WACV-2025-Papers

会议时间：2025年2月28日–3月4日

会议网址：https://wacv2025.thecvf.com/

查看2025年综述文献点这里↘️2025-CV-Surveys

2025 年论文分类汇总戳这里

↘️WACV-2025-Papers ↘️CVPR-2025-Papers ↘️ICCV-2025-Papers

2024 年论文分类汇总戳这里

↘️WACV-2024-Papers ↘️CVPR-2024-Papers ↘️ECCV-2024-Papers

2023 年论文分类汇总戳这里

2022 年论文分类汇总戳这里

2021 年论文分类汇总戳这里

2020 年论文分类汇总戳这里

❣❣❣ WACV 2025 论文分类整理已完成

:loudspeaker::loudspeaker::loudspeaker:获奖论文

:trophy:最佳论文(算法)

• RayGauss: Volumetric Gaussian-Based Ray Casting for Photorealistic Novel View Synthesis
  :star:code
  :house:project

:trophy:最佳论文(应用)

• Optimizing Vision-Language Model for Road Crossing Intention Estimation

:trophy:最佳学生论文

• GeoDiffuser: Geometry-Based Image Editing with Diffusion Models
  :star:code
  :house:project

:trophy:最佳学生论文荣誉提名奖

• Cross-Domain and Cross-Dimension Learning for Image-to-Graph Transformers
  :star:code

:trophy:Time of time award: (tie)

• Deeply-Learned Feature for Age Estimation
• Bayesian Multi-object Tracking Using Motion Context from Multiple Objects

目录

49.计算成像

• FaVoR: Features via Voxel Rendering for Camera Relocalization
• Dense Scene Reconstruction from Light-Field Images Affected by Rolling Shutter光场图像密集场景重建
• PrivateEye: In-Sensor Privacy Preservation Through Optical Feature Separation光学
• Solar Multimodal Transformer: Intraday Solar Irradiance Predictor using Public Cameras and Time Series
• TaCOS: Task-Specific Camera Optimization with Simulation

48.Protecting copyright(保护版权)

• Towards Secure and Usable 3D Assets: A Novel Framework for Automatic Visible Watermarking
• 图像复制检测
  • Relational Self-Supervised Distillation with Compact Descriptors for Image Copy Detection

47.Sketch(草图)

• 3D Edge Sketch from Multiview Images
• PICASSO: A Feed-Forward Framework for Parametric Inference of CAD Sketches via Rendering Self-Supervision
• ColorizeDiffusion: Improving Reference-Based Sketch Colorization with Latent Diffusion Model

46.Sound

• SoundSil-DS: Deep Denoising and Segmentation of Sound-Field Images with Silhouettes利用轮廓对声场图像进行深度去噪和分割
• NarrAD: Automatic Generation of Audio Descriptions for Movies with Rich Narrative Context
• NowYouSee Me: Context-Aware Automatic Audio Description
• EgoSonics: Generating Synchronized Audio for Silent Egocentric Videos
  :house:project
• SoundLoc3D: Invisible 3D Sound Source Localization and Classification Using a Multimodal RGB-D Acoustic Camera
• Temporally Streaming Audio-Visual Synchronization for Real-World Videos
• Multimodal Interpretable Depression Analysis using Visual Physiological Audio and Textual Data
• Unsupervised Video Highlight Detection by Learning from Audio and Visual Recurrence
• VioPose: Violin Performance 4D Pose Estimation by Hierarchical Audiovisual Inference
• VMAs: Video-to-Music Generation via Semantic Alignment in Web Music Videos

45.Transformer

• LowFormer: Hardware Efficient Design for Convolutional Transformer Backbones
• Bandit Based Attention Mechanism in Vision Transformers
• AMP-ViT: Optimizing Vision Transformer Efficiency with Adaptive Mixed-Precision Post-Training Quantization
• Channel Propagation Networks for Refreshable Vision Transformer
• Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers
• SpectFormer: Frequency and Attention is What You Need in a Vision Transformer
• Image Adaptation for Colour Vision Deficient Viewers using Vision Transformers
• QuantAttack: Exploiting Quantization Techniques to Attack Vision Transformers
• TORE: Token Recycling in Vision Transformers for Efficient Active Visual Exploration
• Adversarial Attention Deficit: Fooling Deformable Vision Transformers with Collaborative Adversarial Patches
• Beyond Grids: Exploring Elastic Input Sampling for Vision Transformers
• Weight Copy and Low-Rank Adaptation for Few-Shot Distillation of Vision Transformers

44.Dense Prediction(密集预测)

• Optimizing Dense Visual Predictions Through Multi-Task Coherence and Prioritization
  :star:code
• Cross-Task Affinity Learning for Multitask Dense Scene Predictions密集场景预测

43.Neural Radiance Fields

• Radiance Field-Based Pose Estimation via Decoupled Optimization Under Challenging Initial Conditions
• MFNeRF: Memory Efficient NeRF with Mixed-Feature Hash Table
• GANESH: Generalizable NeRF for Lensless Imaging
  :star:code
• TRNeRF: Restoring Blurry Rolling Shutter and Noisy Thermal Images with Neural Radiance Fields
• BASED: Bundle-Adjusting Surgical Endoscopic Dynamic Video Reconstruction using Neural Radiance Fields
• ARF-Plus: Controlling Perceptual Factors in Artistic Radiance Fields for 3D Scene Stylization
• Self-Aligning Depth-Regularized Radiance Fields for Asynchronous RGB-D Sequences
• 新视图合成
  • RendBEV: Semantic Novel View Synthesis for Self-Supervised Bird's Eye View Segmentation
  • VaLID: Variable-Length Input Diffusion for Novel View Synthesis
  • RayGauss: Volumetric Gaussian-Based Ray Casting for Photorealistic Novel View Synthesis
  • GauFRe: Gaussian Deformation Fields for Real-Time Dynamic Novel View Synthesis
  • MSI-NeRF: Linking Omni-Depth with View Synthesis through Multi-Sphere Image Aided Generalizable Neural Radiance Field
  • FluoNeRF: Fluorescent Novel-View Synthesis under Novel Light Source Colors
  • SGD: Street View Synthesis with Gaussian Splatting and Diffusion Prior
• 渲染
  • Global-Guided Focal Neural Radiance Field for Large-Scale Scene Rendering
  • OccFlowNet: Occupancy Estimation via Differentiable Rendering and Occupancy Flow
  • PRoGS: Progressive Rendering of Gaussian Splats渲染
  • NeuManifold: Neural Watertight Manifold Reconstruction with Efficient and High-Quality Rendering Support

42.Industrial Anomaly Detection(工业缺陷检测)

• SPACE: SPAtial-aware Consistency rEgularization for anomaly detection in Industrial applications
• Adaptive Deviation Learning for Visual Anomaly Detection with Data Contamination
• Anomaly Detection for People with Visual Impairments Using an Egocentric 360-Degree Camera
• ROADS: Robust Prompt-driven Multi-Class Anomaly Detection under Domain Shift
• FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Data
  :star:code
• Single-Layer Distillation with Fourier Convolutions for Texture Anomaly Detection
• Looking at Model Debiasing through the Lens of Anomaly Detection
• AnomalyDINO: Boosting Patch-Based Few-Shot Anomaly Detection with DINOv2
• Removing Geometric Bias in One-Class Anomaly Detection with Adaptive Feature Perturbation
• 图像异常检测
  • Heterogeneous Datasets for Unsupervised Image Anomaly Detection
• 异常定位
  • Towards Zero-shot 3D Anomaly Localization
• 异常分割
  • Towards Accurate Unified Anomaly Segmentation

41.Anomaly Detection(异常检测)

• 奇异值检测
  • Robust Novelty Detection through Style-Conscious Feature Ranking
• OOD
  • Exploiting Inter-Sample Information for Long-Tailed Out-of-Distribution Detection
  • Identity Curvature Laplace Approximation for Improved Out-of-Distribution Detection
  • CRAFT: Class Ranking Aware Fine-Tuning for Enhanced Out-of-Distribution Detection
  • Finding Dino: A Plug-and-Play Framework for Zero-Shot Detection of Out-of-Distribution Objects using Prototypes
  • CLIPScope: Enhancing Zero-Shot OOD Detection with Bayesian Scoring

40.Deepfake

• DeCLIP: Decoding CLIP representations for deepfake localization
  :star:code
• Texture Shape and Order Matter: A New Transformer Design for Sequential DeepFake Detection
• AI生成图像检测
  • Reducing the Content Bias for AI-Generated Image Detection
  • InvisMark: Invisible and Robust Watermarking for AI-Generated Image Provenance
• 错误信息检测
  • Similarity over Factuality: Are we Making Progress on Multimodal Out-of-Context Misinformation Detection?
  • Can Out-of-Domain Data Help to Learn Domain-Specific Prompts for Multimodal Misinformation Detection?错误信息检测

39.Robots(机器人)

• Transferring Foundation Models for Generalizable Robotic Manipulation机器人操作
• Avatar
  • Gaussian Déjà-vu: Creating Controllable 3D Gaussian Head-Avatars with Enhanced Generalization and Personalization Abilities
  • DivAvatar: Diverse 3D Avatar Generation with a Single Prompt
• Try-On
  • Street TryOn: Learning In-the-Wild Virtual Try-On from Unpaired Person Images
• SLAM
  • Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstruction
    :star:code
• 室内定位
  • Multi-Surrogate-Teacher Assistance for Representation Alignment in Fingerprint-based Indoor Localization
• 视觉位置识别
  • Breaking the Frame: Visual Place Recognition by Overlap Prediction

38.HOI Detection(交互检测)

• Unleashing Potentials of Vision-Language Models for Zero-Shot HOI Detection
• 手物交互
  • A Versatile and Differentiable Hand-Object Interaction Representation

37.Scene(场景)

• LLaVA-SpaceSGG: Visual Instruct Tuning for Open-vocabulary Scene Graph Generation with Enhanced Spatial Relations
  :star:code
• DDS: Decoupled Dynamic Scene-Graph Generation Network
• Enhancing Scene Graph Generation with Hierarchical Relationships and Commonsense Knowledge
• Effective Scene Graph Generation by Statistical Relation Distillation

36.Object Pose Estimation(物体姿态估计)//

• Generalizable Single-view Object Pose Estimation by Two-side Generating and Matching
  :star:code
• 重识别
  • DMPT: Decoupled Modality-Aware Prompt Tuning for Multi-Modal Object Re-Identification

35.Dataset/Benchmark(数据集/基准)

