论文检索

Xin Zhang, Yanzhao Zhang, Wen Xie, Mingxin Li, Ziqi Dai, Dingkun Long, Pengjun Xie, Meishan Zhang, Wenjie Li, Min Zhang

Universal Multimodal Retrieval (UMR) aims to enable search across various modalities using a unified model, where queries and candidates can consist of pure text, images, or a combination of both. Previous work has attempted to adopt multimodal large language models (MLLMs) to realize UMR using only text data. However, our preliminary experiments demonstrate that more diverse multimodal training data can further unlock the potential of MLLMs. Despite its effectiveness, the existing multimodal training data is highly imbalanced in terms of modality, which motivates us to develop a training data synthesis pipeline and construct a large-scale, high-quality fused-modal training dataset. Based on the synthetic training data, we develop the General Multimodal Embedder (GME), an MLLM-based dense retriever designed for UMR. Furthermore, we construct a comprehensive UMR Benchmark (UMRB) to evaluate the effectiveness of our approach. Experimental results show that our method achieves state-of-the-art performance among existing UMR methods. Last, we provide in-depth analyses of model scaling and training strategies, and perform ablation studies on both the model and synthetic data.

Minhyeok Lee, Suhwan Cho, Jungho Lee, Sunghun Yang, Heeseung Choi, Ig-Jae Kim, Sangyoun Lee

Open-vocabulary semantic segmentation aims to assign pixel-level labels to images across an unlimited range of classes. Traditional methods address this by sequentially connecting a powerful mask proposal generator, such as the Segment Anything Model (SAM), with a pre-trained vision-language model like CLIP. But these two-stage approaches often suffer from high computational costs, memory inefficiencies. In this paper, we propose ESC-Net, a novel one-stage open-vocabulary segmentation model that leverages the SAM decoder blocks for class-agnostic segmentation within an efficient inference framework. By embedding pseudo prompts generated from image-text correlations into SAM's promptable segmentation framework, ESC-Net achieves refined spatial aggregation for accurate mask predictions. Additionally, a Vision-Language Fusion (VLF) module enhances the final mask prediction through image and text guidance. ESC-Net achieves superior performance on standard benchmarks, including ADE20K, PASCAL-VOC, and PASCAL-Context, outperforming prior methods in both efficiency and accuracy. Comprehensive ablation studies further demonstrate its robustness across challenging conditions.

Yexing Xu, Longguang Wang, Minglin Chen, Sheng Ao, Li Li, Yulan Guo

Although 3D Gaussian Splatting (3DGS) has demonstrated promising results in novel view synthesis, its performance degrades dramatically with sparse inputs and generates undesirable artifacts. As the number of training views decreases, the novel view synthesis task degrades to a highly under-determined problem such that existing methods suffer from the notorious overfitting issue. Interestingly, we observe that models with fewer Gaussian primitives exhibit less overfitting under spare inputs. Inspired by this observation, we propose a Random Dropout Regularization (RDR) to exploit the advantages of low-complexity models to alleviate overfitting. In addition, to remedy the lack of high-frequency details for these models, an Edge-guided Splitting Strategy (ESS) is developed. With these two techniques, our method (termed DropoutGS) provides a simple yet effective plug-in approach to improve the generalization performance of existing 3DGS methods. Extensive experiments show that our DropoutGS produces state-of-the-art performance under sparse views on benchmark datasets including Blender, LLFF, and DTU.

Yuezhi Yang, Qimin Chen, Vladimir G. Kim, Siddhartha Chaudhuri, Qixing Huang, Zhiqin Chen

We introduce the first method for generating Vector Displacement Maps (VDMs): parameterized, detailed geometric stamps commonly used in 3D modeling. Given a single input image, our method first generates multi-view normal maps and then reconstructs a VDM from the normals via a novel reconstruction pipeline. We also propose an efficient algorithm for extracting VDMs from 3D objects, and present the first academic VDM dataset. Compared to existing 3D generative models focusing on complete shapes, we focus on generating parts that can be seamlessly attached to shape surfaces. The method gives artists rich control over adding geometric details to a 3D shape. Experiments demonstrate that our approach outperforms existing baselines. Generating VDMs offers additional benefits, such as using 2D image editing to customize and refine 3D details.

