical imaging techniques, with MRI being the most commonly used. It is necessary to perform automatic segmentation of brain tumors on MRI images. This project intends to build an MRI algorithm based on U-Net. The residual network and the module used to enhance the context information are combined, and the void space convolution pooling pyramid is added to the network for processing. The brain glioma MRI image dataset provided by cancer imaging archives was experimentally verified. A multi-scale segmentation method based on a weighted least squares filter was used to complete the 3D reconstruction of brain tumors. Thus, the accuracy of three-dimensional reconstruction is further improved. Experiments show that the local texture features obtained by the proposed algorithm are similar to those obtained by laser scanning. The algorithm is improved by using the U-Net method and an accuracy of 0.9851 is obtained. This approach significantly enhances the precision of image segmentation and boosts the efficiency of image classification.
Authors: Qishi Zhan, Dan Sun, Erdi Gao, Yuhan Ma, Yaxin Liang, Haowei Yang
Abstract: This study introduces a novel unsupervised medical image feature extraction method that employs spatial stratification techniques. An objective function based on weight is proposed to achieve the purpose of fast image recognition. The algorithm divides the pixels of the image into multiple subdomains and uses a quadtree to access the image. A technique for threshold optimization utilizing a simplex algorithm is presented. Aiming at the nonlinear characteristics of hyperspectral images, a generalized discriminant analysis algorithm based on kernel function is proposed. In this project, a hyperspectral remote sensing image is taken as the object, and we investigate its mathematical modeling, solution methods, and feature extraction techniques. It is found that different types of objects are independent of each other and compact in image processing. Compared with the traditional linear discrimination method, the result of image segmentation is better. This method can not only overcome the disadvantage of the traditional method which is easy to be affected by light, but also extract the features of the object quickly and accurately. It has important reference significance for clinical diagnosis.
Authors: Tyler Morris, Ziming Liu, Longjian Liu, Xiaopeng Zhao
Abstract: As the number of dementia patients rises, the need for accurate diagnostic procedures rises as well. Current methods, like using an MRI scan, rely on human input, which can be inaccurate. However, the decision logic behind machine learning algorithms and their outputs cannot be explained, as most operate in black-box models. Therefore, to increase the accuracy of diagnosing dementia through MRIs, a convolution neural network has been developed and trained using an open-source database of 6400 MRI scans divided into 4 dementia classes. The model, which attained a 98 percent validation accuracy, was shown to be well fit and able to generalize to new data. Furthermore, to aid in the visualization of the model output, an explainable AI algorithm was developed by visualizing the outputs of individual filters in each convolution layer, which highlighted regions of interest in the scan. These outputs do a great job of identifying the image features that contribute most to the model classification, thus allowing users to visualize and understand the results. Altogether, this combination of the convolution neural network and explainable AI algorithm creates a system that can be used in the medical field to not only aid in the proper classification of dementia but also allow everyone involved to visualize and understand the results.
Authors: Xiaomin Lv, Chong Lai, Liya Ding, Maode Lai, Qingrong Sun
Abstract: Large models have become mainstream, yet their applications in digital pathology still require exploration. Meanwhile renal pathology images play an important role in the diagnosis of renal diseases. We conducted image segmentation and paired corresponding text descriptions based on 60 books for renal pathology, clustering analysis for all image and text description features based on large models, ultimately building a retrieval system based on the semantic features of large models. Based above analysis, we established a knowledge base of 10,317 renal pathology images and paired corresponding text descriptions, and then we evaluated the semantic feature capabilities of 4 large models, including GPT2, gemma, LLma and Qwen, and the image-based feature capabilities of dinov2 large model. Furthermore, we built a semantic retrieval system to retrieve pathological images based on text descriptions, and named RppD (aidp.zjsru.edu.cn).
Authors: Tao Li, Chin-Yi Cheng, Amber Xie, Gang Li, Yang Li
Abstract: Layout design, such as user interface or graphical layout in general, is fundamentally an iterative revision process. Through revising a design repeatedly, the designer converges on an ideal layout. In this paper, we investigate how revision edits from human designer can benefit a multimodal generative model. To do so, we curate an expert dataset that traces how human designers iteratively edit and improve a layout generation with a prompted language goal. Based on such data, we explore various supervised fine-tuning task setups on top of a Gemini multimodal backbone, a large multimodal model. Our results show that human revision plays a critical role in iterative layout refinement. While being noisy, expert revision edits lead our model to a surprisingly strong design FID score ~10 which is close to human performance (~6). In contrast, self-revisions that fully rely on model's own judgement, lead to an echo chamber that prevents iterative improvement, and sometimes leads to generative degradation. Fortunately, we found that providing human guidance plays at early stage plays a critical role in final generation. In such human-in-the-loop scenario, our work paves the way for iterative design revision based on pre-trained large multimodal models.
Authors: Chloe R. Bennett, Heather Cole-Lewis, Stephanie Farquhar, Naama Haamel, Boris Babenko, Oran Lang, Mat Fleck, Ilana Traynis, Charles Lau, Ivor Horn, Courtney Lyles
Abstract: The field of artificial intelligence (AI) is rapidly influencing health and healthcare, but bias and poor performance persists for populations who face widespread structural oppression. Previous work has clearly outlined the need for more rigorous attention to data representativeness and model performance to advance equity and reduce bias. However, there is an opportunity to also improve the explainability of AI by leveraging best practices of social epidemiology and health equity to help us develop hypotheses for associations found. In this paper, we focus on explainable AI (XAI) and describe a framework for interdisciplinary expert panel review to discuss and critically assess AI model explanations from multiple perspectives and identify areas of bias and directions for future research. We emphasize the importance of the interdisciplinary expert panel to produce more accurate, equitable interpretations which are historically and contextually informed. Interdisciplinary panel discussions can help reduce bias, identify potential confounders, and identify opportunities for additional research where there are gaps in the literature. In turn, these insights can suggest opportunities for AI model improvement.
Authors: Aditi Ramaswamy, Melane Navaratnarajah, Hana Chockler
Abstract: With the rise of freely available image generators, AI-generated art has become the centre of a series of heated debates, one of which concerns the concept of human creativity. Can an image generation AI exhibit ``creativity'' of the same type that artists do, and if so, how does that manifest? Our paper attempts to define and empirically measure one facet of creative behavior in AI, by conducting an experiment to quantify the "fluidity of prompt interpretation", or just "fluidity", in a series of selected popular image generators. To study fluidity, we (1) introduce a clear definition for it, (2) create chains of auto-generated prompts and images seeded with an initial "ground-truth: image, (3) measure these chains' breakage points using preexisting visual and semantic metrics, and (4) use both statistical tests and visual explanations to study these chains and determine whether the image generators used to produce them exhibit significant fluidity.
Authors: Esmaeil Seraj, Harsh Bhate, Walter Talamonti
Abstract: The integration of Transparent Displays (TD) in various applications, such as Heads-Up Displays (HUDs) in vehicles, is a burgeoning field, poised to revolutionize user experiences. However, this innovation brings forth significant challenges in realtime human-device interaction, particularly in accurately identifying and tracking a user's gaze on dynamically changing TDs. In this paper, we present a two-fold robust and efficient systematic solution for realtime gaze monitoring, comprised of: (1) a tree-based algorithm for identifying and dynamically tracking gaze targets (i.e., moving, size-changing, and overlapping 2D content) projected on a transparent display, in realtime; (2) a multi-stream self-attention architecture to estimate the depth-level of human gaze from eye tracking data, to account for the display's transparency and preventing undesired interactions with the TD. We collected a real-world eye-tracking dataset to train and test our gaze monitoring system. We present extensive results and ablation studies, including inference experiments on System on Chip (SoC) evaluation boards, demonstrating our model's scalability, precision, and realtime feasibility in both static and dynamic contexts. Our solution marks a significant stride in enhancing next-generation user-device interaction and experience, setting a new benchmark for algorithmic gaze monitoring technology in dynamic transparent displays.
Authors: Georgios Tziafas, Hamidreza Kasaei
Abstract: The ability to grasp objects in-the-wild from open-ended language instructions constitutes a fundamental challenge in robotics. An open-world grasping system should be able to combine high-level contextual with low-level physical-geometric reasoning in order to be applicable in arbitrary scenarios. Recent works exploit the web-scale knowledge inherent in large language models (LLMs) to plan and reason in robotic context, but rely on external vision and action models to ground such knowledge into the environment and parameterize actuation. This setup suffers from two major bottlenecks: a) the LLM's reasoning capacity is constrained by the quality of visual grounding, and b) LLMs do not contain low-level spatial understanding of the world, which is essential for grasping in contact-rich scenarios. In this work we demonstrate that modern vision-language models (VLMs) are capable of tackling such limitations, as they are implicitly grounded and can jointly reason about semantics and geometry. We propose OWG, an open-world grasping pipeline that combines VLMs with segmentation and grasp synthesis models to unlock grounded world understanding in three stages: open-ended referring segmentation, grounded grasp planning and grasp ranking via contact reasoning, all of which can be applied zero-shot via suitable visual prompting mechanisms. We conduct extensive evaluation in cluttered indoor scene datasets to showcase OWG's robustness in grounding from open-ended language, as well as open-world robotic grasping experiments in both simulation and hardware that demonstrate superior performance compared to previous supervised and zero-shot LLM-based methods.
Authors: Yannik Glaser, Justin E. Stopa, Linnea M. Wolniewicz, Ralph Foster, Doug Vandemark, Alexis Mouche, Bertrand Chapron, Peter Sadowski
Abstract: The European Space Agency's Copernicus Sentinel-1 (S-1) mission is a constellation of C-band synthetic aperture radar (SAR) satellites that provide unprecedented monitoring of the world's oceans. S-1's wave mode (WV) captures 20x20 km image patches at 5 m pixel resolution and is unaffected by cloud cover or time-of-day. The mission's open data policy has made SAR data easily accessible for a range of applications, but the need for manual image annotations is a bottleneck that hinders the use of machine learning methods. This study uses nearly 10 million WV-mode images and contrastive self-supervised learning to train a semantic embedding model called WV-Net. In multiple downstream tasks, WV-Net outperforms a comparable model that was pre-trained on natural images (ImageNet) with supervised learning. Experiments show improvements for estimating wave height (0.50 vs 0.60 RMSE using linear probing), estimating near-surface air temperature (0.90 vs 0.97 RMSE), and performing multilabel-classification of geophysical and atmospheric phenomena (0.96 vs 0.95 micro-averaged AUROC). WV-Net embeddings are also superior in an unsupervised image-retrieval task and scale better in data-sparse settings. Together, these results demonstrate that WV-Net embeddings can support geophysical research by providing a convenient foundation model for a variety of data analysis and exploration tasks.
Authors: Jiafei Duan, Wentao Yuan, Wilbert Pumacay, Yi Ru Wang, Kiana Ehsani, Dieter Fox, Ranjay Krishna
Abstract: Large-scale endeavors like RT-1 and widespread community efforts such as Open-X-Embodiment have contributed to growing the scale of robot demonstration data. However, there is still an opportunity to improve the quality, quantity, and diversity of robot demonstration data. Although vision-language models have been shown to automatically generate demonstration data, their utility has been limited to environments with privileged state information, they require hand-designed skills, and are limited to interactions with few object instances. We propose Manipulate-Anything, a scalable automated generation method for real-world robotic manipulation. Unlike prior work, our method can operate in real-world environments without any privileged state information, hand-designed skills, and can manipulate any static object. We evaluate our method using two setups. First, Manipulate-Anything successfully generates trajectories for all 5 real-world and 12 simulation tasks, significantly outperforming existing methods like VoxPoser. Second, Manipulate-Anything's demonstrations can train more robust behavior cloning policies than training with human demonstrations, or from data generated by VoxPoser and Code-As-Policies. We believe \methodLong\ can be the scalable method for both generating data for robotics and solving novel tasks in a zero-shot setting.