• SynDRA: Synthetic Dataset for Railway Applications
• High-Fidelity Document Stain Removal via A Large-Scale Real-World Dataset and A Memory-Augmented Transformer
• Needles & Haystacks: Dataset and Benchmark for Domain-Agnostic Image-Based Rigid Slice-to-Volume Registration
• The FineView Dataset:A 3D Scanned Multi-View Object Dataset of Fine-Grained Category Instances
• PureForest: A Large-Scale Aerial Lidar and Aerial Imagery Dataset for Tree Species Classification in Monospecific Forests
• IRIS-VIS: A New Dataset for Visibility Estimation in an Industrial Environment
• GTA-HDR: A Large-Scale Synthetic Dataset for HDR Image Reconstruction
• CT to PET Translation: A Large-scale Dataset and Domain-Knowledge-Guided Diffusion Approach
  :star:code
• PV-VTT: A Privacy-Centric Dataset for Mission-Specific Anomaly Detection and Natural Language Interpretation
• CLIPping Imbalances: A Novel Evaluation Baseline and PEARL Dataset for Pedestrian Attribute Recognition
• SynDroneVision: A Synthetic Dataset for Image-Based Drone Detection
• SEED4D: A Synthetic Ego--Exo Dynamic 4D Data Generator, Driving Dataset and Benchmark
  :star:code
  :star:code
• SEED4D: A Synthetic Ego-Exo Dynamic 4D Data Generator Driving Dataset and Benchmark
• DrIFT: Autonomous Drone Dataset with Integrated Real and Synthetic Data, Flexible Views, and Transformed Domains
  :star:code
• DrIFT: Autonomous Drone Dataset with Integrated Real and Synthetic Data Flexible Views and Transformed Domains
• TimberVision: A Multi-Task Dataset and Framework for Log-Component Segmentation and Tracking in Autonomous Forestry Operations
  :star:code
• A Pipeline and NIR-Enhanced Dataset for Parking Lot Segmentation
• 3D Understanding of Deformable Linear Objects: Datasets and Transferability Benchmark
• CISOL: An Open and Extensible Dataset for Table Structure Recognition in the Construction Industry
• SANPO: A Scene Understanding Accessibility and Human Navigation Dataset
• Sign Language Recognition: A Large-Scale Multi-View Dataset and Comprehensive Evaluation
• CycleCrash: A Dataset of Bicycle Collision Videos for Collision Prediction and Analysis
• A Semantically Impactful Image Manipulation Dataset: Characterizing Image Manipulations using Semantic Significance
• 基准
  • GazeSearch: Radiology Findings Search Benchmark
  • ARTeFACT: Benchmarking Segmentation Models on Diverse Analogue Media Damage
    :star:code
  • CardioSyntax: End-to-End SYNTAX Score Prediction - Dataset Benchmark and Method
  • Oriented Cell Dataset: A Dataset and Benchmark for Oriented Cell Detection and Applications
  • ANTHROPOS-V: Benchmarking the Novel Task of Crowd Volume Estimation
  • SALVE: A 3D Reconstruction Benchmark of Wounds from Consumer-Grade Videos
  • Mind the Prompt: A Novel Benchmark for Prompt-Based Class-Agnostic Counting
  • VG-SSL: Benchmarking Self-Supervised Representation Learning Approaches for Visual Geo-Localization
  • Multi-Label Continual Learning for the Medical Domain: A Novel Benchmark
  • OpenCapBench: A Benchmark to Bridge Pose Estimation and Biomechanics
  • UAL-Bench: The First Comprehensive Unusual Activity Localization Benchmark
  • MIP-GAF: A MLLM-Annotated Benchmark for Most Important Person Localization and Group Context Understanding

34.Vision-Language(视觉语言)

• Active Learning for Vision-Language Models
• Active Learning for Vision Language Models
• Generalist YOLO: Towards Real-Time End-to-End Multi-Task Visual Language Models
• LLM-Generated Rewrite and Context Modulation for Enhanced Vision Language Models in Digital Pathology
• @Bench: Benchmarking Vision-Language Models for Human-centered Assistive Technology
  :star:code
  :house:project
• Style-Pro: Style-Guided Prompt Learning for Generalizable Vision-Language Models
• Who Brings the Frisbee: Probing Hidden Hallucination Factors in Large Vision-Language Model via Causality Analysis
• Retaining and Enhancing Pre-trained Knowledge in Vision-Language Models with Prompt Ensembling
• Enhancing Vision-Language Few-Shot Adaptation with Negative Learning
• DPA: Dual Prototypes Alignment for Unsupervised Adaptation of Vision-Language Models
• Optimizing Vision-Language Model for Road Crossing Intention Estimation
• OpenCity3D: What do Vision-Language Models Know About Urban Environments?
• Automated Evaluation of Large Vision-Language Models on Self-Driving Corner Cases
• 视频语言
  • ACE: Action Concept Enhancement of Video-Language Models in Procedural Videos
• VLN
  • To Ask or Not to Ask? Detecting Absence of Information in Vision and Language Navigation
  • Hijacking Vision-and-Language Navigation Agents with Adversarial Environmental Attacks
  • GroundingMate: Aiding Object Grounding for Goal-Oriented Vision-and-Language Navigation
  • ELBA: Learning by Asking for Embodied Visual Navigation and Task Completion
• LLM
  • Learning Multiple Object States from Actions via Large Language Models
  • LLM-RSPF: Large Language Model-Based Robotic System Planning Framework for Domain Specific Use-Cases
• MLLM
  • Can Multimodal Large Language Models Truly Perform Multimodal In-Context Learning?
  • User-in-the-Loop Evaluation of Multimodal LLMs for Activity Assistance
  • Multi-Modal Large Language Model with RAG Strategies in Soccer Commentary Generation
  • MLLM-Tool: A Multimodal Large Language Model for Tool Agent Learning
  • Multi-Modal Large Language Models are Effective Vision Learners
• Visual Grounding
  • Fine-Grained Spatial and Verbal Losses for 3D Visual Grounding
  • Data-Efficient 3D Visual Grounding via Order-Aware Referring
  • Learning Visual Grounding from Generative Vision and Language Model
• 农业+视觉语言
  • Leveraging Vision Language Models for Specialized Agricultural Tasks
  • AgroGPT: Efficient Agricultural Vision-Language Model with Expert Tuning
  • Flowering Time Prediction of Wheat from DIA-MS Data预测小麦开花期

33.Semi/self-supervised learning(半/自监督)//

• 自监督
  • HEX: Hierarchical Emergence Exploitation in Self-Supervised Algorithms
  • Self-Supervised Learning with Probabilistic Density Labeling for Rainfall Probability Estimation
    :star:code
  • Local Masked Reconstruction for Efficient Self-Supervised Learning on High-Resolution Images
• 半监督
  • PivotAlign: Improve Semi-Supervised Learning by Learning Intra-Class Heterogeneity and Aligning with Pivots
  • Defending Against Repetitive Backdoor Attacks on Semi-Supervised Learning through Lens of Rate-Distortion-Perception Trade-Off
• 新类别发现
  • Towards On-the-Fly Novel Category Discovery in Dynamic Long-Tailed Distributions

32.MC/KD/Pruning(模型压缩/知识蒸馏/剪枝)//

• A Multi-Task Supervised Compression Model for Split Computing
• KD
  • On Explaining Knowledge Distillation: Measuring and Visualising the Knowledge Transfer Process
  • KDC-MAE: Knowledge Distilled Contrastive Mask Auto-Encoder
  • Dropout Connects Transformers and CNNs: Transfer General Knowledge for Knowledge Distillation
  • InDistill: Information Flow-Preserving Knowledge Distillation for Model Compression
  • Information Extraction from Heterogeneous Documents without Ground Truth Labels using Synthetic Label Generation and Knowledge Distillation
  • EchoDFKD: Data-Free Knowledge Distillation for Cardiac Ultrasound Segmentation using Synthetic Data
  • Comparative Knowledge Distillation
  • SMDAF: A Scalable Sidewalk Material Data Acquisition Framework with Bidirectional Cross-Modal Knowledge Distillation
  • ChromaDistill : Colorizing Monochrome Radiance Fields with Knowledge Distillation
• 剪枝
  • VLTP: Vision-Language Guided Token Pruning for Task-Oriented Segmentation
  • Shapley Consensus Deep Learning for Ensemble Pruning
  • Patch Ranking: Token Pruning as Ranking Prediction for Efficient CLIP
  • Pruning One More Token is Enough: Leveraging Latency-Workload Non-Linearities for Vision Transformers on the Edge
  • Information Theoretic Pruning of Coupled Channels in Deep Neural Networks
• 量化
  • PTQ4VM: Post-Training Quantization for Visual Mamba
    :star:code
  • Dequantization and Color Transfer with Diffusion Models
  • Data Generation for Hardware-Friendly Post-Training Quantization
  • Difficulty Diversity and Plausibility: Dynamic Data-Free Quantization
  • Q-TempFusion: Quantization-Aware Temporal Multi-Sensor Fusion on Bird's-Eye View Representation

31.Neural Architecture Search(神经架构搜索)

• MONAS-ESNN: Multi-Objective Neural Architecture Search for Efficient Spiking Neural Networks
• Delta-NAS: Difference of Architecture Encoding for Predictor-Based Evolutionary Neural Architecture Search

30.Few/Zero-Shot Learning/DG/A(小/零样本/域泛化/域适应)

• 域泛化
  • ERM++: An Improved Baseline for Domain Generalization
  • Crafting Distribution Shifts for Validation and Training in Single Source Domain Generalization
  • FRAUD-Net: Fraud News Detection using Sample Uncertainty & Domain Aware Generalized Network
  • Domain-Generalized Object Anti-Spoofing: Bridging Gaps and Patch Selection for Robust Detection Across Domains
  • Fair Domain Generalization with Heterogeneous Sensitive Attributes Across Domains
  • Domain-Guided Weight Modulation for Semi-Supervised Domain Generalization
  • FDS: Feedback-Guided Domain Synthesis with Multi-Source Conditional Diffusion Models for Domain Generalization
  • Domain Generalization using Large Pretrained Models with Mixture-of-Adapters
  • ConDiSR: Contrastive Disentanglement and Style Regularization for Single Domain Generalizatio
• 域适应
  • Label Calibration in Source Free Domain Adaptation
  • AH-OCDA: Amplitude-based Curriculum Learning and Hopfield Segmentation Model for Open Compound Domain Adaptation
  • Feature Fusion Transferability Aware Transformer for Unsupervised Domain Adaptation
  • Instance-Warp: Saliency Guided Image Warping for Unsupervised Domain Adaptation
  • Memory-Efficient Pseudo-Labeling for Online Source-Free Universal Domain Adaptation using a Gaussian Mixture Model
  • Combining Inherent Knowledge of Vision-Language Models with Unsupervised Domain Adaptation through Strong-Weak Guidance
  • Transferable-Guided Attention is All You Need for Video Domain Adaptation
  • When Cars Meet Drones: Hyperbolic Federated Learning for Source-Free Domain Adaptation in Adverse Weather
  • Ad^2mix: Adversarial and Adaptive Mixup for Unsupervised Domain Adaptation
• 零样本
  • Unified Framework for Open-World Compositional Zero-Shot Learning
  • Just Shift It: Test-Time Prototype Shifting for Zero-Shot Generalization with Vision-Language Models
  • Test-Time Low Rank Adaptation via Confidence Maximization for Zero-Shot Generalization of Vision-Language Models
  • SenCLIP: Enhancing zero-shot land-use mapping for Sentinel-2 with ground-level prompting
  • HOPE: A Memory-Based and Composition-Aware Framework for Zero-Shot Learning with Hopfield Network and Soft Mixture of Experts
  • Learning to Identify Seen Unseen and Unknown in the Open World: A Practical Setting for Zero-Shot Learning
  • PC-GZSL: Prior Correction for Generalized Zero Shot Learning