Sheng Fan, Rui Liu, Wenguan Wang, Yi Yang

Navigation instruction generation (NIG), which provides interactive feedback and guidance to humans along a trajectory, is vital for developing embodied agents capable of human-machine communication and collaboration through natural language. Early data-driven methods directly map sequences of past observations to trajectory descriptions on limited datasets, lacking the necessary spatial understanding in complex 3D environments. While recent approaches leverage Large Language Models (LLMs) to improve NIG, they often overlook the global spatial context in navigation, such as the inherent space discretization in maps. Instead of straightforwardly feeding textual descriptions of the map into LLMs, we propose a scene map-based prompt tuning framework for NIG, MAPInstructor, which incorporates map context for parameter-efficient updating of LLMs. MAPInstructor comprises three key components: (i) scene representation encoding, where egocentric observations are projected into 3D voxels for fine-grained scene understanding; (ii) map prompt tuning, which integrates a topological map representation of the entire trajectory into an LLM-based decoder; and (iii) landmark uncertainty assessment, which mitigates hallucinations in landmark predictions, thereby enhancing the reliability and coherence of instruction generation. Extensive experiments on three navigation datasets (i.e., R2R, REVERIE, RxR) confirm the generalization and effectiveness of our algorithm.

Koushik Srivatsan, Fahad Shamshad, Muzammal Naseer, Vishal M. Patel, Karthik Nandakumar

The rapid proliferation of large-scale text-to-image diffusion (T2ID) models has raised serious concerns about their potential misuse in generating harmful content. Although numerous methods have been proposed for erasing undesired concepts from T2ID models, they often provide a false sense of security; concept-erased models (CEMs) can still be manipulated via adversarial attacks to regenerate the erased concept. While a few robust concept erasure methods based on adversarial training have emerged recently, they compromise on utility (generation quality for benign concepts) to achieve robustness and/or remain vulnerable to advanced embedding space attacks. These limitations stem from the failure of robust CEMs to thoroughly search for blind spots in the embedding space. To bridge this gap, we propose STEREO, a novel two-stage framework that employs adversarial training as a first step rather than the only step for robust concept erasure. In the first stage, STEREO employs adversarial training as a vulnerability identification mechanism to search thoroughly enough. In the second robustly erase once stage, STEREO introduces an anchor-concept-based compositional objective to robustly erase the target concept in a single fine-tuning stage, while minimizing the degradation of model utility. We benchmark STEREO against seven state-of-the-art concept erasure methods, demonstrating its superior robustness to both white-box and black-box attacks, while largely preserving utility.

Huakun Liu, Hiroki Ota, Xin Wei, Yutaro Hirao, Monica Perusquia-Hernandez, Hideaki Uchiyama, Kiyoshi Kiyokawa

Sparse wearable inertial measurement units (IMUs) have gained popularity for estimating 3D human motion. However, challenges such as pose ambiguity, data drift, and limited adaptability to diverse bodies persist. To address these issues, we propose UMotion, an uncertainty-driven, online fusing-all state estimation framework for 3D human shape and pose estimation, supported by six integrated, body-worn ultra-wideband (UWB) distance sensors with IMUs. UWB sensors measure inter-node distances to infer spatial relationships, aiding in resolving pose ambiguities and body shape variations when combined with anthropometric data. Unfortunately, IMUs are prone to drift, and UWB sensors are affected by body occlusions. Consequently, we develop a tightly coupled Unscented Kalman Filter (UKF) framework that fuses uncertainties from sensor data and estimated human motion based on individual body shape. The UKF iteratively refines IMU and UWB measurements by aligning them with uncertain human motion constraints in real-time, producing optimal estimates for each. Experiments on both synthetic and real-world datasets demonstrate the effectiveness of UMotion in stabilizing sensor data and the improvement over state of the art in pose accuracy. Code is available at: https://github.com/kk9six/umotion.