Authors: Takeshi Yoshidomi, Shinji Kume, Hiroaki Aizawa, Akira Furui
Abstract: In carotid arteries, plaque can develop as localized elevated lesions. The Jellyfish sign, marked by fluctuating plaque surfaces with blood flow pulsation, is a dynamic characteristic of these plaques that has recently attracted attention. Detecting this sign is vital, as it is often associated with cerebral infarction. This paper proposes an ultrasound video-based classification method for the Jellyfish sign, using deep neural networks. The proposed method first preprocesses carotid ultrasound videos to separate the movement of the vascular wall from plaque movements. These preprocessed videos are then combined with plaque surface information and fed into a deep learning model comprising convolutional and recurrent neural networks, enabling the efficient classification of the Jellyfish sign. The proposed method was verified using ultrasound video images from 200 patients. Ablation studies demonstrated the effectiveness of each component of the proposed method.
Authors: Jiwan Chung, Sungjae Lee, Minseo Kim, Seungju Han, Ashkan Yousefpour, Jack Hessel, Youngjae Yu
Abstract: Visual arguments, often used in advertising or social causes, rely on images to persuade viewers to do or believe something. Understanding these arguments requires selective vision: only specific visual stimuli within an image are relevant to the argument, and relevance can only be understood within the context of a broader argumentative structure. While visual arguments are readily appreciated by human audiences, we ask: are today's AI capable of similar understanding? We collect and release VisArgs, an annotated corpus designed to make explicit the (usually implicit) structures underlying visual arguments. VisArgs includes 1,611 images accompanied by three types of textual annotations: 5,112 visual premises (with region annotations), 5,574 commonsense premises, and reasoning trees connecting them to a broader argument. We propose three tasks over VisArgs to probe machine capacity for visual argument understanding: localization of premises, identification of premises, and deduction of conclusions. Experiments demonstrate that 1) machines cannot fully identify the relevant visual cues. The top-performing model, GPT-4-O, achieved an accuracy of only 78.5%, whereas humans reached 98.0%. All models showed a performance drop, with an average decrease in accuracy of 19.5%, when the comparison set was changed from objects outside the image to irrelevant objects within the image. Furthermore, 2) this limitation is the greatest factor impacting their performance in understanding visual arguments. Most models improved the most when given relevant visual premises as additional inputs, compared to other inputs, for deducing the conclusion of the visual argument.
Authors: Jing Zou, Lanqing Liu, Qi Chen, Shujun Wang, Xiaohan Xing, Jing Qin
Abstract: Multi-contrast MRI acceleration has become prevalent in MR imaging, enabling the reconstruction of high-quality MR images from under-sampled k-space data of the target modality, using guidance from a fully-sampled auxiliary modality. The main crux lies in efficiently and comprehensively integrating complementary information from the auxiliary modality. Existing methods either suffer from quadratic computational complexity or fail to capture long-range correlated features comprehensively. In this work, we propose MMR-Mamba, a novel framework that achieves comprehensive integration of multi-contrast features through Mamba and spatial-frequency information fusion. Firstly, we design the \textit{Target modality-guided Cross Mamba} (TCM) module in the spatial domain, which maximally restores the target modality information by selectively absorbing useful information from the auxiliary modality. Secondly, leveraging global properties of the Fourier domain, we introduce the \textit{Selective Frequency Fusion} (SFF) module to efficiently integrate global information in the frequency domain and recover high-frequency signals for the reconstruction of structure details. Additionally, we present the \textit{Adaptive Spatial-Frequency Fusion} (ASFF) module, which enhances fused features by supplementing less informative features from one domain with corresponding features from the other domain. These innovative strategies ensure efficient feature fusion across spatial and frequency domains, avoiding the introduction of redundant information and facilitating the reconstruction of high-quality target images. Extensive experiments on the BraTS and fastMRI knee datasets demonstrate the superiority of the proposed MMR-Mamba over state-of-the-art MRI reconstruction methods.
Authors: Fanfan Liu, Feng Yan, Liming Zheng, Chengjian Feng, Yiyang Huang, Lin Ma
Abstract: Utilizing Vision-Language Models (VLMs) for robotic manipulation represents a novel paradigm, aiming to enhance the model's ability to generalize to new objects and instructions. However, due to variations in camera specifications and mounting positions, existing methods exhibit significant performance disparities across different robotic platforms. To address this challenge, we propose RoboUniView in this paper, an innovative approach that decouples visual feature extraction from action learning. We first learn a unified view representation from multi-perspective views by pre-training on readily accessible data, and then derive actions from this unified view representation to control robotic manipulation. This unified view representation more accurately mirrors the physical world and is not constrained by the robotic platform's camera parameters. Thanks to this methodology, we achieve state-of-the-art performance on the demanding CALVIN benchmark, enhancing the success rate in the $D \to D$ setting from 88.7% to 96.2%, and in the $ABC \to D$ setting from 82.4% to 94.2%. Moreover, our model exhibits outstanding adaptability and flexibility: it maintains high performance under unseen camera parameters, can utilize multiple datasets with varying camera parameters, and is capable of joint cross-task learning across datasets. Code is provided for re-implementation. https://github.com/liufanfanlff/RoboUniview
Authors: Zi Wang, Fanwen Wang, Chen Qin, Jun Lyu, Ouyang Cheng, Shuo Wang, Yan Li, Mengyao Yu, Haoyu Zhang, Kunyuan Guo, Zhang Shi, Qirong Li, Ziqiang Xu, Yajing Zhang, Hao Li, Sha Hua, Binghua Chen, Longyu Sun, Mengting Sun, Qin Li, Ying-Hua Chu, Wenjia Bai, Jing Qin, Xiahai Zhuang, Claudia Prieto, Alistair Young, Michael Markl, He Wang, Lianming Wu, Guang Yang, Xiaobo Qu, Chengyan Wang
Abstract: Cardiac magnetic resonance imaging (MRI) has emerged as a clinically gold-standard technique for diagnosing cardiac diseases, thanks to its ability to provide diverse information with multiple modalities and anatomical views. Accelerated cardiac MRI is highly expected to achieve time-efficient and patient-friendly imaging, and then advanced image reconstruction approaches are required to recover high-quality, clinically interpretable images from undersampled measurements. However, the lack of publicly available cardiac MRI k-space dataset in terms of both quantity and diversity has severely hindered substantial technological progress, particularly for data-driven artificial intelligence. Here, we provide a standardized, diverse, and high-quality CMRxRecon2024 dataset to facilitate the technical development, fair evaluation, and clinical transfer of cardiac MRI reconstruction approaches, towards promoting the universal frameworks that enable fast and robust reconstructions across different cardiac MRI protocols in clinical practice. To the best of our knowledge, the CMRxRecon2024 dataset is the largest and most diverse publicly available cardiac k-space dataset. It is acquired from 330 healthy volunteers, covering commonly used modalities, anatomical views, and acquisition trajectories in clinical cardiac MRI workflows. Besides, an open platform with tutorials, benchmarks, and data processing tools is provided to facilitate data usage, advanced method development, and fair performance evaluation.
Authors: Daniel Reisenb\"uchler, Lucas Luttner, Nadine S. Schaadt, Friedrich Feuerhake, Dorit Merhof
Abstract: In digital pathology, deep learning (DL) models for tasks such as segmentation or tissue classification are known to suffer from domain shifts due to different staining techniques. Stain adaptation aims to reduce the generalization error between different stains by training a model on source stains that generalizes to target stains. Despite the abundance of target stain data, a key challenge is the lack of annotations. To address this, we propose a joint training between artificially labeled and unlabeled data including all available stained images called Unsupervised Latent Stain Adaption (ULSA). Our method uses stain translation to enrich labeled source images with synthetic target images in order to increase supervised signals. Moreover, we leverage unlabeled target stain images using stain-invariant feature consistency learning. With ULSA we present a semi-supervised strategy for efficient stain adaption without access to annotated target stain data. Remarkably, ULSA is task agnostic in patch-level analysis for whole slide images (WSIs). Through extensive evaluation on external datasets, we demonstrate that ULSA achieves state-of-the-art (SOTA) performance in kidney tissue segmentation and breast cancer classification across a spectrum of staining variations. Our findings suggest that ULSA is an important framework towards stain adaption in digital pathology.
Authors: Lukas Malte Kemeter, Rasmus Hvingelby, Paulina Sierak, Tobias Sch\"on, Bishwajit Gosswam
Abstract: In many manufacturing settings, annotating data for machine learning and computer vision is costly, but synthetic data can be generated at significantly lower cost. Substituting the real-world data with synthetic data is therefore appealing for many machine learning applications that require large amounts of training data. However, relying solely on synthetic data is frequently inadequate for effectively training models that perform well on real-world data, primarily due to domain shifts between the synthetic and real-world data. We discuss approaches for dealing with such a domain shift when detecting defects in X-ray scans of aluminium wheels. Using both simulated and real-world X-ray images, we train an object detection model with different strategies to identify the training approach that generates the best detection results while minimising the demand for annotated real-world training samples. Our preliminary findings suggest that the sim-2-real domain adaptation approach is more cost-efficient than a fully supervised oracle - if the total number of available annotated samples is fixed. Given a certain number of labeled real-world samples, training on a mix of synthetic and unlabeled real-world data achieved comparable or even better detection results at significantly lower cost. We argue that future research into the cost-efficiency of different training strategies is important for a better understanding of how to allocate budget in applied machine learning projects.
Authors: Minghan Li, Heng Li, Zhi-Qi Cheng, Yifei Dong, Yuxuan Zhou, Jun-Yan He, Qi Dai, Teruko Mitamura, Alexander G. Hauptmann
Abstract: Vision-and-Language Navigation (VLN) aims to develop embodied agents that navigate based on human instructions. However, current VLN frameworks often rely on static environments and optimal expert supervision, limiting their real-world applicability. To address this, we introduce Human-Aware Vision-and-Language Navigation (HA-VLN), extending traditional VLN by incorporating dynamic human activities and relaxing key assumptions. We propose the Human-Aware 3D (HA3D) simulator, which combines dynamic human activities with the Matterport3D dataset, and the Human-Aware Room-to-Room (HA-R2R) dataset, extending R2R with human activity descriptions. To tackle HA-VLN challenges, we present the Expert-Supervised Cross-Modal (VLN-CM) and Non-Expert-Supervised Decision Transformer (VLN-DT) agents, utilizing cross-modal fusion and diverse training strategies for effective navigation in dynamic human environments. A comprehensive evaluation, including metrics considering human activities, and systematic analysis of HA-VLN's unique challenges, underscores the need for further research to enhance HA-VLN agents' real-world robustness and adaptability. Ultimately, this work provides benchmarks and insights for future research on embodied AI and Sim2Real transfer, paving the way for more realistic and applicable VLN systems in human-populated environments.