29.Deep Learning

• Prior2Posterior: Model Prior Correction for Long-Tailed Learning
• SADDLe: Sharpness-Aware Decentralized Deep Learning with Heterogeneous Data
• DNN
  • Guardian of the Ensembles: Introducing Pairwise Adversarially Robust Loss for Resisting Adversarial Attacks in DNN Ensembles
  • DARDA: Domain-Aware Real-Time Dynamic Neural Network Adaptation动态神经网络

28.GNN/GCN

• WiGNet: Windowed Vision Graph Neural Network
• SIGNN - Star Identification using Graph Neural Networks

27.Machine Learning(机器学习)

• 度量学习
  • Uncertainty-Guided Metric Learning without Labels
• 迁移学习
  • Learning Unified Distance Metric Across Diverse Data Distributions with Parameter-Efficient Transfer Learning
• 机器遗忘
  • Revisiting Machine Unlearning with Dimensional Alignment
• 增量学习
  • Self-Supervised Incremental Learning of Object Representations from Arbitrary Image Sets
• 类增量
  • Covariance-based Space Regularization for Few-shot Class Incremental Learning
  • A Reality Check on Pre-training for Exemplar-free Class-Incremental Learning
  • Dynamic Adapter Tuning for Long-Tailed Class-Incremental Learning
  • ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learning
  • Strategic Base Representation Learning via Feature Augmentations for Few-Shot Class Incremental Learning
  • Are Exemplar-Based Class Incremental Learning Models Victim of Black-Box Poison Attacks?
  • TACLE: Task and Class-Aware Exemplar-Free Semi-Supervised Class Incremental Learning
• 主动学习
  • CRAAC: Consistency Regularised Active Learning with Automatic Corrections for Real-Life Road Image Annotations
• 联邦学习
  • Navigating Heterogeneity and Privacy in One-Shot Federated Learning with Diffusion Models
  • Predicting Event Memorability using Personalized Federated Learning
  • Achieving Byzantine-Resilient Federated Learning via Layer-Adaptive Sparsified Model Aggregation
  • Identify Backdoored Model in Federated Learning via Individual Unlearning
  • MLLM-LLaVA-FL: Multimodal Large Language Model Assisted Federated Learning
• 对比学习
  • MOOSS: Mask-Enhanced Temporal Contrastive Learning for Smooth State Evolution in Visual Reinforcement Learning
  • Tuned Contrastive Learning
  • Contrastive Learning of Image Representations Guided by Spatial Relations
  • CATALOG: A Camera Trap Language-Guided Contrastive Learning Model
  • PLReMix: Combating Noisy Labels with Pseudo-Label Relaxed Contrastive Representation Learning
• 持续学习
  • Online-LoRA: Task-free Online Continual Learning via Low Rank Adaptation
    :star:code
  • Memory-efficient Continual Learning with Neural Collapse Contrastive
  • Exploring the Stability Gap in Continual Learning: The Role of the Classification Head
  • Semantic Prompting with Image Token for Continual Learning
  • Towards Unbiased Continual Learning: Avoiding Forgetting in the Presence of Spurious Correlations
  • EvoCL: Continual Learning over Evolving Domains
  • AdaPrefix++: Integrating Adapters Prefixes and Hypernetwork for Continual Learning
• 多任务学习
  • Diffusion-based Visual Anagram as Multi-task Learning
    :star:code
• 对抗
  • FAIR-TAT: Improving Model Fairness using Targeted Adversarial Training
  • PoolAtnRes: Towards Generalisable Differential Morphing Attack Detection变形攻击检测
  • Knockoff Branch: Model Stealing Attack via Adding Neurons in the Pre-Trained Model
  • Low-Frequency Black-Box Backdoor Attack via Evolutionary Algorithm
  • Can Adversarial Examples Be Parsed to Reveal Victim Model Information?
  • When Visual State Space Model Meets Backdoor Attacks后门攻击
  • Pre-Trained Multiple Latent Variable Generative Models are Good Defenders Against Adversarial Attacks

26.Motion Generation(人体运动生成)

• SyncViolinist: Music-Oriented Violin Motion Generation Based on Bowing and Fingering
  :star:code
• GHOST: Grounded Human Motion Generation with Open Vocabulary Scene-and-Text Contexts
• Generation of Complex 3D Human Motion by Temporal and Spatial Composition of Diffusion Models
• MoRAG - Multi-Fusion Retrieval Augmented Generation for Human Motion
• UniTMGE: Uniform Text-Motion Generation and Editing Model via Diffusion
• 基于骨架的运动预测
  • Geometry-Aware Deep Learning for 3D Skeleton-Based Motion Prediction

25.Style Transfer(风格迁移)

• Meta-Learning for Color-to-Infrared Cross-Modal Style Transfer
• D-LUT: Photorealistic Style Transfer via Diffusion Process
• Mamba-ST: State Space Model for Efficient Style Transfer

24.GAN/Image Synthesis(图像生成)

• Unsupervised Single-Image Intrinsic Image Decomposition with LiDAR Intensity Enhanced Training图像分解
• ZeroComp: Zero-Shot Object Compositing from Image Intrinsics via Diffusion
• MixDiff: Mixing Natural and Synthetic Images for Robust Self-Supervised Representations合成图像
• 3D Synthesis for Architectural Design
• ARTIST: Improving the Generation of Text-Rich Images with Disentangled Diffusion Models and Large Language Models
• 360PanT: Training-Free Text-Driven 360-Degree Panorama-to-Panorama Translation文本驱动的360度全景到全景翻译
• Clarity Amidst Blur: A Deterministic Method for Synthetic Generation of Water Droplets on Camera Lenses水滴合成
• SpotDiffusion: A Fast Approach for Seamless Panorama Generation Over Time全景图生成
• Attribute Diffusion: Diffusion Driven Diverse Attribute Editing多样化属性编辑
• DiffQRCoder: Diffusion-Based Aesthetic QR Code Generation with Scanning Robustness Guided Iterative Refinement美学二维码生成
• GAN
  • WINE : Wavelet-Guided GAN Inversion and Editing for High-Fidelity Refinement
• 图像合成
  • GeoPos: A Minimal Positional Encoding for Enhanced Fine-Grained Details in Image Synthesis using Convolutional Neural Networks
  • Rethinking Cluster-Conditioned Diffusion Models for Label-Free Image Synthesis
• 纹理生成
  • Make-A-Texture: Fast Shape-Aware Texture Generation in 3 Seconds
  • Make-A-Texture: Fast Shape-Aware 3D Texture Generation in 3 Seconds
• 图像生成
  • RAW-Diffusion: RGB-Guided Diffusion Models for High-Fidelity RAW Image Generation(https://github.com/SonyResearch/RAW-Diffusion)
  • MegaFusion: Extend Diffusion Models towards Higher-resolution Image Generation without Further Tuning
    :star:code
    :house:project
  • Beta Sampling is All You Need: Efficient Image Generation Strategy for Diffusion Models using Stepwise Spectral Analysis
  • Refining Text-to-Image Generation: Towards Accurate Training-Free Glyph-Enhanced Image Generation
  • Skip-and-Play: Depth-Driven Pose-Preserved Image Generation for Any Objects
  • FineControlNet: Fine-Level Text Control for Image Generation with Spatially Aligned Text Control Injection
• 食谱生成
  • Retrieval Augmented Recipe Generation
• 图像编辑
  • Uniform Attention Maps: Boosting Image Fidelity in Reconstruction and Editing
    :star:code
  • Diffusion-Based Conditional Image Editing through Optimized Inference with Guidance
  • Dual-Schedule Inversion: Training- and Tuning-Free Inversion for Real Image Editing
  • LIME: Localized Image Editing via Attention Regularization in Diffusion Models
  • ReEdit: Multimodal Exemplar-Based Image Editing
  • GeoDiffuser: Geometry-Based Image Editing with Diffusion Models
  • DragText: Rethinking Text Embedding in Point-Based Image Editing
  • Incorporating Task Progress Knowledge for Subgoal Generation in Robotic Manipulation through Image Edits
• 文本-图像
  • DreamBlend: Advancing Personalized Fine-tuning of Text-to-Image Diffusion Models
  • Disentangling Subject-Irrelevant Elements in Personalized Text-to-Image Diffusion via Filtered Self-Distillation
  • Counting Guidance for High Fidelity Text-to-Image Synthesis
  • Elucidating Optimal Reward-Diversity Tradeoffs in Text-to-Image Diffusion Models
  • Detecting Origin Attribution for Text-to-Image Diffusion Models
  • AlignIT: Enhancing Prompt Alignment in Customization of Text-to-Image Models
  • Good Seed Makes a Good Crop: Discovering Secret Seeds in Text-to-Image Diffusion Models
  • Improving Faithfulness of Text-to-Image Diffusion Models through Inference Intervention
  • Structured Human Assessment of Text-to-Image Generative Models
  • Controlling Human Shape and Pose in Text-to-Image Diffusion Models via Domain Adaptation
  • An Image is Worth Multiple Words: Multi-Attribute Inversion for Constrained Text-to-Image Synthesis
• 布局到图像生成
  • STAY Diffusion: Styled Layout Diffusion Model for Diverse Layout-to-Image Generation
• 三维生成
  • My3DGen: A Scalable Personalized 3D Generative Model
• 文本-3D
  • GANFusion: Feed-Forward Text-to-3D with Diffusion in GAN Space
    :star:code
  • Deep Geometric Moments Promote Shape Consistency in Text-to-3D Generation
  • HexaGen3D: StableDiffusion is One Step Away from Fast and Diverse Text-to-3D Generation
• 图像-图像翻译
  • Uncertainty-Aware Regularization for Image-to-Image Translation
• 视频编辑
  • Ada-VE: Training-Free Consistent Video Editing using Adaptive Motion Prior
  • MagicStick: Controllable Video Editing via Control Handle Transformations
  • 文本-视频编辑
    • FastVideoEdit: Leveraging Consistency Models for Efficient Text-to-Video Editing
• 视频生成
  • Fine-Grained Controllable Video Generation via Object Appearance and Context
  • Generating Long-Take Videos via Effective Keyframes and Guidance生成长镜头视频
  • Dance Any Beat: Blending Beats with Visuals in Dance Video Generation
  • Corgi: Cached Memory Guided Video Generation
  • TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion Models
• 视频合成
  • Contrastive Sequential-Diffusion Learning: Non-Linear and Multi-Scene Instructional Video Synthesis
• 轮胎足迹生成
  • CTIP: Towards Accurate Tabular-to-Image Generation for Tire Footprint Generation
• 扩散模型
  • Enhancing Image Layout Control with Loss-Guided Diffusion Models
  • GeoGuide: Geometric Guidance of Diffusion Models
  • Inverse Problems with Diffusion Models: A MAP Estimation Perspective
  • Diffusion Model Guided Sampling with Pixel-Wise Aleatoric Uncertainty Estimation
  • SimuScope: Realistic Endoscopic Synthetic Dataset Generation through Surgical Simulation and Diffusion Models
    :star:code
  • Negative-Prompt Inversion: Fast Image Inversion for Editing with Text-Guided Diffusion Models
  • SODA: Spectral Orthogonal Decomposition Adaptation for Diffusion Models
  • Improving Conditional Diffusion Models through Re-Noising from Unconditional Diffusion Priors
  • MemControl: Mitigating Memorization in Diffusion Models via Automated Parameter Selection
  • Disentangling Disentangled Representations: Towards Improved Latent Units via Diffusion Models
  • Elucidating the Solution Space of Extended Reverse-Time SDE for Diffusion Models
  • DiffuseKronA: A Parameter Efficient Fine-Tuning Method for Personalized Diffusion Models
  • CusConcept: Customized Visual Concept Decomposition with Diffusion Models
  • CharDiff: Improving Sampling Convergence via Characteristic Function Consistency in Diffusion Models
  • Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think