Dongseob Kim, Hyunjung Shim

Multi-label classification is crucial for comprehensive image understanding, yet acquiring accurate annotations is challenging and costly. To address this, a recent study suggests exploiting unsupervised multi-label classification leveraging CLIP, a powerful vision-language model. Despite CLIP's proficiency, it suffers from view-dependent predictions and inherent bias, limiting its effectiveness. We propose a novel method that addresses these issues by leveraging multiple views near target objects, guided by Class Activation Mapping (CAM) of the classifier, and debiasing pseudo-labels derived from CLIP predictions. Our Classifier-guided CLIP Distillation (CCD) enables selecting multiple local views without extra labels and debiasing predictions to enhance classification performance. Experimental results validate our method's superiority over existing techniques across diverse datasets. The code is available at https://github.com/k0u-id/CCD.

Fengyi Fu, Lei Zhang, Mengqi Huang, Zhendong Mao

Text-based image editing which aims at generating rigid or non-rigid changes to images conditioned on the given text, has recently attracted considerable interest. Previous works mainly follow the multi-step denoising diffusion paradigm, which adopts a fixed text guidance intensity (i.e., editing intensity) to inject textual features, while ignoring the step-specific editing requirements. This work argues that the editing intensity at each denoising step should be adaptively adjusted conditioned on the historical editing degree, to provide accurate text guidance for the whole denoising process. We thereby propose a novel feedback editing framework (FeedEdit), a training-free method which, explicitly exploits the feedback regulation on editing intensity to ensure precise and harmonious editing at all steps. Specifically, we design (1) Dynamic Editing Degree Perceiving module, which is based on specific frequency-domain filtering, to enhance and exploit the correlation between feature differences and editing degree for perceiving. (2) Proportional-Integral feedback controller, to automatically map the perceived editing errors into appropriate feedback control signals. (3) Phrase-level Regulating Strategy, to achieve fine-grained function-specific regulation of textual features. Extensive experiments demonstrate the superiority of FeedEdit over existing methods in both editability and quality, especially for multi-function editing scenarios.

Aditya Chetan, Guandao Yang, Zichen Wang, Steve Marschner, Bharath Hariharan

Neural fields have become widely used in various fields, from shape representation to neural rendering, and for solving partial differential equations (PDEs). With the advent of hybrid neural field representations like Instant NGP that leverage small MLPs and explicit representations, these models train quickly and can fit large scenes. Yet in many applications like rendering and simulation, hybrid neural fields can cause noticeable and unreasonable artifacts. This is because they do not yield accurate spatial derivatives needed for these downstream applications. In this work, we propose two ways to circumvent these challenges. Our first approach is a post hoc operator that uses local polynomial fitting to obtain more accurate derivatives from pre-trained hybrid neural fields. Additionally, we also propose a self-supervised fine-tuning approach that refines the hybrid neural field to yield accurate derivatives directly while preserving the initial signal. We show applications of our method to rendering, collision simulation, and solving PDEs. We observe that using our approach yields more accurate derivatives, reducing artifacts and leading to more accurate simulations in downstream applications.

Aniruddha Ganguly, Debolina Chatterjee, Wentao Huang, Jie Zhang, Alisa Yurovsky, Travis Steele Johnson, Chao Chen

Recent advances in Spatial Transcriptomics (ST) pair histology images with spatially resolved gene expression profiles, enabling predictions of gene expression across different tissue locations based on image patches. This opens up new possibilities for enhancing whole slide image (WSI) prediction tasks with localized gene expression. However, existing methods fail to fully leverage the interactions between different tissue locations, which are crucial for accurate joint prediction. To address this, we introduce MERGE (Multi-faceted hiErarchical gRaph for Gene Expressions), which combines a multi-faceted hierarchical graph construction strategy with graph neural networks (GNN) to improve gene expression predictions from WSIs. By clustering tissue image patches based on both spatial and morphological features, and incorporating intra- and inter-cluster edges, our approach fosters interactions between distant tissue locations during GNN learning. As an additional contribution, we evaluate different data smoothing techniques that are necessary to mitigate artifacts in ST data, often caused by technical imperfections. We advocate for adopting gene-aware smoothing methods that are more biologically justified. Experimental results on gene expression prediction show that our GNN method outperforms state-of-the-art techniques across multiple metrics.