Authors: Shubhankar Singh, Purvi Chaurasia, Yerram Varun, Pranshu Pandya, Vatsal Gupta, Vivek Gupta, Dan Roth
Abstract: Existing benchmarks for visual question answering lack in visual grounding and complexity, particularly in evaluating spatial reasoning skills. We introduce FlowVQA, a novel benchmark aimed at assessing the capabilities of visual question-answering multimodal language models in reasoning with flowcharts as visual contexts. FlowVQA comprises 2,272 carefully generated and human-verified flowchart images from three distinct content sources, along with 22,413 diverse question-answer pairs, to test a spectrum of reasoning tasks, including information localization, decision-making, and logical progression. We conduct a thorough baseline evaluation on a suite of both open-source and proprietary multimodal language models using various strategies, followed by an analysis of directional bias. The results underscore the benchmark's potential as a vital tool for advancing the field of multimodal modeling, providing a focused and challenging environment for enhancing model performance in visual and logical reasoning tasks.
Authors: Gianluca Giacchi, Isidoros Iakovidis, Bastien Milani, Matthias Stuber, Micah Murray, Benedetta Franceschiello
Abstract: Magnetic Resonance Imaging (MRI) is a powerful technique employed for non-invasive in vivo visualization of internal structures. Sparsity is often deployed to accelerate the signal acquisition or overcome the presence of motion artifacts, improving the quality of image reconstruction. Image reconstruction algorithms use TV-regularized LASSO (Total Variation-regularized LASSO) to retrieve the missing information of undersampled signals, by cleaning the data of noise and while optimizing sparsity. A tuning parameter moderates the balance between these two aspects; its choice affecting the quality of the reconstructions. Currently, there is a lack of general deterministic techniques to choose these parameters, which are oftentimes manually selected and thus hinder the reliability of the reconstructions. Here, we present ALMA (Algorithm for Lagrange Multipliers Approximation), an iterative mathematics-inspired technique that computes tuning parameters for generalized LASSO problems during MRI reconstruction. We analyze quantitatively the performance of these parameters for imaging reconstructions via TV-LASSO in an MRI context on phantoms. Although our study concentrates on TV-LASSO, the techniques developed here hold significant promise for a wide array of applications. ALMA is not only adaptable to more generalized LASSO problems but is also robust to accommodate other forms of regularization beyond total variation. Moreover, it extends effectively to handle non-Cartesian sampling trajectories, broadening its utility in complex data reconstruction scenarios. More generally, ALMA provides a powerful tool for numerically solving constrained optimization problems across various disciplines, offering a versatile and impactful solution for advanced computational challenges.
Authors: Yue Fan, Lei Ding, Ching-Chen Kuo, Shan Jiang, Yang Zhao, Xinze Guan, Jie Yang, Yi Zhang, Xin Eric Wang
Abstract: Graphical User Interfaces (GUIs) are central to our interaction with digital devices. Recently, growing efforts have been made to build models for various GUI understanding tasks. However, these efforts largely overlook an important GUI-referring task: screen reading based on user-indicated points, which we name the Screen Point-and-Read (SPR) task. This task is predominantly handled by rigid accessible screen reading tools, in great need of new models driven by advancements in Multimodal Large Language Models (MLLMs). In this paper, we propose a Tree-of-Lens (ToL) agent, utilizing a novel ToL grounding mechanism, to address the SPR task. Based on the input point coordinate and the corresponding GUI screenshot, our ToL agent constructs a Hierarchical Layout Tree. Based on the tree, our ToL agent not only comprehends the content of the indicated area but also articulates the layout and spatial relationships between elements. Such layout information is crucial for accurately interpreting information on the screen, distinguishing our ToL agent from other screen reading tools. We also thoroughly evaluate the ToL agent against other baselines on a newly proposed SPR benchmark, which includes GUIs from mobile, web, and operating systems. Last but not least, we test the ToL agent on mobile GUI navigation tasks, demonstrating its utility in identifying incorrect actions along the path of agent execution trajectories. Code and data: screen-point-and-read.github.io
Authors: Vincent Micheli, Eloi Alonso, Fran\c{c}ois Fleuret
Abstract: Scaling up deep Reinforcement Learning (RL) methods presents a significant challenge. Following developments in generative modelling, model-based RL positions itself as a strong contender. Recent advances in sequence modelling have led to effective transformer-based world models, albeit at the price of heavy computations due to the long sequences of tokens required to accurately simulate environments. In this work, we propose $\Delta$-IRIS, a new agent with a world model architecture composed of a discrete autoencoder that encodes stochastic deltas between time steps and an autoregressive transformer that predicts future deltas by summarizing the current state of the world with continuous tokens. In the Crafter benchmark, $\Delta$-IRIS sets a new state of the art at multiple frame budgets, while being an order of magnitude faster to train than previous attention-based approaches. We release our code and models at https://github.com/vmicheli/delta-iris.
Authors: Kaushalya Sivayogaraj, Sahan T. Guruge, Udari Liyanage, Jeevani Udupihille, Saroj Jayasinghe, Gerard Fernando, Ranga Rodrigo, M. Rukshani Liyanaarachchi
Abstract: 3D reconstruction of the liver for volumetry is important for qualitative analysis and disease diagnosis. Liver volumetry using ultrasound (US) scans, although advantageous due to less acquisition time and safety, is challenging due to the inherent noisiness in US scans, blurry boundaries, and partial liver visibility. We address these challenges by using the segmentation masks of a few incomplete sagittal-plane US scans of the liver in conjunction with a statistical shape model (SSM) built using a set of CT scans of the liver. We compute the shape parameters needed to warp this canonical SSM to fit the US scans through a parametric regression network. The resulting 3D liver reconstruction is accurate and leads to automatic liver volume calculation. We evaluate the accuracy of the estimated liver volumes with respect to CT segmentation volumes using RMSE. Our volume computation is statistically much closer to the volume estimated using CT scans than the volume computed using Childs' method by radiologists: p-value of 0.094 (>0.05) says that there is no significant difference between CT segmentation volumes and ours in contrast to Childs' method. We validate our method using investigations (ablation studies) on the US image resolution, the number of CT scans used for SSM, the number of principal components, and the number of input US scans. To the best of our knowledge, this is the first automatic liver volumetry system using a few incomplete US scans given a set of CT scans of livers for SSM.
Authors: Moayed Haji-Ali, Willi Menapace, Aliaksandr Siarohin, Guha Balakrishnan, Sergey Tulyakov, Vicente Ordonez
Abstract: Generating ambient sounds and effects is a challenging problem due to data scarcity and often insufficient caption quality, making it difficult to employ large-scale generative models for the task. In this work, we tackle the problem by introducing two new models. First, we propose AutoCap, a high-quality and efficient automatic audio captioning model. We show that by leveraging metadata available with the audio modality, we can substantially improve the quality of captions. AutoCap reaches CIDEr score of 83.2, marking a 3.2% improvement from the best available captioning model at four times faster inference speed. We then use AutoCap to caption clips from existing datasets, obtaining 761,000 audio clips with high-quality captions, forming the largest available audio-text dataset. Second, we propose GenAu, a scalable transformer-based audio generation architecture that we scale up to 1.25B parameters and train with our new dataset. When compared to state-of-the-art audio generators, GenAu obtains significant improvements of 15.7% in FAD score, 22.7% in IS, and 13.5% in CLAP score, indicating significantly improved quality of generated audio compared to previous works. This shows that the quality of data is often as important as its quantity. Besides, since AutoCap is fully automatic, new audio samples can be added to the training dataset, unlocking the training of even larger generative models for audio synthesis.
Authors: Chengyin Li, Prashant Khanduri, Yao Qiang, Rafi Ibn Sultan, Indrin Chetty, Dongxiao Zhu
Abstract: Segment Anything Model (SAM) is one of the pioneering prompt-based foundation models for image segmentation and has been rapidly adopted for various medical imaging applications. However, in clinical settings, creating effective prompts is notably challenging and time-consuming, requiring the expertise of domain specialists such as physicians. This requirement significantly diminishes SAM's primary advantage - its interactive capability with end users - in medical applications. Moreover, recent studies have indicated that SAM, originally designed for 2D natural images, performs sub optimally on 3D medical image segmentation tasks. This subpar performance is attributed to the domain gaps between natural and medical images and the disparities in spatial arrangements between 2D and 3D images, particularly in multi-organ segmentation applications. To overcome these challenges, we present a novel technique termed AutoProSAM. This method automates 3D multi-organ CT-based segmentation by leveraging SAM's foundational model capabilities without relying on domain experts for prompts. The approach utilizes parameter-efficient adaptation techniques to adapt SAM for 3D medical imagery and incorporates an effective automatic prompt learning paradigm specific to this domain. By eliminating the need for manual prompts, it enhances SAM's capabilities for 3D medical image segmentation and achieves state-of-the-art (SOTA) performance in CT-based multi-organ segmentation tasks.
Authors: Ruyang Liu, Chen Li, Yixiao Ge, Ying Shan, Thomas H. Li, Ge Li
Abstract: The recent progress in Large Language Models (LLM) has spurred various advancements in image-language conversation agents, while how to build a proficient video-based dialogue system is still under exploration. Considering the extensive scale of LLM and visual backbone, minimal GPU memory is left for facilitating effective temporal modeling, which is crucial for comprehending and providing feedback on videos. To this end, we propose Branching Temporal Adapter (BT-Adapter), a novel method for extending image-language pretrained models into the video domain. Specifically, BT-Adapter serves as a plug-and-use temporal modeling branch alongside the pretrained visual encoder, which is tuned while keeping the backbone frozen. Just pretrained once, BT-Adapter can be seamlessly integrated into all image conversation models using this version of CLIP, enabling video conversations without the need for video instructions. Besides, we develop a unique asymmetric token masking strategy inside the branch with tailor-made training tasks for BT-Adapter, facilitating faster convergence and better results. Thanks to BT-Adapter, we are able to empower existing multimodal dialogue models with strong video understanding capabilities without incurring excessive GPU costs. Without bells and whistles, BT-Adapter achieves (1) state-of-the-art zero-shot results on various video tasks using thousands of fewer GPU hours. (2) better performance than current video chatbots without any video instruction tuning. (3) state-of-the-art results of video chatting using video instruction tuning, outperforming previous SOTAs by a large margin.
Authors: Sergio Izquierdo, Javier Civera
Abstract: The task of Visual Place Recognition (VPR) aims to match a query image against references from an extensive database of images from different places, relying solely on visual cues. State-of-the-art pipelines focus on the aggregation of features extracted from a deep backbone, in order to form a global descriptor for each image. In this context, we introduce SALAD (Sinkhorn Algorithm for Locally Aggregated Descriptors), which reformulates NetVLAD's soft-assignment of local features to clusters as an optimal transport problem. In SALAD, we consider both feature-to-cluster and cluster-to-feature relations and we also introduce a 'dustbin' cluster, designed to selectively discard features deemed non-informative, enhancing the overall descriptor quality. Additionally, we leverage and fine-tune DINOv2 as a backbone, which provides enhanced description power for the local features, and dramatically reduces the required training time. As a result, our single-stage method not only surpasses single-stage baselines in public VPR datasets, but also surpasses two-stage methods that add a re-ranking with significantly higher cost. Code and models are available at https://github.com/serizba/salad.