23.Visual Question Answering(视觉问答)

• 视频问答
  • DAM: Dynamic Adapter Merging for Continual Video QA Learning
  • Foundation Models and Adaptive Feature Selection: A Synergistic Approach to Video Question Answering
  • Perceive, Query & Reason: Enhancing Video QA with Question-Guided Temporal Queries
• 视觉问答
  • CL-Cross VQA: A Continual Learning Benchmark for Cross-Domain Visual Question Answering
  • AdQuestA: Knowledge-Guided Visual Question Answer Framework for Advertisements
  • One VLM to Keep it Learning: Generation and Balancing for Data-Free Continual Visual Question Answering
  • Unsupervised Domain Adaptive Visual Question Answering in the Era of Multi-Modal Large Language Models
  • Visual Robustness Benchmark for Visual Question Answering (VQA)
• 图表问答
  • Advancing Chart Question Answering with Robust Chart Component Recognition
• 表格问答
  • TRH2TQA: Table Recognition with Hierarchical Relationships to Table Question-Answering on Business Table Images

22.OCR

• 手写文档识别
  • DocTTT: Test-Time Training for Handwritten Document Recognition using Meta-Auxiliary Learning
• 场景文本识别
  • Stratified Domain Adaptation: A Progressive Self-Training Approach for Scene Text Recognition
• 场景文本编辑
  • FASTER: A Font-Agnostic Scene Text Editing and Rendering Framework
• 文本变化检测
  • Text Change Detection in Multilingual Documents Using Image Comparison
• 文本多边形检测
  • TPD-STR: Text Polygon Detection with Split Transformers
• 表结构识别
  • Treading Towards Privacy-Preserving Table Structure Recognition

21.3D(三维重建\三维视觉)

• NPL-MVPS: Neural Point-Light Multi-View Photometric Stereo多视角光度立体
• HybridDepth: Robust Metric Depth Fusion by Leveraging Depth from Focus and Single-Image Priors
• LIPIDS: Learning-based Illumination Planning In Discretized (Light) Space for Photometric Stereo
• Instructive3D: Editing Large Reconstruction Models with Text Instructions
• Scene-LLM: Extending Language Model for 3D Visual Reasoning
• Towards a Training Free Approach for 3D Scene Editing
• CRAFT: Designing Creative and Functional 3D Objects
• NeRFs are Mirror Detectors: using Structural Similarity for Multi-View Mirror Scene Reconstruction with 3D Surface Primitives
• VortSDF: 3D Modeling with Centroidal Voronoi Tesselation on Signed Distance Field
• EfficientMorph: Parameter-Efficient Transformer-Based Architecture for 3D Image Registration
• 3DGS
  • Planar Gaussian Splatting
  • UW-GS: Distractor-Aware 3D Gaussian Splatting for Enhanced Underwater Scene Reconstruction
  • EdgeGaussians - 3D Edge Mapping via Gaussian Splatting
  • Localized Gaussian Splatting Editing with Contextual Awareness
  • DN-Splatter: Depth and Normal Priors for Gaussian Splatting and Meshing
  • ELMGS: Enhancing Memory and Computation Scalability through Compression for 3D Gaussian Splatting
  • OmniGS: Fast Radiance Field Reconstruction using Omnidirectional Gaussian Splatting
• 三维重建
  • Assessing the Quality of 3D Reconstruction in the Absence of Ground Truth: Application to a Multimodal Archaeological Dataset
  • Multi-HexPlanes: A Lightweight Map Representation for Rendering and 3D Reconstruction
  • Semantic Segmentation Method for Automated Indoor 3D Reconstruction Based on Architectural-Knowledge-Aware Features
  • DreaMo: Articulated 3D Reconstruction from a Single Casual Video
  • Sparse-View 3D Reconstruction of Clothed Humans via Normal Maps
  • Comparative Evaluation of 3D Reconstruction Methods for Object Pose Estimation
  • ALSTER: A Local Spatio-Temporal Expert for Online 3D Semantic Reconstruction
• 表面重建
  • Spatially-Adaptive Hash Encodings for Neural Surface Reconstruction
  • Transientangelo: Few-Viewpoint Surface Reconstruction using Single-Photon Lidar
  • PVT: An Implicit Surface Reconstruction Framework via Point Voxel Geometric-Aware Transformer
• 深度估计
  • GET-UP: GEomeTric-aware Depth Estimation with Radar Points UPsampling
  • MDCN-PS: Monocular-Depth-Guided Coarse Normal Attention for Robust Photometric Stereo
  • Revisiting Disparity from Dual-Pixel Images: Physics-Informed Lightweight Depth Estimation
  • MonoPP: Metric-Scaled Self-Supervised Monocular Depth Estimation by Planar-Parallax Geometry in Automotive Applications
    :star:code
  • OmniDiffusion: Reformulating 360 Monocular Depth Estimation using Semantic and Surface Normal Conditioned Diffusion
  • Enhancing Monocular Depth Estimation with Multi-Source Auxiliary Tasks
  • CabNIR: A Benchmark for In-Vehicle Infrared Monocular Depth Estimation
• 房屋布局估计
  • uLayout: Unified Room Layout Estimation for Perspective and Panoramic Images
• 三维场景理解
  • Calib3D: Calibrating Model Preferences for Reliable 3D Scene Understanding
• 三维语义场景补全
  • DepthSSC: Monocular 3D Semantic Scene Completion via Depth-Spatial Alignment and Voxel Adaptation
• 3D形状补全
  • A Recipe for Geometry-Aware 3D Mesh Transformers
  • 3D Shape Completion using Multi-Resolution Spectral Encoding
  • ShapeMorph: 3D Shape Completion via Blockwise Discrete Diffusion

20.Point Cloud(点云)

• BioNet and NeFF: Crop Biomass Prediction from Point Clouds to Drone Imagery
• Point Cloud Color Upsampling with Attention-Based Coarse Colorization and Refinement
• On-the-Fly Object-aware Representative Point Selection in Point Cloud
• PocoLoco: A Point Cloud Diffusion Model of Human Shape in Loose Clothing
  :star:code
• Test-Time Adaptation in Point Clouds: Leveraging Sampling Variation with Weight Averaging
• Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud
• 3D 点云
  • Test-Time Adaptation of 3D Point Clouds via Denoising Diffusion Models
    :star:code
  • Learning under Noisy Labels Spurious Points and Diverse Structures: TS40K a 3D Point Cloud Dataset of Rural Terrain and Electrical Transmission Systems
  • Learning Semantic Part-Based Graph Structure for 3D Point Cloud Domain Generalization
  • Adversarial Learning Based Knowledge Distillation on 3D Point Clouds
  • RGB2Point: 3D Point Cloud Generation from Single RGB Images
  • Continual Learning in 3D Point Clouds: Employing Spectral Techniques for Exemplar Selection
• 点云分类
  • Point-GN: A Non-Parametric Network Using Gaussian Positional Encoding for Point Cloud Classification
• 点云分割
  • ZAHA: Introducing the Level of Facade Generalization and the Large-Scale Point Cloud Facade Semantic Segmentation Benchmark Dataset
• 点云配准
  • XR-MBT: Multi-modal Full Body Tracking for XR through Self-Supervision with Learned Depth Point Cloud Registration

19.Video

• NeuroViG - Integrating Event Cameras for Resource-Efficient Video Grounding
• MVFNet: Multipurpose Video Forensics Network using Multiple Forms of Forensic Evidence
• GEXIA: Granularity Expansion and Iterative Approximation for Scalable Multi-Grained Video-Language Learning视频语言学习
• 视频监控
  • DashCop: Automated E-Ticket Generation for Two-Wheeler Traffic Violations using Dashcam Videos使用Dashcam视频自动生成两轮车交通违章电子票
• 视频理解
  • Ego-VPA: Egocentric Video Understanding with Parameter-Efficient Adaptation
  • Paladin: Understanding Video Intentions in Political Advertisement Videos
  • Frame by Familiar Frame: Understanding Replication in Video Diffusion Models
• 视频时许定位
  • FlashVTG: Feature Layering and Adaptive Score Handling Network for Video Temporal Grounding
    :star:code
• 视频异常检测
  • Graph-Jigsaw Conditioned Diffusion Model for Skeleton-Based Video Anomaly Detection
  • Guess Future Anomalies from Normalcy: Forecasting Abnormal Behavior in Real-World Videos
  • Distilling Aggregated Knowledge for Weakly-Supervised Video Anomaly Detection
  • Discriminative Score Suppression for Weakly Supervised Video Anomaly Detection
  • MissionGNN: Hierarchical Multimodal GNN-Based Weakly Supervised Video Anomaly Recognition with Mission-Specific Knowledge Graph Generation
• 视频镜像检测
  • RGB-D Video Mirror Detection
• 视频时刻检索
  • Moment of Untruth: Dealing with Negative Queries in Video Moment Retrieval
  • Background-Aware Moment Detection for Video Moment Retrieval
• 视频帧插值
  • Explicit Guidance for Robust Video Frame Interpolation against Discontinuous Motions
• 视频稳定
  • SensorFlow: Sensor and Image Fused Video Stabilization