Wen Li, Chen Liu, Shangshu Yu, Dunqiang Liu, Yin Zhou, Siqi Shen, Chenglu Wen, Cheng Wang

Scene coordinate regression achieves impressive results in outdoor LiDAR localization but requires days of training. Since training needs to be repeated for each new scene, long training times make these impractical for applications requiring time-sensitive system upgrades, such as autonomous driving, drones, robotics, etc. We identify large coverage areas and vast amounts of data in large-scale outdoor scenes as key challenges that limit fast training. In this paper, we propose LightLoc, the first method capable of efficiently learning localization in a new scene at light speed. Beyond freezing the scene-agnostic feature backbone and training only the scene-specific prediction heads, we introduce two novel techniques to address these challenges. First, we introduce sample classification guidance to assist regression learning, reducing ambiguity from similar samples and improving training efficiency. Second, we propose redundant sample downsampling to remove well-learned frames during training, reducing training time without compromising accuracy. In addition, the fast training and confidence estimation characteristics of sample classification enable its integration into SLAM, effectively eliminating error accumulation. Extensive experiments on large-scale outdoor datasets demonstrate that LightLoc achieves state-of-the-art performance with just 1 hour of training--50x faster than existing methods. Our Code is available at https://github.com/liw95/LightLoc.

Jianhui Zhang, Yizhi Luo, Zicheng Zhang, Xuecheng Nie, Bonan Li

Local features aggregation and global information perception are the fundamental to point cloud segmentation. However, existing works often fall short in effectively identifying semantic relevant neighbors and face challenges in endowing each point with high-level information. Here, we propose CamPoint, an innovative method that employs virtual cameras to solve above problems. The core of CamPoint lies in introducing the novel camera visibility feature for points, where each dimension encodes the visibility of that point from a specific camera. Leveraging this feature, we propose the camera perspective slice distance for accurate relevant neighbor searching and design the camera parameter embedding to deliver rich feature representations for global interaction. Specifically, the camera perspective slice distance between two points is defined as a similarity metric derived from their camera visibility features, whereby an increased number of shared cameras observing both points corresponds to a reduced distance between them. To effectively facilitate global semantic perception, we assign each camera an optimizable embedding and then integrate these embeddings into the original spatial features based on visibility attributes, thereby obtaining high-level features enriched with camera priors. Additionally, the state space model characterized by linear computational complexity is employed as the operator to achieve global learning with efficiency. Comprehensive experiments on multiple datasets shows that our CamPoint surpasses the current state-of-the-art in multiple datasets, achieving low training cost and fast inference speed.

Aishik Konwer, Zhijian Yang, Erhan Bas, Cao Xiao, Prateek Prasanna, Parminder Bhatia, Taha Kass-Hout

Foundational models such as the Segment Anything Model (SAM) are gaining traction in medical imaging segmentation, supporting multiple downstream tasks. However, such models are supervised in nature, still relying on large annotated datasets or prompts supplied by experts. Conventional techniques such as active learning to alleviate such limitations are limited in scope and still necessitate continuous human involvement and complex domain knowledge for label refinement or establishing reward ground truth. To address these challenges, we propose an enhanced Segment Anything Model (SAM) framework that utilizes annotation-efficient prompts generated in a fully unsupervised fashion, while still capturing essential semantic, location, and shape information through contrastive language-image pretraining and visual question answering. We adopt the direct preference optimization technique to design an optimal policy that enables the model to generate high-fidelity segmentations with simple ratings or rankings provided by a virtual annotator simulating the human annotation process. State-of-the-art performance of our framework in tasks such as lung segmentation, breast tumor segmentation, and organ segmentation across various modalities, including X-ray, ultrasound, and abdominal CT, justifies its effectiveness in low-annotation data scenarios.