Authors: Pingyi Chen, Honglin Li, Chenglu Zhu, Sunyi Zheng, Zhongyi Shui, Lin Yang
Abstract: Whole slide images are the foundation of digital pathology for the diagnosis and treatment of carcinomas. Writing pathology reports is laborious and error-prone for inexperienced pathologists. To reduce the workload and improve clinical automation, we investigate how to generate pathology reports given whole slide images. On the data end, we curated the largest WSI-text dataset (PathText). In specific, we collected nearly 10000 high-quality WSI-text pairs for visual-language models by recognizing and cleaning pathology reports which narrate diagnostic slides in TCGA. On the model end, we propose the multiple instance generative model (MI-Gen) which can produce pathology reports for gigapixel WSIs. We benchmark our model on the largest subset of TCGA-PathoText. Experimental results show our model can generate pathology reports which contain multiple clinical clues and achieve competitive performance on certain slide-level tasks. We observe that simple semantic extraction from the pathology reports can achieve the best performance (0.838 of F1 score) on BRCA subtyping surpassing previous state-of-the-art approaches. Our collected dataset and related code are available.
Authors: Alonso Urbano, David W. Romero
Abstract: Group equivariance can overly constrain models if the symmetries in the group differ from those observed in data. While common methods address this by determining the appropriate level of symmetry at the dataset level, they are limited to supervised settings and ignore scenarios in which multiple levels of symmetry co-exist in the same dataset. In this paper, we propose a method able to detect the level of symmetry of each input without the need for labels. Our framework is general enough to accommodate different families of both continuous and discrete symmetry distributions, such as arbitrary unimodal, symmetric distributions and discrete groups. We validate the effectiveness of our approach on synthetic datasets with different per-class levels of symmetries, and demonstrate practical applications such as the detection of out-of-distribution symmetries. Our code is publicly available at https://github.com/aurban0/ssl-sym.
Authors: Dvir Samuel, Barak Meiri, Nir Darshan, Shai Avidan, Gal Chechik, Rami Ben-Ari
Abstract: Diffusion inversion is the problem of taking an image and a text prompt that describes it and finding a noise latent that would generate the image. Most current inversion techniques operate by approximately solving an implicit equation and may converge slowly or yield poor reconstructed images. Here, we formulate the problem as finding the roots of an implicit equation and design a method to solve it efficiently. Our solution is based on Newton-Raphson (NR), a well-known technique in numerical analysis. A naive application of NR may be computationally infeasible and tends to converge to incorrect solutions. We describe an efficient regularized formulation that converges quickly to a solution that provides high-quality reconstructions. We also identify a source of inconsistency stemming from prompt conditioning during the inversion process, which significantly degrades the inversion quality. To address this, we introduce a prompt-aware adjustment of the encoding, effectively correcting this issue. Our solution, Regularized Newton-Raphson Inversion, inverts an image within 0.5 sec for latent consistency models, opening the door for interactive image editing. We further demonstrate improved results in image interpolation and generation of rare objects.
Authors: Liang Peng, Songyue Cai, Zongqian Wu, Huifang Shang, Xiaofeng Zhu, Xiaoxiao Li
Abstract: Prompt learning has demonstrated impressive efficacy in the fine-tuning of multimodal large models to a wide range of downstream tasks. Nonetheless, applying existing prompt learning methods for the diagnosis of neurological disorder still suffers from two issues: (i) existing methods typically treat all patches equally, despite the fact that only a small number of patches in neuroimaging are relevant to the disease, and (ii) they ignore the structural information inherent in the brain connection network which is crucial for understanding and diagnosing neurological disorders. To tackle these issues, we introduce a novel prompt learning model by learning graph prompts during the fine-tuning process of multimodal large models for diagnosing neurological disorders. Specifically, we first leverage GPT-4 to obtain relevant disease concepts and compute semantic similarity between these concepts and all patches. Secondly, we reduce the weight of irrelevant patches according to the semantic similarity between each patch and disease-related concepts. Moreover, we construct a graph among tokens based on these concepts and employ a graph convolutional network layer to extract the structural information of the graph, which is used to prompt the pre-trained multimodal large models for diagnosing neurological disorders. Extensive experiments demonstrate that our method achieves superior performance for neurological disorder diagnosis compared with state-of-the-art methods and validated by clinicians.
Authors: Xun Guo, Mingwu Zheng, Liang Hou, Yuan Gao, Yufan Deng, Pengfei Wan, Di Zhang, Yufan Liu, Weiming Hu, Zhengjun Zha, Haibin Huang, Chongyang Ma
Abstract: Text-guided image-to-video (I2V) generation aims to generate a coherent video that preserves the identity of the input image and semantically aligns with the input prompt. Existing methods typically augment pretrained text-to-video (T2V) models by either concatenating the image with noised video frames channel-wise before being fed into the model or injecting the image embedding produced by pretrained image encoders in cross-attention modules. However, the former approach often necessitates altering the fundamental weights of pretrained T2V models, thus restricting the model's compatibility within the open-source communities and disrupting the model's prior knowledge. Meanwhile, the latter typically fails to preserve the identity of the input image. We present I2V-Adapter to overcome such limitations. I2V-Adapter adeptly propagates the unnoised input image to subsequent noised frames through a cross-frame attention mechanism, maintaining the identity of the input image without any changes to the pretrained T2V model. Notably, I2V-Adapter only introduces a few trainable parameters, significantly alleviating the training cost and also ensures compatibility with existing community-driven personalized models and control tools. Moreover, we propose a novel Frame Similarity Prior to balance the motion amplitude and the stability of generated videos through two adjustable control coefficients. Our experimental results demonstrate that I2V-Adapter is capable of producing high-quality videos. This performance, coupled with its agility and adaptability, represents a substantial advancement in the field of I2V, particularly for personalized and controllable applications.
Authors: Yue Fan, Jing Gu, Kaiwen Zhou, Qianqi Yan, Shan Jiang, Ching-Chen Kuo, Xinze Guan, Xin Eric Wang
Abstract: Multipanel images, commonly seen as web screenshots, posters, etc., pervade our daily lives. These images, characterized by their composition of multiple subfigures in distinct layouts, effectively convey information to people. Toward building advanced multimodal AI applications, such as agents that understand complex scenes and navigate through webpages, the skill of multipanel visual reasoning is essential, and a comprehensive evaluation of models in this regard is important. Therefore, we introduce Multipanel Visual Question Answering (MultipanelVQA), a novel benchmark comprising 6,600 triplets of questions, answers, and multipanel images that specifically challenge models in comprehending multipanel images. Our evaluation shows that questions in the MultipanelVQA benchmark pose significant challenges to the state-of-the-art Multimodal Large Language Models (MLLMs) tested, even though humans can attain approximately 99% accuracy on these questions. Distinctively, the MultipanelVQA benchmark features synthetically generated multipanel images specifically crafted to isolate and assess the impact of various factors, such as the layout, on MLLMs' multipanel image comprehension abilities. As a result, in addition to benchmarking the capabilities of MLLMs in understanding multipanel images, we analyze various factors of the multipanel image that affect MLLMs' performance with synthetic data and offer insights for enhancement. Code and data are released at https://sites.google.com/view/multipanelvqa/home.
Authors: Hamed Amini Amirkolaee, Miaojing Shi, Lianghua He, Mark Mulligan
Abstract: The process of estimating and counting tree density using only a single aerial or satellite image is a difficult task in the fields of photogrammetry and remote sensing. However, it plays a crucial role in the management of forests. The huge variety of trees in varied topography severely hinders tree counting models to perform well. The purpose of this paper is to propose a framework that is learnt from the source domain with sufficient labeled trees and is adapted to the target domain with only a limited number of labeled trees. Our method, termed as AdaTreeFormer, contains one shared encoder with a hierarchical feature extraction scheme to extract robust features from the source and target domains. It also consists of three subnets: two for extracting self-domain attention maps from source and target domains respectively and one for extracting cross-domain attention maps. For the latter, an attention-to-adapt mechanism is introduced to distill relevant information from different domains while generating tree density maps; a hierarchical cross-domain feature alignment scheme is proposed that progressively aligns the features from the source and target domains. We also adopt adversarial learning into the framework to further reduce the gap between source and target domains. Our AdaTreeFormer is evaluated on six designed domain adaptation tasks using three tree counting datasets, \ie Jiangsu, Yosemite, and London. Experimental results show that AdaTreeFormer significantly surpasses the state of the art, \eg in the cross domain from the Yosemite to Jiangsu dataset, it achieves a reduction of 15.9 points in terms of the absolute counting errors and an increase of 10.8\% in the accuracy of the detected trees' locations. The codes and datasets are available at https://github.com/HAAClassic/AdaTreeFormer.
Authors: Liyun Zhu, Lei Wang, Arjun Raj, Tom Gedeon, Chen Chen
Abstract: Video Anomaly Detection (VAD) finds widespread applications in security surveillance, traffic monitoring, industrial monitoring, and healthcare. Despite extensive research efforts, there remains a lack of concise reviews that provide insightful guidance for researchers. Such reviews would serve as quick references to grasp current challenges, research trends, and future directions. In this paper, we present such a review, examining models and datasets from various perspectives. We emphasize the critical relationship between model and dataset, where the quality and diversity of datasets profoundly influence model performance, and dataset development adapts to the evolving needs of emerging approaches. Our review identifies practical issues, including the absence of comprehensive datasets with diverse scenarios. To address this, we introduce a new dataset, Multi-Scenario Anomaly Detection (MSAD), comprising 14 distinct scenarios captured from various camera views. Our dataset has diverse motion patterns and challenging variations, such as different lighting and weather conditions, providing a robust foundation for training superior models. We conduct an in-depth analysis of recent representative models using MSAD and highlight its potential in addressing the challenges of detecting anomalies across diverse and evolving surveillance scenarios. Our dataset is available here.
Authors: Shivang Chopra, Suraj Kothawade, Houda Aynaou, Aman Chadha
Abstract: This paper introduces a novel approach to leverage the generalizability of Diffusion Models for Source-Free Domain Adaptation (DM-SFDA). Our proposed DMSFDA method involves fine-tuning a pre-trained text-to-image diffusion model to generate source domain images using features from the target images to guide the diffusion process. Specifically, the pre-trained diffusion model is fine-tuned to generate source samples that minimize entropy and maximize confidence for the pre-trained source model. We then use a diffusion model-based image mixup strategy to bridge the domain gap between the source and target domains. We validate our approach through comprehensive experiments across a range of datasets, including Office-31, Office-Home, and VisDA. The results demonstrate significant improvements in SFDA performance, highlighting the potential of diffusion models in generating contextually relevant, domain-specific images.
Authors: Yaoyao Zhu, Xiuding Cai, Xueyao Wang, Xiaoqing Chen, Yu Yao, Zhongliang Fu
Abstract: Data augmentation is a crucial regularization technique for deep neural networks, particularly in medical image classification. Mainstream data augmentation (DA) methods are usually applied at the image level. Due to the specificity and diversity of medical imaging, expertise is often required to design effective DA strategies, and improper augmentation operations can degrade model performance. Although automatic augmentation methods exist, they are computationally intensive. Semantic data augmentation can implemented by translating features in feature space. However, over-translation may violate the image label. To address these issues, we propose \emph{Bayesian Random Semantic Data Augmentation} (BSDA), a computationally efficient and handcraft-free feature-level DA method. BSDA uses variational Bayesian to estimate the distribution of the augmentable magnitudes, and then a sample from this distribution is added to the original features to perform semantic data augmentation. We performed experiments on nine 2D and five 3D medical image datasets. Experimental results show that BSDA outperforms current DA methods. Additionally, BSDA can be easily assembled into CNNs or Transformers as a plug-and-play module, improving the network's performance. The code is available online at \url{https://github.com/YaoyaoZhu19/BSDA}.