18.Person Re-id(行人重识别)

• 重识别
  • ReMix: Training Generalized Person Re-identification on a Mixture of Data
  • Re-Identifying People in Video via Learned Temporal Attention and Multi-Modal Foundation Models重识别
  • VILLS : Video-Image Learning to Learn Semantics for Person Re-Identification
  • AnonyNoise: Anonymizing Event Data with Smart Noise to Outsmart Re-Identification and Preserve Privacy
    :star:code
  • 换衣重识别
    • DLCR: A Generative Data Expansion Framework via Diffusion for Clothes-Changing Person Re-ID
      :star:code
  • 可见光和红外重识别
    • Bidirectional Multi-Step Domain Generalization for Visible-Infrared Person Re-Identification
• 行人搜索
  • Swap Path Network for Robust Person Search Pre-training
    :star:code
• 步态识别
  • VM-Gait: Multi-Modal 3D Representation Based on Virtual Marker for Gait Recognition
  • GaitCloud: Leveraging Spatial-Temporal Information for LiDAR-Base Gait Recognition with A True-3D Gait Representation
  • GaitContour: Efficient Gait Recognition Based on a Contour-Pose Representation
  • MimicGait: A Model Agnostic Approach for Occluded Gait Recognition using Correlational Knowledge Distillation
• 人群密度完成
  • CrowdMAC: Masked Crowd Density Completion for Robust Crowd Density Forecasting

17.Action Detection(动作检测)

• Learning to Visually Connect Actions and their Effects
• ActionDiffusion: An Action-Aware Diffusion Model for Procedure Planning in Instructional Videos
• Inferring Past Human Actions in Homes with Abductive Reasoning
• 动作检测
  • VAD
    • Boosting Semi-Supervised Video Action Detection with Temporal Context
  • 零样本动作检测
    • Spatio-Temporal Context Prompting for Zero-Shot Action Detection
  • 开放词汇动作检测
    • Exploiting VLM Localizability and Semantics for Open Vocabulary Action Detection
      :star:code
• 动作识别
  • Differentially Private Integrated Decision Gradients (IDG-DP) for Radar-Based Human Activity Recognition
  • Importance-Guided Interpretability and Pruning for Video Transformers in Driver Action Recognition
  • 基于骨架的动作识别
    • Autoregressive Adaptive Hypergraph Transformer for Skeleton-based Activity Recognition
• 群组动作识别
  • LiGAR: LiDAR-Guided Hierarchical Transformer for Multi-Modal Group Activity Recognition
• 时许动作定位
  • SADA: Semantic Adversarial Unsupervised Domain Adaptation for Temporal Action Localization
  • LoSA: Long-Short-Range Adapter for Scaling End-to-End Temporal Action Localization
  • Exploring Scalability of Self-Training for Open-Vocabulary Temporal Action Localization
• 动作质量评估
  • I3D-AE-LSTM: A 2-Stream Autoencoder for Action Quality Assessment using a Newly Created Cricket Batsman Video Dataset
• 社交互动识别
  • Loose Social-Interaction Recognition in Real-World Therapy Scenarios

16.Human Pose Estimation(人体姿态估计)

• EgoCast: Forecasting Egocentric Human Pose in the Wild
• Recoverable Anonymization for Pose Estimation: A Privacy-Enhancing Approach
• 人体解析
  • D2FP: Learning Implicit Prior for Human Parsing
• 人体重塑
  • Structure-Aware Human Body Reshaping with Adaptive Affinity-Graph Network
• 人体重建
  • DiHuR: Diffusion-Guided Generalizable Human Reconstruction
  • EasyRet3D: Uncalibrated Multi-View Multi-Human 3D Reconstruction and Tracking
• 人体网格恢复
  • Utilizing Uncertainty in 2D Pose Detectors for Probabilistic 3D Human Mesh Recovery
    :star:code
  • PostoMETRO: Pose Token Enhanced Mesh Transformer for Robust 3D Human Mesh Recovery
  • DiffMesh: A Motion-Aware Diffusion Framework for Human Mesh Recovery from Videos
• 三维姿态估计
  • ReMP: Reusable Motion Prior for Multi-domain 3D Human Pose Estimation and Motion Inbetweening
    :star:code
  • LiCamPose: Combining Multi-View LiDAR and RGB Cameras for Robust Single-Timestamp 3D Human Pose Estimation
  • Event-Guided Video Transformer for End-to-End 3D Human Pose Estimation
  • Event-Guided Fusion-Mamba for Context-Aware 3D Human Pose Estimation
  • STRIDE: Single-Video Based Temporally Continuous Occlusion-Robust 3D Pose Estimation
  • BioPose: Biomechanically-Accurate 3D Pose Estimation from Monocular Videos
• 人体运动恢复
  • RopeTP: Global Human Motion Recovery via Integrating Robust Pose Estimation with Diffusion Trajectory Prior
• 手势生成
  • Conditional GAN for Enhancing Diffusion Models in Efficient and Authentic Global Gesture Generation from Audios
• 手势识别
  • Automated Patient Positioning with Learned 3D Hand Gestures
  • ConvMixFormer- A Resource-Efficient Convolution Mixer for Transformer-Based Dynamic Hand Gesture Recognition
• 手部姿态估计
  • XPose: Towards Extreme Low Light Hand Pose Estimation
• 婴儿动作生成
  • Infant Action Generative Modeling
• 手语翻译
  • A Spatio-Temporal Representation Learning as an Alternative to Traditional Glosses in Sign Language Translation and Production

15.Medical Image Progress(医学影响处理)

• Survival Prediction in Lung Cancer through Multi-Modal Representation Learning预测肺癌患者生存率
• Physiology-Aware PolySnake for Coronary Vessel Segmentation血管分割
• TRUST: Time-Domain Residual Unsupervised Stability Technique for Improved Heart Rate Estimation心率估计
• Attention-Guided Masked Autoencoders for Learning Image Representations
• Multi-Aperture Transformers for 3D (MAT3D) Segmentation of Clinical and Microscopic Images3D分割
• Tumor Synthesis Conditioned on Radiomics肿瘤合成
• Data Augmentation for Surgical Scene Segmentation with Anatomy-Aware Diffusion Models手术场景分割
• MFTrans: A Multi-Resolution Fusion Transformer for Robust Tumor Segmentation in Whole Slide Images肿瘤分割
• AutoProSAM: Automated Prompting SAM for 3D Multi-Organ Segmentation3D多器官分割
• Learning Anatomy-Disease Entangled Representation学习解剖学疾病纠缠表征
• GAUDA: Generative Adaptive Uncertainty-Guided Diffusion-Based Augmentation for Surgical Segmentation手术分割
• Endoscopic Scoring and Localization in Unconstrained Clinical Trial Videos内镜评分和定位
• SAM-Mamba: Mamba Guided SAM Architecture for Generalized Zero-Shot Polyp Segmentation息肉分割
• Uncertainty Awareness Enables Efficient Labeling for Cancer Subtyping in Digital Pathology数字病理学
• Federated-Continual Dynamic Segmentation of Histopathology Guided by Barlow Continuity组织病理学
• F2FLDM: Latent Diffusion Models with Histopathology Pre-Trained Embeddings for Unpaired Frozen Section to FFPE Translation
• Investigating Imaging Annotation and Self-Supervision for the Classification of Continuously Developing Cells in Histological Whole Slide Images
• PK-YOLO: Pretrained Knowledge Guided YOLO for Brain Tumor Detection in Multiplanar MRI Slices
  :star:code
• Volumetric Conditioning Module to Control Pretrained Diffusion Models for 3D Medical Images
  :star:code
• AMNCutter: Affinity-Attention-Guided Multi-View Normalized Cutter for Unsupervised Surgical Instrument Segmentation
  :star:code
• SAM-DA: Decoder Adapter for Efficient Medical Domain Adaptation
• Multimodal Fusion Learning with Dual Attention for Medical Imaging
  :star:code
• LQ-Adapter: ViT-Adapter with Learnable Queries for Gallbladder Cancer Detection from Ultrasound Image
  :star:code
• Reviving Poor Object Segmentations in OOD Medical Images using Variational-Deep-PCA Modeling on Segmentation Maps with Sampling-Free Learning
• Training-Free Medical Image Inverses via Bi-Level Guided Diffusion Models
• Sli2Vol+: Segmenting 3D Medical Images Based on an Object Estimation Guided Correspondence Flow Network
• Data-Efficient Alignment in Medical Imaging via Reconfigurable Generative Networks
• Multi-Resolution Guided 3D GANs for Medical Image Translation
• Relaxing Binary Constraints in Contrastive Vision-Language Medical Representation Learning
• MAISI: Medical AI for Synthetic Imaging
• FMD: Comprehensive Data Compression in Medical Domain via Fused Matching Distillation
• 皮肤病分类
  • A Novel Perspective for Multi-Modal Multi-Label Skin Lesion Classification
  • Enhancing Skin Disease Diagnosis: Interpretable Visual Concept Discovery with SAM皮肤病诊断
• 青光眼检测
  • DiffuPT: Class Imbalance Mitigation for Glaucoma Detection via Diffusion Based Generation and Model Pretraining
• 异常分割
  • Self-Supervised Anomaly Segmentation via Diffusion Models with Dynamic Transformer UNet
• 心脏图像的合成分割
  • Compositional Segmentation of Cardiac Images Leveraging Metadata
    :star:code
• 视网膜眼底图像增强
  • CUNSB-RFIE: Context-Aware Unpaired Neural Schrodinger Bridge in Retinal Fundus Image Enhancement
  • Context-Aware Optimal Transport Learning for Retinal Fundus Image Enhancement
• 医学图像超分辨率
  • SeCo-INR: Semantically Conditioned Implicit Neural Representations for Improved Medical Image Super-Resolution
• 医学图像配准
  • NestedMorph: Enhancing Deformable Medical Image Registration with Nested Attention Mechanisms
• 医学图像分类
  • CAMEL: Confidence-Aware Multi-Task Ensemble Learning with Spatial Information for Retina OCT Image Classification and Segmentation
• 医学图像分割
  • Generalizable Single-Source Cross-modality Medical Image Segmentation via Invariant Causal Mechanisms
    :star:code
  • MulModSeg: Enhancing Unpaired Multi-Modal Medical Image Segmentation with Modality-Conditioned Text Embedding and Alternating Training
    :star:code
  • Effective and Efficient Medical Image Segmentation with Hierarchical Context Interaction
  • Personalized Mixture of Experts for Multi-Site Medical Image Segmentation
  • Frequency-Domain Refinement of Vision Transformers for Robust Medical Image Segmentation under Degradation
  • 半监督医学图像分割
    • Uncertainty-Guided Cross Attention Ensemble Mean Teacher for Semi-supervised Medical Image Segmentation
      :star:code
    • Cross Image Feature Perturbation with Pseudo Label Fusion for Semi-Supervised Medical Image Segmentation
    • Learning Semi-Supervised Medical Image Segmentation from Spatial Registration
• 医学放射科报告生成
  • ORID: Organ-Regional Information Driven Framework for Radiology Report Generation
• MRI
  • Continuous Spatio-Temporal Memory Networks for 4D Cardiac Cine MRI Segmentation
  • Feasibility of Federated Learning from Client Databases with Different Brain Diseases and MRI Modalities
  • DiaMond: Dementia Diagnosis with Multi-Modal Vision Transformers Using MRI and PET
  • MRI Reconstruction with Regularized 3D Diffusion Model (R3DM)
  • MambaRecon: MRI Reconstruction with Structured State Space Models
  • McCaD: Multi-Contrast MRI Conditioned Adaptive Adversarial Diffusion Model for High-Fidelity MRI Synthesis
• X-ray
  • DeepCA: Deep Learning-Based 3D Coronary Artery Tree Reconstruction from Two 2D Non-Simultaneous X-ray Angiography Projections
  • Self-Supervised Pre-Training with Diffusion Model for Few-Shot Landmark Detection in X-Ray Images
  • Foundation X: Integrating Classification Localization and Segmentation through Lock-Release Pretraining Strategy for Chest X-ray Analysis
  • TempA-VLP: Temporal-Aware Vision-Language Pretraining for Longitudinal Exploration in Chest X-ray Image
  • OTCXR: Rethinking Self-Supervised Alignment using Optimal Transport for Chest X-ray Analysis
• CT
  • AC-IND: Sparse CT Reconstruction Based on Attenuation Coefficient Estimation and Implicit Neural Distribution
  • Improving Pelvic MR-CT Image Alignment with Self-Supervised Reference-Augmented Pseudo-CT Generation Framework
• 海马体分割
  • NCAdapt: Dynamic Adaptation with Domain-Specific Neural Cellular Automata for Continual Hippocampus Segmentation
• 肺气道分割
  • DMRN: A Dynamical Multi-Order Response Network for the Robust Lung Airway Segmentation