Qiang Li, Jian Ruan, Fanghao Wu, Yuchi Chen, Zhihua Wei, Wen Shen

Recently, many self-supervised pre-training methods have been proposed to improve the performance of deep neural networks (DNNs) for 3D point clouds processing. However, the common mechanism underlying the effectiveness of different pre-training methods remains unclear. In this paper, we use game-theoretic interactions as a unified approach to explore the common mechanism of pre-training methods. Specifically, we decompose the output score of a DNN into the sum of numerous effects of interactions, with each interaction representing a distinct 3D substructure of the input point cloud. Based on the decomposed interactions, we draw the following conclusions. (1) The common mechanism across different pre-training methods is that they enhance the strength of high-order interactions encoded by DNNs, which represent complex and global 3D structures, while reducing the strength of low-order interactions, which represent simple and local 3D structures. (2) Sufficient pre-training and adequate fine-tuning data for downstream tasks further reinforce the mechanism described above. (3) Pre-training methods carry a potential risk of reducing the transferability of features encoded by DNNs. Inspired by the observed common mechanism, we propose a new method to directly enhance the strength of high-order interactions and reduce the strength of low-order interactions encoded by DNNs, improving performance without the need for pre-training on large-scale datasets. Experiments show that our method achieves performance comparable to traditional pre-training methods.

Seungmin Baek, Soyul Lee, Hayeon Jo, Hyesong Choi, Dongbo Min

Transfer learning paradigm has driven substantial advancements in various vision tasks. However, as state-of-the-art models continue to grow, classical full fine-tuning often becomes computationally impractical, particularly in multi-task learning (MTL) setup where training complexity increases proportional to the number of tasks. Consequently, recent studies have explored Parameter-Efficient Fine-Tuning (PEFT) for MTL architectures. Despite some progress, these approaches still exhibit limitations in capturing fine-grained, task-specific features that are crucial to MTL. In this paper, we introduce Task-Adaptive Dynamic transFormer, termed TADFormer, a novel PEFT framework that performs task-aware feature adaptation in the fine-grained manner by dynamically considering task-specific input contexts. TADFormer proposes the parameter-efficient prompting for task adaptation and the Dynamic Task Filter (DTF) to capture task information conditioned on input contexts. Experiments on the PASCAL-Context benchmark demonstrate that the proposed method achieves higher accuracy in dense scene understanding tasks, while reducing the number of trainable parameters by up to 8.4 times when compared to full fine-tuning of MTL models. TADFormer also demonstrates superior parameter efficiency and accuracy compared to recent PEFT methods. Our code is available at supplementary material.

Zhenpeng Huang, Xinhao Li, Jiaqi Li, Jing Wang, Xiangyu Zeng, Cheng Liang, Tao Wu, Xi Chen, Liang Li, Limin Wang

Multimodal Large Language Models (MLLMs) have significantly progressed in offline video understanding. However, applying these models to real-world scenarios, such as autonomous driving and human-computer interaction, presents unique challenges due to the need for real-time processing of continuous online video streams. To this end, this paper presents systematic efforts from three perspectives: evaluation benchmark, model architecture, and training strategy. First, we introduce OVBench, a comprehensive question-answering benchmark designed to evaluate models' ability to perceive, memorize, and reason within online video contexts. It features 6 core task types across three temporal contexts--past, current, and future--forming 16 subtasks from diverse datasets. Second, we propose a new Pyramid Memory Bank (PMB) that effectively retains key spatiotemporal information in video streams. Third, we proposed an offline-to-online learning paradigm, designing an interleaved dialogue format for online video data and constructing an instruction-tuning dataset tailored for online video training. This framework led to the development of VideoChat-Online, a robust and efficient model for online video understanding. Despite the lower computational cost and higher efficiency, VideoChat-Online outperforms existing state-of-the-art offline and online models across popular offline video benchmarks and OVBench, demonstrating the effectiveness of our model architecture and training strategy.