Authors: Yilin Leng (for the Alzheimer's Disease Neuroimaging Initiative), Wenju Cui (for the Alzheimer's Disease Neuroimaging Initiative), Bai Chen (for the Alzheimer's Disease Neuroimaging Initiative), Xi Jiang (for the Alzheimer's Disease Neuroimaging Initiative), Shuangqing Chen (for the Alzheimer's Disease Neuroimaging Initiative), Jian Zheng (for the Alzheimer's Disease Neuroimaging Initiative)
Abstract: Prediction the conversion to early-stage dementia is critical for mitigating its progression but remains challenging due to subtle cognitive impairments and structural brain changes. Traditional T1-weighted magnetic resonance imaging (T1-MRI) research focus on identifying brain atrophy regions but often fails to address the intricate connectivity between them. This limitation underscores the necessity of focuing on inter-regional connectivity for a comprehensive understand of the brain's complex network. Moreover, there is a pressing demand for methods that adaptively preserve and extract critical information, particularly specialized subgraph mining techniques for brain networks. These are essential for developing high-quality feature representations that reveal critical spatial impacts of structural brain changes and its topology. In this paper, we propose Brain-SubGNN, a novel graph representation network to mine and enhance critical subgraphs based on T1-MRI. This network provides a subgraph-level interpretation, enhancing interpretability and insights for graph analysis. The process begins by extracting node features and a correlation matrix between nodes to construct a task-oriented brain network. Brain-SubGNN then adaptively identifies and enhances critical subgraphs, capturing both loop and neighbor subgraphs. This method reflects the loop topology and local changes, indicative of long-range connections, and maintains local and global brain attributes. Extensive experiments validate the effectiveness and advantages of Brain-SubGNN, demonstrating its potential as a powerful tool for understanding and diagnosing early-stage dementia. Source code is available at https://github.com/Leng-10/Brain-SubGNN.
Authors: Yoni Kasten, Wuyue Lu, Haggai Maron
Abstract: This paper addresses the long-standing challenge of reconstructing 3D structures from videos with dynamic content. Current approaches to this problem were not designed to operate on casual videos recorded by standard cameras or require a long optimization time. Aiming to significantly improve the efficiency of previous approaches, we present TracksTo4D, a learning-based approach that enables inferring 3D structure and camera positions from dynamic content originating from casual videos using a single efficient feed-forward pass. To achieve this, we propose operating directly over 2D point tracks as input and designing an architecture tailored for processing 2D point tracks. Our proposed architecture is designed with two key principles in mind: (1) it takes into account the inherent symmetries present in the input point tracks data, and (2) it assumes that the movement patterns can be effectively represented using a low-rank approximation. TracksTo4D is trained in an unsupervised way on a dataset of casual videos utilizing only the 2D point tracks extracted from the videos, without any 3D supervision. Our experiments show that TracksTo4D can reconstruct a temporal point cloud and camera positions of the underlying video with accuracy comparable to state-of-the-art methods, while drastically reducing runtime by up to 95\%. We further show that TracksTo4D generalizes well to unseen videos of unseen semantic categories at inference time.
Authors: Kai Chen, Yanze Li, Wenhua Zhang, Yanxin Liu, Pengxiang Li, Ruiyuan Gao, Lanqing Hong, Meng Tian, Xinhai Zhao, Zhenguo Li, Dit-Yan Yeung, Huchuan Lu, Xu Jia
Abstract: Large Vision-Language Models (LVLMs) have received widespread attention in advancing the interpretable self-driving. Existing evaluations of LVLMs primarily focus on the multi-faceted capabilities in natural circumstances, lacking automated and quantifiable assessment for self-driving, let alone the severe road corner cases. In this paper, we propose CODA-LM, the very first benchmark for the automatic evaluation of LVLMs for self-driving corner cases. We adopt a hierarchical data structure to prompt powerful LVLMs to analyze complex driving scenes and generate high-quality pre-annotation for human annotators, and for LVLM evaluation, we show that using the text-only large language models (LLMs) as judges reveals even better alignment with human preferences than the LVLM judges. Moreover, with CODA-LM, we build CODA-VLM, a new driving LVLM surpassing all the open-sourced counterparts on CODA-LM. Our CODA-VLM performs comparably with GPT-4V, even surpassing GPT-4V by +21.42% on the regional perception task. We hope CODA-LM can become the catalyst to promote interpretable self-driving empowered by LVLMs.
Authors: Zhicheng Ding, Panfeng Li, Qikai Yang, Siyang Li, Qingtian Gong
Abstract: This paper presents a novel contribution to the field of regional style transfer. Existing methods often suffer from the drawback of applying style homogeneously across the entire image, leading to stylistic inconsistencies or foreground object twisted when applied to image with foreground elements such as person figures. To address this limitation, we propose a new approach that leverages a segmentation network to precisely isolate foreground objects within the input image. Subsequently, style transfer is applied exclusively to the background region. The isolated foreground objects are then carefully reintegrated into the style-transferred background. To enhance the visual coherence between foreground and background, a color transfer step is employed on the foreground elements prior to their rein-corporation. Finally, we utilize feathering techniques to achieve a seamless amalgamation of foreground and background, resulting in a visually unified and aesthetically pleasing final composition. Extensive evaluations demonstrate that our proposed approach yields significantly more natural stylistic transformations compared to conventional methods.
Authors: Jayanth Shenoy, Xingjian Davis Zhang, Shlok Mehrotra, Bill Tao, Rem Yang, Han Zhao, Deepak Vasisht
Abstract: Satellite image time series (SITS) segmentation is crucial for many applications like environmental monitoring, land cover mapping and agricultural crop type classification. However, training models for SITS segmentation remains a challenging task due to the lack of abundant training data, which requires fine grained annotation. We propose S4 a new self-supervised pre-training approach that significantly reduces the requirement for labeled training data by utilizing two new insights: (a) Satellites capture images in different parts of the spectrum such as radio frequencies, and visible frequencies. (b) Satellite imagery is geo-registered allowing for fine-grained spatial alignment. We use these insights to formulate pre-training tasks in S4. We also curate m2s2-SITS, a large-scale dataset of unlabeled, spatially-aligned, multi-modal and geographic specific SITS that serves as representative pre-training data for S4. Finally, we evaluate S4 on multiple SITS segmentation datasets and demonstrate its efficacy against competing baselines while using limited labeled data.
Authors: Manish Dhakal, Rabin Adhikari, Safal Thapaliya, Bishesh Khanal
Abstract: Foundation Vision-Language Models (VLMs) trained using large-scale open-domain images and text pairs have recently been adapted to develop Vision-Language Segmentation Models (VLSMs) that allow providing text prompts during inference to guide image segmentation. If robust and powerful VLSMs can be built for medical images, it could aid medical professionals in many clinical tasks where they must spend substantial time delineating the target structure of interest. VLSMs for medical images resort to fine-tuning base VLM or VLSM pretrained on open-domain natural image datasets due to fewer annotated medical image datasets; this fine-tuning is resource-consuming and expensive as it usually requires updating all or a significant fraction of the pretrained parameters. Recently, lightweight blocks called adapters have been proposed in VLMs that keep the pretrained model frozen and only train adapters during fine-tuning, substantially reducing the computing resources required. We introduce a novel adapter, VLSM-Adapter, that can fine-tune pretrained vision-language segmentation models using transformer encoders. Our experiments in widely used CLIP-based segmentation models show that with only 3 million trainable parameters, the VLSM-Adapter outperforms state-of-the-art and is comparable to the upper bound end-to-end fine-tuning. The source code is available at: https://github.com/naamiinepal/vlsm-adapter.
Authors: Shengqiong Wu, Hao Fei, Xiangtai Li, Jiayi Ji, Hanwang Zhang, Tat-Seng Chua, Shuicheng Yan
Abstract: Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in processing vision-language tasks. One of the crux of MLLMs lies in vision tokenization, which involves efficiently transforming input visual signals into feature representations that are most beneficial for LLMs. However, existing vision tokenizers, essential for semantic alignment between vision and language, remain problematic. Existing methods aggressively fragment visual input, corrupting the visual semantic integrity. To address this, this paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok), which groups visual features into semantic units via a dynamic clustering algorithm, flexibly determining the number of tokens based on image complexity. The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features. The proposed MLLM (Setokim) equipped with SeTok significantly demonstrates superior performance across various tasks, as evidenced by our experimental results. The project page is at https://chocowu.github.io/SeTok-web/.
Authors: Hongjia Chen, Xin Xu, Fangling Pu
Abstract: Change detection (CD) in remote sensing imagery is a crucial task with applications in environmental monitoring, urban development, and disaster management. CD involves utilizing bi-temporal images to identify changes over time. The bi-temporal spatial relationships between features at the same location at different times play a key role in this process. However, existing change detection networks often do not fully leverage these spatial relationships during bi-temporal feature extraction and fusion. In this work, we propose SRC-Net: a bi-temporal spatial relationship concerned network for CD. The proposed SRC-Net includes a Perception and Interaction Module that incorporates spatial relationships and establishes a cross-branch perception mechanism to enhance the precision and robustness of feature extraction. Additionally, a Patch-Mode joint Feature Fusion Module is introduced to address information loss in current methods. It considers different change modes and concerns about spatial relationships, resulting in more expressive fusion features. Furthermore, we construct a novel network using these two relationship concerned modules and conducted experiments on the LEVIR-CD and WHU Building datasets. The experimental results demonstrate that our network outperforms state-of-the-art (SOTA) methods while maintaining a modest parameter count. We believe our approach sets a new paradigm for change detection and will inspire further advancements in the field. The code and models are publicly available at https://github.com/Chnja/SRCNet.
Authors: Zigeng Chen, Xinyin Ma, Gongfan Fang, Zhenxiong Tan, Xinchao Wang
Abstract: Diffusion models have garnered significant interest from the community for their great generative ability across various applications. However, their typical multi-step sequential-denoising nature gives rise to high cumulative latency, thereby precluding the possibilities of parallel computation. To address this, we introduce AsyncDiff, a universal and plug-and-play acceleration scheme that enables model parallelism across multiple devices. Our approach divides the cumbersome noise prediction model into multiple components, assigning each to a different device. To break the dependency chain between these components, it transforms the conventional sequential denoising into an asynchronous process by exploiting the high similarity between hidden states in consecutive diffusion steps. Consequently, each component is facilitated to compute in parallel on separate devices. The proposed strategy significantly reduces inference latency while minimally impacting the generative quality. Specifically, for the Stable Diffusion v2.1, AsyncDiff achieves a 2.7x speedup with negligible degradation and a 4.0x speedup with only a slight reduction of 0.38 in CLIP Score, on four NVIDIA A5000 GPUs. Our experiments also demonstrate that AsyncDiff can be readily applied to video diffusion models with encouraging performances. The code is available at https://github.com/czg1225/AsyncDiff.
Authors: Wenxiao Cai, Yaroslav Ponomarenko, Jianhao Yuan, Xiaoqi Li, Wankou Yang, Hao Dong, Bo Zhao
Abstract: Vision Language Models (VLMs) have achieved impressive performance in 2D image understanding, however they are still struggling with spatial understanding which is the foundation of Embodied AI. In this paper, we propose SpatialBot for better spatial understanding by feeding both RGB and depth images. Additionally, we have constructed the SpatialQA dataset, which involves multi-level depth-related questions to train VLMs for depth understanding. Finally, we present SpatialBench to comprehensively evaluate VLMs' capabilities in spatial understanding at different levels. Extensive experiments on our spatial-understanding benchmark, general VLM benchmarks and Embodied AI tasks, demonstrate the remarkable improvements of SpatialBot trained on SpatialQA. The model, code and data are available at https://github.com/BAAI-DCAI/SpatialBot.