14.Autonomous Driving(自动驾驶)

• A Generic Vehicle-to-Sensor Calibration Framework车辆到传感器校准框架
• CoVLA: Comprehensive Vision-Language-Action Dataset for Autonomous Driving
  :house:project
• S3PT: Scene Semantics and Structure Guided Clustering to Boost Self-Supervised Pre-Training for Autonomous Driving
• LORD: Large Models Based Opposite Reward Design for Autonomous Driving
• DyRoNet: Dynamic Routing and Low-Rank Adapters for Autonomous Driving Streaming Perception
• Robust Long-Range Perception Against Sensor Misalignment in Autonomous Vehicles
• Bandwidth-Efficient Communication Modelling for Autonomous Vehicle Collaborative Perception
• 轨迹预测
  • Socially-Informed Reconstruction for Pedestrian Trajectory Forecasting
  • Hierarchical Light Transformer Ensembles for Multimodal Trajectory Forecasting
• 车道线检测
  • ElasticLaneNet: An Efficient Geometry-Flexible Lane Detection Framework
  • HeightLane: BEV Heightmap Guided 3D Lane Detection
• 3D占用预测
  • OccLoff: Learning Optimized Feature Fusion for 3D Occupancy Prediction
  • PVP: Polar Representation Boost for 3D Semantic Occupancy Prediction

13.Biomedical(生物特征识别)

• On Which Data Distribution (Synthetic or Real) We Should Rely for Soft Biometric Classification软生物特征分类
• 指纹检测
  • ColFigPhotoAttnNet: Reliable Finger Photo Presentation Attack Detection Leveraging Window-Attention on Color Spaces
• 虹膜检测
  • A Parametric Approach to Adversarial Augmentation for Cross-Domain Iris Presentation Attack Detection
• 基于虹膜图像的死后间隔估计
  • Forensic Iris Image-Based Post-Mortem Interval Estimation

12.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)

• PGRID: Power Grid Reconstruction in Informal Developments Using High-Resolution Aerial Imagery
• Pix2Poly: A Sequence Prediction Method for End-to-end Polygonal Building Footprint Extraction from Remote Sensing Imagery
  :star:code
• Skyeyes: Ground Roaming using Aerial View Images
• Aerial Mirage: Unmasking Hallucinations in Large Vision Language Models
• OPTIMUS: Observing Persistent Transformations in Multi-Temporal Unlabeled Satellite-Data
• UAV
  • Detecting Wildfires on UAVs with Real-Time Segmentation Trained by Larger Teacher Models
  • WeedsGalore: A Multispectral and Multitemporal UAV-Based Dataset for Crop and Weed Segmentation in Agricultural Maize Fields
• 分割
  • SegDesicNet: Lightweight Semantic Segmentation in Remote Sensing with Geo-Coordinate Embeddings for Domain Adaptation
• 超分辨率
  • Can Location Embeddings Enhance Super-Resolution of Satellite Imagery?
• 跟踪/检测
  • Click&Describe: Multimodal Grounding and Tracking for Aerial Objects
  • Psych-Occlusion: using Visual Psychophysics for Aerial Detection of Occluded Persons during Search and Rescue
• 航空图像合成
  • Cross-View Meets Diffusion: Aerial Image Synthesis with Geometry and Text Guidance
• 遥感变化检测
  • Deep Metric Learning for Unsupervised Remote Sensing Change Detection
  • DDPM-CD: Denoising Diffusion Probabilistic Models as Feature Extractors for Remote Sensing Change Detection
  • A Mamba-Based Siamese Network for Remote Sensing Change Detection
  • Improving Zero-Shot Object-Level Change Detection by Incorporating Visual Correspondence变化检测
• 稀有动物行为
  • Sifting through the Haystack - Efficiently Finding Rare Animal Behaviors in Large-Scale Datasets

11.Object Tracking(目标跟踪)

• BroadTrack: Broadcast Camera Tracking for Soccer
• MFTIQ: Multi-Flow Tracker with Independent Matching Quality Estimation
  :star:code
• Improving Accuracy and Generalization for Efficient Visual Tracking
• Vision-Based Landing Guidance through Tracking and Orientation Estimation
• TAM-VT: Transformation-Aware Multi-Scale Video Transformer for Segmentation and Tracking
• 多人跟踪
  • DragonTrack: Transformer-Enhanced Graphical Multi-Person Tracking in Complex Scenarios
• 3D跟踪
  • Context-Aware Outlier Rejection for Robust Multi-View 3D Tracking of Similar Small Birds in An Outdoor Aviary
• 点跟踪
  • Segment Anything Meets Point Tracking
  • EgoPoints: Advancing Point Tracking for Egocentric Videos
    :star:code

10.Object Detection(目标检测)

• Uncertainty Aware Interest Point Detection and Description兴趣点检测与描述
• DT-LSD: Deformable Transformer-Based Line Segment Detection线段检测
• EDMB: Edge Detector with Mamba边缘检测
• MVMD: A Multi-View Approach for Enhanced Mirror Detection镜像检测
• Recurrence-based Vanishing Point Detection
• No Annotations for Object Detection in Art through Stable Diffusion
  :star:code
• Label Convergence: Defining an Upper Performance Bound in Object Recognition through Contradictory Annotations
• Bit-Flip Induced Latency Attacks in Object Detection
• Multispectral Object Detection Enhanced by Cross-Modal Information Complementary and Cosine Similarity Channel Resampling Modules
• DiL: An Explainable and Practical Metric for Abnormal Uncertainty in Object Detection
• Enhancing Novel Object Detection via Cooperative Foundational Models
• ERUP-YOLO: Enhancing Object Detection Robustness for Adverse Weather Condition by Unified Image-Adaptive Processing
• Mixed Patch Visible-Infrared Modality Agnostic Object Detection
  :star:code
  :house:project
• Enhancing Embodied Object Detection with Spatial Feature Memory
• Interactive Object Detection for Tiny Objects in Large Remotely Sensed Images
• Shape-Biased Texture Agnostic Representations for Improved Textureless and Metallic Object Detection and 6D Pose Estimation
• ECF-YOLOv7-Tiny: Improving Feature Fusion and the Receptive Field for Lightweight Object Detectors
• Noise-Aware Evaluation of Object Detectors
• Attention-Based Class-Conditioned Alignment for Multi-Source Domain Adaptation of Object Detectors
• 3D OD
  • VADet: Multi-frame LiDAR 3D Object Detection using Variable Aggregation
  • V-MIND: Building Versatile Monocular Indoor 3D Detector with Diverse 2D Annotations
  • DSTR: Dual Scenes Transformer for Cross-Modal Fusion in 3D Object Detection
  • ALPI: Auto-Labeller with Proxy Injection for 3D Object Detection using 2D Labels Only
  • AIC3DOD: Advancing Indoor Class-Incremental 3D Object Detection with Point Transformer Architecture and Room Layout Constraints
  • Reflective Teacher: Semi-Supervised Multimodal 3D Object Detection in Bird's-Eye-View via Uncertainty Measure
• VOD
  • MaskVD: Region Masking for Efficient Video Object Detection
  • Beyond Boxes: Mask-Guided Spatio-Temporal Feature Aggregation for Video Object Detection
  • Disentangling Spatio-Temporal Knowledge for Weakly Supervised Object Detection and Segmentation in Surgical Video
• 伪装目标检测
  • HDPNet: Hourglass Vision Transformer with Dual-Path Feature Pyramid for Camouflaged Object Detection
• 伪装目标发现
  • CIRCOD: Co-Saliency Inspired Referring Camouflaged Object Discovery
• 域泛化目标检测
  • On the Importance of Dual-Space Augmentation for Domain Generalized Object Detection
• 小样本目标检测
  • Elemental Composite Prototypical Network: Few-Shot Object Detection on Outdoor 3D Point Cloud Scenes
  • Cross-Domain Multi-Modal Few-Shot Object Detection via Rich Text
• 零样本目标检测
  • Zero-Shot Detection of Out-of-Context Objects using Foundation Models
• 端到端目标检测
  • RT-DETRv3: Real-Time End-to-End Object Detection with Hierarchical Dense Positive Supervision
• 开发世界目标检测
  • Decoupled PROB: Decoupled Query Initialization Tasks and Objectness-Class Learning for Open World Object Detection
• 裂纹检测
  • CrackStructures and CrackEnsembles: The Power of Multi-View for 2.5D Crack Detection
• 雪检测
  • A Regional-Level Resource-Saving Model for Winter Road Surface Snow Detection in Extreme Weathers
• 目标定位
  • A Realistic Protocol for Evaluation of Weakly Supervised Object Localization
• 视觉伪影检测
  • MVAD: A Multiple Visual Artifact Detector for Video Streaming