Mingze Sun, Junhao Chen, Junting Dong, Yurun Chen, Xinyu Jiang, Shiwei Mao, Puhua Jiang, Jingbo Wang, Bo Dai, Ruqi Huang

Recent advances in generative models have enabled high-quality 3D character reconstruction from multi-modal. However, animating these generated characters remains a challenging task, especially for complex elements like garments and hair, due to the lack of large-scale datasets and effective rigging methods. To address this gap, we curate AnimeRig, a large-scale dataset with detailed skeleton and skinning annotations. Building upon this, we propose DRiVE, a novel framework for generating and rigging 3D human characters with intricate structures. Unlike existing methods, DRiVE utilizes a 3D Gaussian representation, facilitating efficient animation and high-quality rendering. We further introduce GSDiff, a 3D Gaussian-based diffusion module that predicts joint positions as spatial distributions, overcoming the limitations of regression-based approaches. Extensive experiments demonstrate that DRiVE achieves precise rigging results, enabling realistic dynamics for clothing and hair, and surpassing previous methods in both quality and versatility. Code, dataset and visualization results are available at https://DRiVEAvatar.github.io/.

Zixuan Wang, Duo Peng, Feng Chen, Yuwei Yang, Yinjie Lei

Conditional image synthesis is a crucial task with broad applications, such as artistic creation and virtual reality. However, current generative methods are often task-oriented with a narrow scope, handling a restricted condition with constrained applicability. In this paper, we propose a novel approach that treats conditional image synthesis as the modular combination of diverse fundamental condition units. Specifically, we divide conditions into three primary units: text, layout, and drag. To enable effective control over these conditions, we design a dedicated alignment module for each. For the text condition, we introduce a Dense Concept Alignment (DCA) module, which achieves dense visual-text alignment by drawing on diverse textual concepts. For the layout condition, we propose a Dense Geometry Alignment (DGA) module to enforce comprehensive geometric constraints that preserve the spatial configuration. For the drag condition, we introduce a Dense Motion Alignment (DMA) module to apply multi-level motion regularization, ensuring that each pixel follows its desired trajectory without visual artifacts. By flexibly inserting and combining these alignment modules, our framework enhances the model's adaptability to diverse conditional generation tasks and greatly expands its application range. Extensive experiments demonstrate the superior performance of our framework across a variety of conditions, including textual description, segmentation mask (bounding box), drag manipulation, and their combinations. Code is available at https://github.com/ZixuanWang0525/DADG

Chengwei Zheng, Lixin Xue, Juan Zarate, Jie Song

3D Gaussian Splatting techniques have enabled efficient photo-realistic rendering of static scenes. Recent works have extended these approaches to support surface reconstruction and tracking. However, tracking dynamic surfaces with 3D Gaussians remains challenging due to complex topology changes, such as surfaces appearing, disappearing, or splitting. To address these challenges, we propose GauSTAR, a novel method that achieves photo-realistic rendering, accurate surface reconstruction, and reliable 3D tracking for general dynamic scenes with changing topology. Given multi-view captures as input, GauSTAR binds Gaussians to mesh faces to represent dynamic objects. For surfaces with consistent topology, GauSTAR maintains the mesh topology and tracks the meshes using Gaussians. For regions where topology changes, GauSTAR adaptively unbinds Gaussians from the mesh, enabling accurate registration and generation of new surfaces based on these optimized Gaussians. Additionally, we introduce a surface-based scene flow method that provides robust initialization for tracking between frames. Experiments demonstrate that our method effectively tracks and reconstructs dynamic surfaces, enabling a range of applications. Our project page with the code release is available at https://eth-ait.github.io/GauSTAR/.