Authors: Guido Di Federico, Louis J. Durlofsky
Abstract: Geological parameterization entails the representation of a geomodel using a small set of latent variables and a mapping from these variables to grid-block properties such as porosity and permeability. Parameterization is useful for data assimilation (history matching), as it maintains geological realism while reducing the number of variables to be determined. Diffusion models are a new class of generative deep-learning procedures that have been shown to outperform previous methods, such as generative adversarial networks, for image generation tasks. Diffusion models are trained to "denoise", which enables them to generate new geological realizations from input fields characterized by random noise. Latent diffusion models, which are the specific variant considered in this study, provide dimension reduction through use of a low-dimensional latent variable. The model developed in this work includes a variational autoencoder for dimension reduction and a U-net for the denoising process. Our application involves conditional 2D three-facies (channel-levee-mud) systems. The latent diffusion model is shown to provide realizations that are visually consistent with samples from geomodeling software. Quantitative metrics involving spatial and flow-response statistics are evaluated, and general agreement between the diffusion-generated models and reference realizations is observed. Stability tests are performed to assess the smoothness of the parameterization method. The latent diffusion model is then used for ensemble-based data assimilation. Two synthetic "true" models are considered. Significant uncertainty reduction, posterior P$_{10}$-P$_{90}$ forecasts that generally bracket observed data, and consistent posterior geomodels, are achieved in both cases.
Authors: Yingying Fang, Shuang Wu, Zihao Jin, Caiwen Xu, Shiyi Wang, Simon Walsh, Guang Yang
Abstract: In the field of medical imaging, particularly in tasks related to early disease detection and prognosis, understanding the reasoning behind AI model predictions is imperative for assessing their reliability. Conventional explanation methods encounter challenges in identifying decisive features in medical image classifications, especially when discriminative features are subtle or not immediately evident. To address this limitation, we propose an agent model capable of generating counterfactual images that prompt different decisions when plugged into a black box model. By employing this agent model, we can uncover influential image patterns that impact the black model's final predictions. Through our methodology, we efficiently identify features that influence decisions of the deep black box. We validated our approach in the rigorous domain of medical prognosis tasks, showcasing its efficacy and potential to enhance the reliability of deep learning models in medical image classification compared to existing interpretation methods. The code will be publicly available at https://github.com/ayanglab/DiffExplainer.
Authors: Zhiyu Tan, Xiaomeng Yang, Luozheng Qin, Mengping Yang, Cheng Zhang, Hao Li
Abstract: The recent advancements in text-to-image generative models have been remarkable. Yet, the field suffers from a lack of evaluation metrics that accurately reflect the performance of these models, particularly lacking fine-grained metrics that can guide the optimization of the models. In this paper, we propose EvalAlign, a metric characterized by its accuracy, stability, and fine granularity. Our approach leverages the capabilities of Multimodal Large Language Models (MLLMs) pre-trained on extensive datasets. We develop evaluation protocols that focus on two key dimensions: image faithfulness and text-image alignment. Each protocol comprises a set of detailed, fine-grained instructions linked to specific scoring options, enabling precise manual scoring of the generated images. We Supervised Fine-Tune (SFT) the MLLM to align closely with human evaluative judgments, resulting in a robust evaluation model. Our comprehensive tests across 24 text-to-image generation models demonstrate that EvalAlign not only provides superior metric stability but also aligns more closely with human preferences than existing metrics, confirming its effectiveness and utility in model assessment.
Authors: Yang Liu, Kowshik Thopalli, Jayaraman Thiagarajan
Abstract: Scaling up neural networks has been a key recipe to the success of large language and vision models. However, in practice, up-scaled models can be disproportionately costly in terms of computations, providing only marginal improvements in performance; for example, EfficientViT-L3-384 achieves <2% improvement on ImageNet-1K accuracy over the base L1-224 model, while requiring $14\times$ more multiply-accumulate operations (MACs). In this paper, we investigate scaling properties of popular families of neural networks for image classification, and find that scaled-up models mostly help with "difficult" samples. Decomposing the samples by difficulty, we develop a simple model-agnostic two-pass Little-Big algorithm that first uses a light-weight "little" model to make predictions of all samples, and only passes the difficult ones for the "big" model to solve. Good little companion achieve drastic MACs reduction for a wide variety of model families and scales. Without loss of accuracy or modification of existing models, our Little-Big models achieve MACs reductions of 76% for EfficientViT-L3-384, 81% for EfficientNet-B7-600, 71% for DeiT3-L-384 on ImageNet-1K. Little-Big also speeds up the InternImage-G-512 model by 62% while achieving 90% ImageNet-1K top-1 accuracy, serving both as a strong baseline and as a simple practical method for large model compression.
Authors: Zhuolin He, Xinrun Li, Heng Gao, Jiachen Tang, Shoumeng Qiu, Wenfu Wang, Lvjian Lu, Xuchong Qiu, Xiangyang Xue, Jian Pu
Abstract: Traditional camera 3D object detectors are typically trained to recognize a predefined set of known object classes. In real-world scenarios, these detectors may encounter unknown objects outside the training categories and fail to identify them correctly. To address this gap, we present OS-Det3D (Open-set Camera 3D Object Detection), a two-stage training framework enhancing the ability of camera 3D detectors to identify both known and unknown objects. The framework involves our proposed 3D Object Discovery Network (ODN3D), which is specifically trained using geometric cues such as the location and scale of 3D boxes to discover general 3D objects. ODN3D is trained in a class-agnostic manner, and the provided 3D object region proposals inherently come with data noise. To boost accuracy in identifying unknown objects, we introduce a Joint Objectness Selection (JOS) module. JOS selects the pseudo ground truth for unknown objects from the 3D object region proposals of ODN3D by combining the ODN3D objectness and camera feature attention objectness. Experiments on the nuScenes and KITTI datasets demonstrate the effectiveness of our framework in enabling camera 3D detectors to successfully identify unknown objects while also improving their performance on known objects.
Authors: Filipe Gama, Matej Misar, Lukas Navara, Sergiu T. Popescu, Matej Hoffmann
Abstract: Automatic markerless estimation of infant posture and motion from ordinary videos carries great potential for movement studies "in the wild", facilitating understanding of motor development and massively increasing the chances of early diagnosis of disorders. There is rapid development of human pose estimation methods in computer vision thanks to advances in deep learning and machine learning. However, these methods are trained on datasets featuring adults in different contexts. This work tests and compares seven popular methods (AlphaPose, DeepLabCut/DeeperCut, Detectron2, HRNet, MediaPipe/BlazePose, OpenPose, and ViTPose) on videos of infants in supine position. Surprisingly, all methods except DeepLabCut and MediaPipe have competitive performance without additional finetuning, with ViTPose performing best. Next to standard performance metrics (object keypoint similarity, average precision and recall), we introduce errors expressed in the neck-mid-hip ratio and additionally study missed and redundant detections and the reliability of the internal confidence ratings of the different methods, which are relevant for downstream tasks. Among the networks with competitive performance, only AlphaPose could run close to real time (27 fps) on our machine. We provide documented Docker containers or instructions for all the methods we used, our analysis scripts, and processed data at https://hub.docker.com/u/humanoidsctu and https://osf.io/x465b/.
URLs: https://hub.docker.com/u/humanoidsctu, https://osf.io/x465b/.
Authors: Xiangyu Zhao, Xiangtai Li, Haodong Duan, Haian Huang, Yining Li, Kai Chen, Hua Yang
Abstract: Multi-modal large language models (MLLMs) have made significant strides in various visual understanding tasks. However, the majority of these models are constrained to process low-resolution images, which limits their effectiveness in perception tasks that necessitate detailed visual information. In our study, we present MG-LLaVA, an innovative MLLM that enhances the model's visual processing capabilities by incorporating a multi-granularity vision flow, which includes low-resolution, high-resolution, and object-centric features. We propose the integration of an additional high-resolution visual encoder to capture fine-grained details, which are then fused with base visual features through a Conv-Gate fusion network. To further refine the model's object recognition abilities, we incorporate object-level features derived from bounding boxes identified by offline detectors. Being trained solely on publicly available multimodal data through instruction tuning, MG-LLaVA demonstrates exceptional perception skills. We instantiate MG-LLaVA with a wide variety of language encoders, ranging from 3.8B to 34B, to evaluate the model's performance comprehensively. Extensive evaluations across multiple benchmarks demonstrate that MG-LLaVA outperforms existing MLLMs of comparable parameter sizes, showcasing its remarkable efficacy. The code will be available at https://github.com/PhoenixZ810/MG-LLaVA.
Authors: Jialun Pei, Ruize Cui, Yaoqian Li, Weixin Si, Jing Qin, Pheng-Ann Heng
Abstract: Laparoscopic liver surgery poses a complex intraoperative dynamic environment for surgeons, where remains a significant challenge to distinguish critical or even hidden structures inside the liver. Liver anatomical landmarks, e.g., ridge and ligament, serve as important markers for 2D-3D alignment, which can significantly enhance the spatial perception of surgeons for precise surgery. To facilitate the detection of laparoscopic liver landmarks, we collect a novel dataset called L3D, which comprises 1,152 frames with elaborated landmark annotations from surgical videos of 39 patients across two medical sites. For benchmarking purposes, 12 mainstream detection methods are selected and comprehensively evaluated on L3D. Further, we propose a depth-driven geometric prompt learning network, namely D2GPLand. Specifically, we design a Depth-aware Prompt Embedding (DPE) module that is guided by self-supervised prompts and generates semantically relevant geometric information with the benefit of global depth cues extracted from SAM-based features. Additionally, a Semantic-specific Geometric Augmentation (SGA) scheme is introduced to efficiently merge RGB-D spatial and geometric information through reverse anatomic perception. The experimental results indicate that D2GPLand obtains state-of-the-art performance on L3D, with 63.52% DICE and 48.68% IoU scores. Together with 2D-3D fusion technology, our method can directly provide the surgeon with intuitive guidance information in laparoscopic scenarios.
Authors: Baoqi Pei, Guo Chen, Jilan Xu, Yuping He, Yicheng Liu, Kanghua Pan, Yifei Huang, Yali Wang, Tong Lu, Limin Wang, Yu Qiao
Abstract: In this report, we present our solutions to the EgoVis Challenges in CVPR 2024, including five tracks in the Ego4D challenge and three tracks in the EPIC-Kitchens challenge. Building upon the video-language two-tower model and leveraging our meticulously organized egocentric video data, we introduce a novel foundation model called EgoVideo. This model is specifically designed to cater to the unique characteristics of egocentric videos and provides strong support for our competition submissions. In the Ego4D challenges, we tackle various tasks including Natural Language Queries, Step Grounding, Moment Queries, Short-term Object Interaction Anticipation, and Long-term Action Anticipation. In addition, we also participate in the EPIC-Kitchens challenge, where we engage in the Action Recognition, Multiple Instance Retrieval, and Domain Adaptation for Action Recognition tracks. By adapting EgoVideo to these diverse tasks, we showcase its versatility and effectiveness in different egocentric video analysis scenarios, demonstrating the powerful representation ability of EgoVideo as an egocentric foundation model. Our codebase and pretrained models are publicly available at https://github.com/OpenGVLab/EgoVideo.