9.Super Resolution(超分辨率)

• Boosting Diffusion Guidance via Learning Degradation-Aware Models for Blind Super Resolution
  :star:code
• ENAF: A Multi-Exit Network with an Adaptive Patch Fusion for Large Image Super Resolution
• Partial Filter-Sharing: Improved Parameter-Sharing Method for Single Image Super-Resolution Networks
• Dynamic Attention-Guided Diffusion for Image Super-Resolution
• 场景文本图像超分辨率
  • NCAP: Scene Text Image Super-Resolution with Non-CAtegorical Prior
• 视频超分辨率
  • RefVSR++: Exploiting Reference Inputs for Reference-Based Video Super-Resolution
  • Unifying Low-Resolution and High-Resolution Alignment by Event Cameras for Space-Time Video Super-Resolution

8.Image/Video Retrieval(图像/视频检索)

• 跨域检索
  • UCDR-Adapter: Exploring Adaptation of Pre-Trained Vision-Language Models for Universal Cross-Domain Retrieval
    :star:code
• 图像检索
  • GlobalDoc: A Cross-Modal Vision-Language Framework for Real-World Document Image Retrieval and Classification
  • Semantic Clustering of Image Retrieval Databases used for Visual Localization
  • 开放词汇图像检索
    • FOR: Finetuning for Object Level Open Vocabulary Image Retrieval
  • 组合图像检索
    • Composed Image Retrieval for Training-Free Domain Conversion
      :star:code
    • DualCIR: Enhancing Training-Free Composed Image Retrieval via Dual-Directional Descriptions
    • SCOT: Self-Supervised Contrastive Pretraining For Zero-Shot Compositional Retrieval
• 视频检索
  • ContextIQ: A Multimodal Expert-Based Video Retrieval System for Contextual Advertising
  • 长视频检索
    • A Video is Worth 10000 Words: Training and Benchmarking with Diverse Captions for Better Long Video Retrieval
  • 文本-视频检索
    • TokenBinder: Text-Video Retrieval with One-to-Many Alignment Paradigm
• 信息检索
  • Patchfinder: Leveraging Visual Language Models for Accurate Information Retrieval using Model Uncertainty

7.Image Captioning(图像字幕)

• Reframing Image Difference Captioning with BLIP2IDC and Synthetic Augmentation
• Image-Caption Encoding for Improving Zero-Shot Generalization
• 密集视频字幕
  • Exo2EgoDVC: Dense Video Captioning of Egocentric Procedural Activities using Web Instructional Videos

6.Image/video Compression(图像/视频压缩)

• All-in-One Image Compression and Restoration
• Efficient Progressive Image Compression with Variance-aware Masking

5.Image Classification(图像分类)

• Federated Source-Free Domain Adaptation for Classification: Weighted Cluster Aggregation for Unlabeled Data
• Multi-Task Learning of Classification and Generation for Set-Structured Data
• 图像分类
  • Invariant Shape Representation Learning for Image Classification
  • TLDR: Text Based Last-Layer Retraining for Debiasing Image Classifiers
  • Pixel-Wise Shuffling with Collaborative Sparsity for Melanoma Hyperspectral Image Classification
  • Data Augmentation for Image Classification using Generative AI
  • CEMIL: Contextual Attention Based Efficient Weakly Supervised Approach for Histopathology Image Classification
• 视觉分类
  • Enhancing Visual Classification using Comparative Descriptors
  • Class-Conditioned Transformation for Enhanced Robust Image Classification
    :star:code
  • Temporal Dynamics in Visual Data: Analyzing the Impact of Time on Classification Accuracy
• 小样本分类
  • Flatness Improves Backbone Generalisation in Few-Shot Classification
• 多标签分类
  • Multi-View Factorizing and Disentangling: A Novel Framework for Incomplete Multi-View Multi-Label Classification
• 开集识别
  • Learning Visual-Semantic Hierarchical Attribute Space for Interpretable Open-Set Recognition

4.Image Progress(图像/视频处理)

• 去噪
  • DiffPAD: Denoising Diffusion-based Adversarial Patch Decontamination
  • Focusing on What to Decode and What to Train: SOV Decoding with Specific Target Guided DeNoising and Vision Language Advisor
  • Inverting the Generation Process of Denoising Diffusion Implicit Models: Empirical Evaluation and a Novel Method
  • VISIONARY: Novel Spatial-Spectral Attention Mechanism for Hyperspectral Image Denoising
  • Unsupervised Denoising for Signal-Dependent and Row-Correlated Imaging Noise
  • J-Invariant Volume Shuffle for Self-Supervised Cryo-Electron Tomogram Denoising on Single Noisy Volume
  • SwinIA: Self-Supervised Blind-Spot Image Denoising without Convolutions
  • Hausdorff Distance Matching with Adaptive Query Denoising for Rotated Detection Transformer
  • Design Principles of Multi-Scale J-Invariant Networks for Self-Supervised Image Denoising
• 去模糊
  • Wavelength- and Depth-Aware Deep Image Prior for Blind Hyperspectral Imagery Deblurring with Coarse Depth Guidance
  • F2former: When Fractional Fourier Meets Deep Wiener Deconvolution and Selective Frequency Transformer for Image Deblurring
  • Blind Image Deblurring with FFT-ReLU Sparsity Prior
• 去阴影
  • Shadow Removal Refinement via Material-Consistent Shadow Edges
• 图像恢复
  • Dropout the High-rate Downsampling: A Novel Design Paradigm for UHD Image Restoration
  • Swin-: Gradient-Based Image Restoration from Image Sequences using Video Swin-Transformers
  • Bayesian Optimal Latent Projection for Noisy Image Restoration
  • Denoising Diffusion Models for High-Resolution Microscopy Image Restoration
  • 水下图像恢复
    • Phaseformer: Phase-Based Attention Mechanism for Underwater Image Restoration and Beyond
• 图像修复
  • SEM-Net: Efficient Pixel Modelling for image inpainting with Spatially Enhanced SSM
    :star:code
  • I Dream My Painting: Connecting MLLMs and Diffusion Models via Prompt Generation for Text-Guided Multi-Mask Inpainting
    :star:code
  • A New Benchmark and Baseline for Real-Time High-Resolution Image Inpainting on Edge Devices
  • Improving Detail in Pluralistic Image Inpainting with Feature Dequantization
• 图像增强
  • DiffuCE: Expert-Level CBCT Image Enhancement using a Novel Conditional Denoising Diffusion Model with Latent Alignment
  • 水下图像增强
    • USWformer: Efficient Sparse Wavelet Transformer for Underwater Image Enhancement
    • CE-VAE: Capsule Enhanced Variational AutoEncoderfor Underwater Image Enhancement
  • 低光图像增强
    • Deep Joint Unrolling for Deblurring and Low-Light Image Enhancement
    • VIIS: Visible and Infrared Information Synthesis for Severe Low-light Image Enhancement
      :star:code
• 图像质量评估
  • Dual-Representation Interaction Driven Image Quality Assessment with Restoration Assistance
    :star:code
• 视频恢复
  • Reversing the Damage: A QP-Aware Transformer-Diffusion Approach for 8K Video Restoration under Codec Compression
• 视频修复
  • VipDiff: Towards Coherent and Diverse Video Inpainting via Training-free Denoising Diffusion Models
• 视频增强
  • ReBotNet: Fast Real-Time Video Enhancement
  • UnDIVE: Generalized Underwater Video Enhancement Using Generative Priors
• 视频去模糊
  • Adaptive High-Pass Kernel Prediction for Efficient Video Deblurring
    :star:code
• 图像重建
  • MS-Glance: Non-semantic context vectors and the applications in supervising image reconstruction
    :star:code
  • MS-Glance: Bio-Inspired Non-Semantic Context Vectors and their Applications in Supervising Image Reconstruction
  • Spk2ImgMamba: Spiking Camera Image Reconstruction with Multi-Scale State Space Models
  • Self-Supervised Learning with Spectral Low-Rank Prior for Hyperspectral Image Reconstruction
• 着色
  • eLIR-Net: An Efficient AI Solution for Image Retouching图像润色
  • Closing the Domain Gap in Manga Colorization via Aligned Paired Dataset漫画着色
• Image Dewarping
  • DocMatcher: Document Image Dewarping via Structural and Textual Line Matching

3.Image Segmentation(图像分割)