Authors: Hao Li, Ming Yuan, Yan Zhang, Chenming Wu, Chen Zhao, Chunyu Song, Haocheng Feng, Errui Ding, Dingwen Zhang, Jingdong Wang
Abstract: Thoroughly testing autonomy systems is crucial in the pursuit of safe autonomous driving vehicles. It necessitates creating safety-critical scenarios that go beyond what can be safely collected from real-world data, as many of these scenarios occur infrequently on public roads. However, the evaluation of most existing NVS methods relies on sporadic sampling of image frames from the training data, comparing the rendered images with ground truth images using metrics. Unfortunately, this evaluation protocol falls short of meeting the actual requirements in closed-loop simulations. Specifically, the true application demands the capability to render novel views that extend beyond the original trajectory (such as cross-lane views), which are challenging to capture in the real world. To address this, this paper presents a novel driving view synthesis dataset and benchmark specifically designed for autonomous driving simulations. This dataset is unique as it includes testing images captured by deviating from the training trajectory by 1-4 meters. It comprises six sequences encompassing various time and weather conditions. Each sequence contains 450 training images, 150 testing images, and their corresponding camera poses and intrinsic parameters. Leveraging this novel dataset, we establish the first realistic benchmark for evaluating existing NVS approaches under front-only and multi-camera settings. The experimental findings underscore the significant gap that exists in current approaches, revealing their inadequate ability to fulfill the demanding prerequisites of cross-lane or closed-loop simulation. Our dataset is released publicly at the project page: https://3d-aigc.github.io/XLD/.
Authors: Tianyu Lin, Zhiguang Chen, Zhonghao Yan, Weijiang Yu, Fudan Zheng
Abstract: Diffusion models have demonstrated their effectiveness across various generative tasks. However, when applied to medical image segmentation, these models encounter several challenges, including significant resource and time requirements. They also necessitate a multi-step reverse process and multiple samples to produce reliable predictions. To address these challenges, we introduce the first latent diffusion segmentation model, named SDSeg, built upon stable diffusion (SD). SDSeg incorporates a straightforward latent estimation strategy to facilitate a single-step reverse process and utilizes latent fusion concatenation to remove the necessity for multiple samples. Extensive experiments indicate that SDSeg surpasses existing state-of-the-art methods on five benchmark datasets featuring diverse imaging modalities. Remarkably, SDSeg is capable of generating stable predictions with a solitary reverse step and sample, epitomizing the model's stability as implied by its name. The code is available at https://github.com/lin-tianyu/Stable-Diffusion-Seg
Authors: Chengkang Shen, Hao Zhu, You Zhou, Yu Liu, Si Yi, Lili Dong, Weipeng Zhao, David J. Brady, Xun Cao, Zhan Ma, Yi Lin
Abstract: Myocardial motion tracking stands as an essential clinical tool in the prevention and detection of cardiovascular diseases (CVDs), the foremost cause of death globally. However, current techniques suffer from incomplete and inaccurate motion estimation of the myocardium in both spatial and temporal dimensions, hindering the early identification of myocardial dysfunction. To address these challenges, this paper introduces the Neural Cardiac Motion Field (NeuralCMF). NeuralCMF leverages implicit neural representation (INR) to model the 3D structure and the comprehensive 6D forward/backward motion of the heart. This method surpasses pixel-wise limitations by offering the capability to continuously query the precise shape and motion of the myocardium at any specific point throughout the cardiac cycle, enhancing the detailed analysis of cardiac dynamics beyond traditional speckle tracking. Notably, NeuralCMF operates without the need for paired datasets, and its optimization is self-supervised through the physics knowledge priors in both space and time dimensions, ensuring compatibility with both 2D and 3D echocardiogram video inputs. Experimental validations across three representative datasets support the robustness and innovative nature of the NeuralCMF, marking significant advantages over existing state-of-the-art methods in cardiac imaging and motion tracking.
Authors: Eshant English, Matthias Kirchler, Christoph Lippert
Abstract: Normalising flows are generative models that transform a complex density into a simpler density through the use of bijective transformations enabling both density estimation and data generation from a single model. %However, the requirement for bijectivity imposes the use of specialised architectures. In the context of image modelling, the predominant choice has been the Glow-based architecture, whereas alternative architectures remain largely unexplored in the research community. In this work, we propose a novel architecture called MixerFlow, based on the MLP-Mixer architecture, further unifying the generative and discriminative modelling architectures. MixerFlow offers an efficient mechanism for weight sharing for flow-based models. Our results demonstrate comparative or superior density estimation on image datasets and good scaling as the image resolution increases, making MixerFlow a simple yet powerful alternative to the Glow-based architectures. We also show that MixerFlow provides more informative embeddings than Glow-based architectures and can integrate many structured transformations such as splines or Kolmogorov-Arnold Networks.
Authors: Cathrin Elich, Lukas Kirchdorfer, Jan M. K\"ohler, Lukas Schott
Abstract: While multi-task learning (MTL) has gained significant attention in recent years, its underlying mechanisms remain poorly understood. Recent methods did not yield consistent performance improvements over single task learning (STL) baselines, underscoring the importance of gaining more profound insights about challenges specific to MTL. In our study, we investigate paradigms in MTL in the context of STL: First, the impact of the choice of optimizer has only been mildly investigated in MTL. We show the pivotal role of common STL tools such as the Adam optimizer in MTL empirically in various experiments. To further investigate Adam's effectiveness, we theoretical derive a partial loss-scale invariance under mild assumptions. Second, the notion of gradient conflicts has often been phrased as a specific problem in MTL. We delve into the role of gradient conflicts in MTL and compare it to STL. For angular gradient alignment we find no evidence that this is a unique problem in MTL. We emphasize differences in gradient magnitude as the main distinguishing factor. Overall, we find surprising similarities between STL and MTL suggesting to consider methods from both fields in a broader context.
Authors: Yunxiang Li, Hua-Chieh Shao, Xiaoxue Qian, You Zhang
Abstract: Diffusion models have demonstrated significant potential in producing high-quality images in medical image translation to aid disease diagnosis, localization, and treatment. Nevertheless, current diffusion models have limited success in achieving faithful image translations that can accurately preserve the anatomical structures of medical images, especially for unpaired datasets. The preservation of structural and anatomical details is essential to reliable medical diagnosis and treatment planning, as structural mismatches can lead to disease misidentification and treatment errors. In this study, we introduce the Frequency Decoupled Diffusion Model (FDDM) for MR-to-CT conversion. FDDM first obtains the anatomical information of the CT image from the MR image through an initial conversion module. This anatomical information then guides a subsequent diffusion model to generate high-quality CT images. Our diffusion model uses a dual-path reverse diffusion process for low-frequency and high-frequency information, achieving a better balance between image quality and anatomical accuracy. We extensively evaluated FDDM using public datasets for brain MR-to-CT and pelvis MR-to-CT translations, demonstrating its superior performance to other GAN-based, VAE-based, and diffusion-based models. The evaluation metrics included Frechet Inception Distance (FID), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM). FDDM achieved the best scores on all metrics for both datasets, particularly excelling in FID, with scores of 25.9 for brain data and 29.2 for pelvis data, significantly outperforming other methods. These results demonstrate that FDDM can generate high-quality target domain images while maintaining the accuracy of translated anatomical structures.
Authors: Shinnosuke Yamamoto, Isso Saito, Eichi Takaya, Ayaka Harigai, Tomomi Sato, Tomoya Kobayashi, Kei Takase, Takuya Ueda
Abstract: [Purpose] To develop a fully automated semantic placenta segmentation model that integrates the U-Net and SegNeXt architectures through ensemble learning. [Methods] A total of 218 pregnant women with suspected placental anomalies who underwent magnetic resonance imaging (MRI) were enrolled, yielding 1090 annotated images for developing a deep learning model for placental segmentation. The images were standardized and divided into training and test sets. The performance of PlaNet-S, which integrates U-Net and SegNeXt within an ensemble framework, was assessed using Intersection over Union (IoU) and counting connected components (CCC) against the U-Net model. [Results] PlaNet-S had significantly higher IoU (0.73 +/- 0.13) than that of U-Net (0.78 +/- 0.010) (p<0.01). The CCC for PlaNet-S was significantly higher than that for U-Net (p<0.01), matching the ground truth in 86.0\% and 56.7\% of the cases, respectively. [Conclusion]PlaNet-S performed better than the traditional U-Net in placental segmentation tasks. This model addresses the challenges of time-consuming physician-assisted manual segmentation and offers the potential for diverse applications in placental imaging analyses.
Authors: Cuong V. Nguyen, Cuong D. Do
Abstract: The adoption of deep learning in ECG diagnosis is often hindered by the scarcity of large, well-labeled datasets in real-world scenarios, leading to the use of transfer learning to leverage features learned from larger datasets. Yet the prevailing assumption that transfer learning consistently outperforms training from scratch has never been systematically validated. In this study, we conduct the first extensive empirical study on the effectiveness of transfer learning in multi-label ECG classification, by investigating comparing the fine-tuning performance with that of training from scratch, covering a variety of ECG datasets and deep neural networks. We confirm that fine-tuning is the preferable choice for small downstream datasets; however, when the dataset is sufficiently large, training from scratch can achieve comparable performance, albeit requiring a longer training time to catch up. Furthermore, we find that transfer learning exhibits better compatibility with convolutional neural networks than with recurrent neural networks, which are the two most prevalent architectures for time-series ECG applications. Our results underscore the importance of transfer learning in ECG diagnosis, yet depending on the amount of available data, researchers may opt not to use it, considering the non-negligible cost associated with pre-training.
Authors: Cary Coglianese, Colton R. Crum
Abstract: Fervent calls for more robust governance of the harms associated with artificial intelligence (AI) are leading to the adoption around the world of what regulatory scholars have called a management-based approach to regulation. Recent initiatives in the United States and Europe, as well as the adoption of major self-regulatory standards by the International Organization for Standardization, share in common a core management-based paradigm. These management-based initiatives seek to motivate an increase in human oversight of how AI tools are trained and developed. Refinements and systematization of human-guided training techniques will thus be needed to fit within this emerging era of management-based regulatory paradigm. If taken seriously, human-guided training can alleviate some of the technical and ethical pressures on AI, boosting AI performance with human intuition as well as better addressing the needs for fairness and effective explainability. In this paper, we discuss the connection between the emerging management-based regulatory frameworks governing AI and the need for human oversight during training. We broadly cover some of the technical components involved in human-guided training and then argue that the kinds of high-stakes use cases for AI that appear of most concern to regulators should lean more on human-guided training than on data-only training. We hope to foster a discussion between legal scholars and computer scientists involving how to govern a domain of technology that is vast, heterogenous, and dynamic in its applications and risks.
Authors: Yidong Zhao, Joao Tourais, Iain Pierce, Christian Nitsche, Thomas A. Treibel, Sebastian Weing\"artner, Artur M. Schweidtmann, Qian Tao
Abstract: Deep learning (DL)-based methods have achieved state-of-the-art performance for many medical image segmentation tasks. Nevertheless, recent studies show that deep neural networks (DNNs) can be miscalibrated and overconfident, leading to "silent failures" that are risky for clinical applications. Bayesian DL provides an intuitive approach to DL failure detection, based on posterior probability estimation. However, the posterior is intractable for large medical image segmentation DNNs. To tackle this challenge, we propose a Bayesian learning framework using Hamiltonian Monte Carlo (HMC), tempered by cold posterior (CP) to accommodate medical data augmentation, named HMC-CP. For HMC computation, we further propose a cyclical annealing strategy, capturing both local and global geometries of the posterior distribution, enabling highly efficient Bayesian DNN training with the same computational budget as training a single DNN. The resulting Bayesian DNN outputs an ensemble segmentation along with the segmentation uncertainty. We evaluate the proposed HMC-CP extensively on cardiac magnetic resonance image (MRI) segmentation, using in-domain steady-state free precession (SSFP) cine images as well as out-of-domain datasets of quantitative T1 and T2 mapping. Our results show that the proposed method improves both segmentation accuracy and uncertainty estimation for in- and out-of-domain data, compared with well-established baseline methods such as Monte Carlo Dropout and Deep Ensembles. Additionally, we establish a conceptual link between HMC and the commonly known stochastic gradient descent (SGD) and provide general insight into the uncertainty of DL. This uncertainty is implicitly encoded in the training dynamics but often overlooked. With reliable uncertainty estimation, our method provides a promising direction toward trustworthy DL in clinical applications.
Authors: Manxi Lin, Nina Weng, Kamil Mikolaj, Zahra Bashir, Morten Bo S{\o}ndergaard Svendsen, Martin Tolsgaard, Anders Nymark Christensen, Aasa Feragen
Abstract: Shortcut learning is a phenomenon where machine learning models prioritize learning simple, potentially misleading cues from data that do not generalize well beyond the training set. While existing research primarily investigates this in the realm of image classification, this study extends the exploration of shortcut learning into medical image segmentation. We demonstrate that clinical annotations such as calipers, and the combination of zero-padded convolutions and center-cropped training sets in the dataset can inadvertently serve as shortcuts, impacting segmentation accuracy. We identify and evaluate the shortcut learning on two different but common medical image segmentation tasks. In addition, we suggest strategies to mitigate the influence of shortcut learning and improve the generalizability of the segmentation models. By uncovering the presence and implications of shortcuts in medical image segmentation, we provide insights and methodologies for evaluating and overcoming this pervasive challenge and call for attention in the community for shortcuts in segmentation. Our code is public at https://github.com/nina-weng/shortcut_skinseg .
Authors: Juan Manuel Zambrano Chaves, Shih-Cheng Huang, Yanbo Xu, Hanwen Xu, Naoto Usuyama, Sheng Zhang, Fei Wang, Yujia Xie, Mahmoud Khademi, Ziyi Yang, Hany Awadalla, Julia Gong, Houdong Hu, Jianwei Yang, Chunyuan Li, Jianfeng Gao, Yu Gu, Cliff Wong, Mu Wei, Tristan Naumann, Muhao Chen, Matthew P. Lungren, Akshay Chaudhari, Serena Yeung-Levy, Curtis P. Langlotz, Sheng Wang, Hoifung Poon
Abstract: The scaling laws and extraordinary performance of large foundation models motivate the development and utilization of such models in biomedicine. However, despite early promising results on some biomedical benchmarks, there are still major challenges that need to be addressed before these models can be used in real-world clinics. Frontier general-domain models such as GPT-4V still have significant performance gaps in multimodal biomedical applications. More importantly, less-acknowledged pragmatic issues, including accessibility, model cost, and tedious manual evaluation make it hard for clinicians to use state-of-the-art large models directly on private patient data. Here, we explore training open-source small multimodal models (SMMs) to bridge competency gaps for unmet clinical needs in radiology. To maximize data efficiency, we adopt a modular approach by incorporating state-of-the-art pre-trained models for image and text modalities, and focusing on training a lightweight adapter to ground each modality to the text embedding space, as exemplified by LLaVA-Med. For training, we assemble a large dataset of over 697 thousand radiology image-text pairs. For evaluation, we propose CheXprompt, a GPT-4-based metric for factuality evaluation, and demonstrate its parity with expert evaluation. For best practice, we conduct a systematic ablation study on various choices in data engineering and multimodal training. The resulting LlaVA-Rad (7B) model attains state-of-the-art results on standard radiology tasks such as report generation and cross-modal retrieval, even outperforming much larger models such as GPT-4V and Med-PaLM M (84B). The inference of LlaVA-Rad is fast and can be performed on a single V100 GPU in private settings, offering a promising state-of-the-art tool for real-world clinical applications.
Authors: Hang Jung Ling, Salom\'e Bru, Julia Puig, Florian Vix\`ege, Simon Mendez, Franck Nicoud, Pierre-Yves Courand, Olivier Bernard, Damien Garcia
Abstract: Intraventricular vector flow mapping (iVFM) seeks to enhance and quantify color Doppler in cardiac imaging. In this study, we propose novel alternatives to the traditional iVFM optimization scheme by utilizing physics-informed neural networks (PINNs) and a physics-guided nnU-Net-based supervised approach. When evaluated on simulated color Doppler images derived from a patient-specific computational fluid dynamics model and in vivo Doppler acquisitions, both approaches demonstrate comparable reconstruction performance to the original iVFM algorithm. The efficiency of PINNs is boosted through dual-stage optimization and pre-optimized weights. On the other hand, the nnU-Net method excels in generalizability and real-time capabilities. Notably, nnU-Net shows superior robustness on sparse and truncated Doppler data while maintaining independence from explicit boundary conditions. Overall, our results highlight the effectiveness of these methods in reconstructing intraventricular vector blood flow. The study also suggests potential applications of PINNs in ultrafast color Doppler imaging and the incorporation of fluid dynamics equations to derive biomarkers for cardiovascular diseases based on blood flow.
Authors: Yan Ju, Chengzhe Sun, Shan Jia, Shuwei Hou, Zhaofeng Si, Soumyya Kanti Datta, Lipeng Ke, Riky Zhou, Anita Nikolich, Siwei Lyu
Abstract: Deepfakes, as AI-generated media, have increasingly threatened media integrity and personal privacy with realistic yet fake digital content. In this work, we introduce an open-source and user-friendly online platform, DeepFake-O-Meter v2.0, that integrates state-of-the-art methods for detecting Deepfake images, videos, and audio. Built upon DeepFake-O-Meter v1.0, we have made significant upgrades and improvements in platform architecture design, including user interaction, detector integration, job balancing, and security management. The platform aims to offer everyday users a convenient service for analyzing DeepFake media using multiple state-of-the-art detection algorithms. It ensures secure and private delivery of the analysis results. Furthermore, it serves as an evaluation and benchmarking platform for researchers in digital media forensics to compare the performance of multiple algorithms on the same input. We have also conducted detailed usage analysis based on the collected data to gain deeper insights into our platform's statistics. This involves analyzing two-month trends in user activity and evaluating the processing efficiency of each detector.
Authors: Feng Wu, Lei Cui, Shaowen Yao, Shui Yu
Abstract: The prosperity of machine learning has also brought people's concerns about data privacy. Among them, inference attacks can implement privacy breaches in various MLaaS scenarios and model training/prediction phases. Specifically, inference attacks can perform privacy inference on undisclosed target training sets based on outputs of the target model, including but not limited to statistics, membership, semantics, data representation, etc. For instance, infer whether the target data has the characteristics of AIDS. In addition, the rapid development of the machine learning community in recent years, especially the surge of model types and application scenarios, has further stimulated the inference attacks' research. Thus, studying inference attacks and analyzing them in depth is urgent and significant. However, there is still a gap in the systematic discussion of inference attacks from taxonomy, global perspective, attack, and defense perspectives. This survey provides an in-depth and comprehensive inference of attacks and corresponding countermeasures in ML-as-a-service based on taxonomy and the latest researches. Without compromising researchers' intuition, we first propose the 3MP taxonomy based on the community research status, trying to normalize the confusing naming system of inference attacks. Also, we analyze the pros and cons of each type of inference attack, their workflow, countermeasure, and how they interact with other attacks. In the end, we point out several promising directions for researchers from a more comprehensive and novel perspective.
Authors: Stefan P. Hein, Manuel Schultheiss, Andrei Gafita, Raphael Zaum, Farid Yagubbayli, Robert Tauber, Isabel Rauscher, Matthias Eiber, Franz Pfeiffer, Wolfgang A. Weber
Abstract: Assessing tumor response to systemic therapies is one of the main applications of PET/CT. Routinely, only a small subset of index lesions out of multiple lesions is analyzed. However, this operator dependent selection may bias the results due to possible significant inter-metastatic heterogeneity of response to therapy. Automated, AI based approaches for lesion tracking hold promise in enabling the analysis of many more lesions and thus providing a better assessment of tumor response. This work introduces a Siamese CNN approach for lesion tracking between PET/CT scans. Our approach is applied on the laborious task of tracking a high number of bone lesions in full-body baseline and follow-up [68Ga]Ga- or [18F]F-PSMA PET/CT scans after two cycles of [177Lu]Lu-PSMA therapy of metastatic castration resistant prostate cancer patients. Data preparation includes lesion segmentation and affine registration. Our algorithm extracts suitable lesion patches and forwards them into a Siamese CNN trained to classify the lesion patch pairs as corresponding or non-corresponding lesions. Experiments have been performed with different input patch types and a Siamese network in 2D and 3D. The CNN model successfully learned to classify lesion assignments, reaching a lesion tracking accuracy of 83 % in its best configuration with an AUC = 0.91. For remaining lesions the pipeline accomplished a re-identification rate of 89 %. We proved that a CNN may facilitate the tracking of multiple lesions in PSMA PET/CT scans. Future clinical studies are necessary if this improves the prediction of the outcome of therapies.
Authors: Xueqing Wu, Zongyu Lin, Songyan Zhao, Te-Lin Wu, Pan Lu, Nanyun Peng, Kai-Wei Chang
Abstract: Visual programs are executable code generated by large language models to address visual reasoning problems. They decompose complex questions into multiple reasoning steps and invoke specialized models for each step to solve the problems. However, these programs are prone to logic errors, with our preliminary evaluation showing that 58% of the total errors are caused by program logic errors. Debugging complex visual programs remains a major bottleneck for visual reasoning. To address this, we introduce VDebugger, a novel critic-refiner framework trained to localize and debug visual programs by tracking execution step by step. VDebugger identifies and corrects program errors leveraging detailed execution feedback, improving interpretability and accuracy. The training data is generated through an automated pipeline that injects errors into correct visual programs using a novel mask-best decoding technique. Evaluations on six datasets demonstrate VDebugger's effectiveness, showing performance improvements of up to 3.2% in downstream task accuracy. Further studies show VDebugger's ability to generalize to unseen tasks, bringing a notable improvement of 2.3% on the unseen COVR task. Code, data and models are made publicly available at https://github.com/shirley-wu/vdebugger/
Authors: Zihao Jin, Yingying Fang, Jiahao Huang, Caiwen Xu, Simon Walsh, Guang Yang
Abstract: The manifestation of symptoms associated with lung diseases can vary in different depths for individual patients, highlighting the significance of 3D information in CT scans for medical image classification. While Vision Transformer has shown superior performance over convolutional neural networks in image classification tasks, their effectiveness is often demonstrated on sufficiently large 2D datasets and they easily encounter overfitting issues on small medical image datasets. To address this limitation, we propose a Diffusion-based 3D Vision Transformer (Diff3Dformer), which utilizes the latent space of the Diffusion model to form the slice sequence for 3D analysis and incorporates clustering attention into ViT to aggregate repetitive information within 3D CT scans, thereby harnessing the power of the advanced transformer in 3D classification tasks on small datasets. Our method exhibits improved performance on two different scales of small datasets of 3D lung CT scans, surpassing the state of the art 3D methods and other transformer-based approaches that emerged during the COVID-19 pandemic, demonstrating its robust and superior performance across different scales of data. Experimental results underscore the superiority of our proposed method, indicating its potential for enhancing medical image classification tasks in real-world scenarios.