• Aggregated Attributions for Explanatory Analysis of 3D Segmentation Models3D分割
• GaussianBeV : 3D Gaussian Representation Meets Perception Models for BeV SegmentationBeV分割
• HSDA: High-Frequency Shuffle Data Augmentation for Bird's-Eye-View Map Segmentation鸟瞰图分割
• EfficientCrackNet: A Lightweight Model for Crack Segmentation裂纹分割
• HSDA: High-frequency Shuffle Data Augmentation for Bird's-Eye-View Map Segmentation
  :star:code
• CAMS: Convolution and Attention-Free Mamba-Based Cardiac Image Segmentation
• Image-Level Regression for Uncertainty-Aware Retinal Image Segmentation
• Active Learning for Image Segmentation with Binary User Feedback
• Task Configuration Impacts Annotation Quality and Model Training Performance in Crowdsourced Image Segmentation
• 指代图像分割
  • Vision-Aware Text Features in Referring Image Segmentation: From Object Understanding to Context Understanding
  • Feature Design for Bridging SAM and CLIP toward Referring Image Segmentation
• 部分分割
  • Few-shot Structure-Informed Machinery Part Segmentation with Foundation Models and Graph Neural Networks
    :star:code
  • 3D Part Segmentation via Geometric Aggregation of 2D Visual Features
• 全景分割
  • Language-Guided Instance-Aware Domain-Adaptive Panoptic Segmentation
• 实例分割
  • GMT: Guided Mask Transformer for Leaf Instance Segmentation
  • Robot Instance Segmentation with Few Annotations for Grasping
  • 3D实例分割
    • Lifting by Gaussians: A Simple, Fast and Flexible Method for 3D Instance Segmentation
• 语义分割
  • COSNet: A Novel Semantic Segmentation Network using Enhanced Boundaries in Cluttered Scenes
  • A Conflict-Guided Evidential Multimodal Fusion for Semantic Segmentation
  • Modality-Incremental Learning with Disjoint Relevance Mapping Networks for Image-based Semantic Segmentation
  • Epipolar Attention Field Transformers for Bird's Eye View Semantic Segmentation
  • Active Learning with Context Sampling and One-vs-Rest Entropy for Semantic Segmentation
  • CCASeg: Decoding Multi-Scale Context with Convolutional Cross-Attention for Semantic Segmentation
  • Multi-Scale Grouped Prototypes for Interpretable Semantic Segmentation
  • DASC-SPT: Towards Self-Supervised Panoramic Semantic Segmentation
  • U-MixFormer: UNet-Like Transformer with Mix-Attention for Efficient Semantic Segmentation
  • Class-Agnostic Visio-Temporal Scene Sketch Semantic Segmentation
  • Sigma: Siamese Mamba Network for Multi-Modal Semantic Segmentation
  • 3D语义分割
    • CACE: Sim-to-Real Indoor 3D Semantic Segmentation via Context-Aware Augmentation and Consistency Enforcement
  • 域适应语义分割
    • Distribution Optimization under Gaussian Hypothesis for Domain Adaptive Semantic Segmentation
  • 弱监督语义分割
    • Feature-Level and Spatial-Level Activation Expansion for Weakly-Supervised Semantic Segmentation
    • Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation
  • 半监督语义分割
    • Uncertainty and Energy based Loss Guided Semi-Supervised Semantic Segmentation
  • 小样本语义分割
    • Generative Model-Based Fusion for Improved Few-Shot Semantic Segmentation of Infrared Images
  • 开放词汇语义分割
    • TFM^2: Training-Free Mask Matching for Open-Vocabulary Semantic Segmentation
    • Pay Attention to Your Neighbours: Training-Free Open-Vocabulary Semantic Segmentation
    • AIDE: Improving 3D Open-Vocabulary Semantic Segmentation by Aligned Vision-Language Learning
• 抠图
  • Robust Portrait Image Matting and Depth-of-Field Synthesis via Multiplane Images
• VSS
  • Event-guided Low-light Video Semantic Segmentation
• VPS
  • Balancing Shared and Task-Specific Representations: A Hybrid Approach to Depth-Aware Video Panoptic Segmentation
    :house:project
• VIS
  • Towards Real-Time Open-Vocabulary Video Instance Segmentation
• VOS
  • Efficient Video Object Segmentation via Modulated Cross-Attention Memory
• 伪装分割
  • CamoFA: A Learnable Fourier-Based Augmentation for Camouflage Segmentation

2.Face(人脸)

• Continual Learning of Personalized Generative Face Models with Experience Replay
  :star:code
• Remote Blood Pressure Estimation from Facial Videos using Transfer Learning: Leveraging PPG to rPPG Conversion
• ComFace: Facial Representation Learning with Synthetic Data for Comparing Faces
• BeautyBank: Encoding Facial Makeup in Latent Space
• VerA: Versatile Anonymization Applicable to Clinical Facial Photographs
• LogicNet: A Logical Consistency Embedded Face Attribute Learning Network
• 情绪识别
  • Leveraging CLIP Encoder for Multimodal Emotion Recognition
• 三维人脸
  • FitDiff: Robust Monocular 3D Facial Shape and Reflectance Estimation using Diffusion Models
• 人脸恢复
  • Copy or Not? Reference-Based Face Image Restoration with Fine Details
  • FLAIR: A Conditional Diffusion Framework with Applications to Face Video Restoration
  • Towards Unsupervised Blind Face Restoration using Diffusion Prior
• 人脸重现
  • Anchored Diffusion for Video Face Reenactment
• 人脸识别
  • PETALface: Parameter Efficient Transfer Learning for Low-resolution Face Recognition
    :house:project
  • A Rapid Test for Accuracy and Bias of Face Recognition Technology
  • FALCON: Fair Face Recognition via Local Optimal Feature Normalization
  • CLFace: A Scalable and Resource-Efficient Continual Learning Framework for Lifelong Face Recognition
  • Beyond Spatial Explanations: Explainable Face Recognition in the Frequency Domain
  • Effective Backdoor Learning on Open-Set Face Recognition Systems
• 人脸验证
  • Fairer Analysis and Demographically Balanced Face Generation for Fairer Face Verification
• 人脸重建
  • SplatFace: Gaussian Splat Face Reconstruction Leveraging an Optimizable Surface
• 人脸交换
  • Realistic and Efficient Face Swapping: A Unified Approach with Diffusion Models
  • PixSwap: High-Resolution Face Swapping for Effective Reflection of Identity via Pixel-Level Supervision with Synthetic Paired Dataset
• 人脸生成
  • Analyzing and Improving the Skin Tone Consistency and Bias in Implicit 3D Relightable Face Generators
• 人脸匿名化
  • Face Anonymization Made Simple
• 人脸活体检测
  • Towards Generalized Face Anti-Spoofing from a Frequency Shortcut View
  • Text-to-Image Synthesis for Domain Generalization in Face Anti-Spoofing
• 说话人脸生成
  • Joint Co-Speech Gesture and Expressive Talking Face Generation using Diffusion with Adapters
  • DisFlowEm : One-Shot Emotional Talking Head Generation using Disentangled Pose and Expression Flow-Guidance
  • SyncDiff: Diffusion-Based Talking Head Synthesis with Bottlenecked Temporal Visual Prior for Improved Synchronization
  • EmoVOCA: Speech-Driven Emotional 3D Talking Heads
  • Talking Head Anime 4: Distillation for Real-Time Performance
  • 文本-说话人脸生成
    • FT2TF: First-Person Statement Text-To-Talking Face Generation
• 人脸表情识别
  • Facial Expression Recognition with Controlled Privacy Preservation and Feature Compensation
  • Enhancing Zero-Shot Facial Expression Recognition by LLM Knowledge Transfer
• 人脸关键点检测
  • Cascaded Dual Vision Transformer for Accurate Facial Landmark Detection
  • ORFormer: Occlusion-Robust Transformer for Accurate Facial Landmark Detection

1.Othere(其它)

• Dense Depth from Event Focal Stack
• MAGMA: Manifold Regularization for MAEs
  :star:code
• Advancing Weight and Channel Sparsification with Enhanced Saliency
• Secrets of Edge-Informed Contrast Maximization for Event-Based Vision
• Metric Compatible Training for Online Backfilling in Large-Scale Retrieval
• PULSE: Physiological Understanding with Liquid Signal Extraction
• CryoMAE: Few-Shot Cryo-EM Particle Picking with Masked Autoencoders
• Adaptive and Temporally Consistent Gaussian Surfels for Multi-View Dynamic Reconstruction
• Enriching Local Patterns with Multi-Token Attention for Broad-Sight Neural Networks
• PALO: A Polyglot Large Multimodal Model for 5B People
• Assessing Visually-Continuous Corruption Robustness of Neural Networks Relative to Human Performance
• Cross-Domain and Cross-Dimension Learning for Image-to-Graph Transformers
• SpiralMLP: A Lightweight Vision MLP Architecture
• Mind the Map! Accounting for Existing Maps When Estimating Online HDMaps from Sensors
• Zero-Shot Class Unlearning in CLIP with Synthetic Samples
• Revisiting Deep Archetypal Analysis for Phenotype Discovery in High Content Imaging
• A 0-Shot Self-Attention Mechanism for Accelerated Diagonal Attention
• Long-Term Ad Memorability: Understanding & Generating Memorable Ads
• Crossroads of Continents: Automated Artifact Extraction for Cultural Adaptation with Large Multimodal Models
• A Two-Head Loss Function for Deep Average-K Classification
• Neural Graph Map: Dense Mapping with Efficient Loop Closure Integration
• Learning to Count from Pseudo-Labeled Segmentation
• Recognizing Unseen States of Unknown Objects by Leveraging Knowledge Graphs
• ReC-TTT: Contrastive Feature Reconstruction for Test-Time Training
• Cross-Aligned Fusion for Multimodal Understanding
• GStex: Per-Primitive Texturing of 2D Gaussian Splatting for Decoupled Appearance and Geometry Modeling
• VHS: High-Resolution Iterative Stereo Matching with Visual Hull Priors
• MemFusionMap: Working Memory Fusion for Online Vectorized HD Map Construction
• Learning the Power of "No": Foundation Models with Negations
• Disentangle Source and Target Knowledge for Continual Test-Time Adaptation
• LiLMaps: Learnable Implicit Language Maps
• SpaGBOL: Spatial-Graph-Based Orientated Localisation
• SegBuilder: A Semi-Automatic Annotation Tool for Segmentation标注工具
• DTA: Dual Temporal-Channel-Wise Attention for Spiking Neural Networks
• LumiGauss: Relightable Gaussian Splatting in the Wild
• Improving Shift Invariance in Convolutional Neural Networks with Translation Invariant Polyphase Sampling
• Multi-Spectral Image Color Reproduction图像颜色再现
• Improving Uncertainty Estimation with Confidence-Aware Training Data
• Decomposed Distribution Matching in Dataset Condensation
• Deduce and Select Evidences with Language Models for Training-Free Video Goal Inference
• SAND: Enhancing Open-Set Neuron Descriptions through Spatial Awareness
• Harmonizing Attention: Training-Free Texture-Aware Geometry Transfer
• SmartKC++: Improving Performance of Smartphone-Based Corneal Topographers
• Partial Texture VAE: Color and Texture Encoder for Rock Particle Images
• Feature Space Perturbation: A Panacea to Enhanced Transferability Estimation
• Token Turing Machines are Efficient Vision Models
• MetaVIn: Meteorological and Visual Integration for Atmospheric Turbulence Strength Estimation大气湍流强度估算
• On Neural BRDFs: A Thorough Comparison of State-of-the-Art Approaches
• Learning Keypoints for Multi-Agent Behavior Analysis using Self-Supervision
• DiTAS: Quantizing Diffusion Transformers via Enhanced Activation Smoothing
• SUM: Saliency Unification through Mamba for Visual Attention Modeling
• Shift Equivariant Pose Network
• SV-data2vec: Guiding Video Representation Learning with Latent Skeleton Targets
• An Investigation on LLMs' Visual Understanding Ability using SVG for Image-Text Bridging
• RiemStega: Covariance-Based Loss for Print-Proof Transmission of Data in Images
• Rethinking Low-Rank Adaptation in Vision: Exploring Head-Level Responsiveness Across Diverse Tasks
• VisualFusion: Enhancing Blog Content with Advanced Infographic Pipeline
• From Visual Explanations to Counterfactual Explanations with Latent Diffusion
• Latency Robust Cooperative Perception using Asynchronous Feature Fusion
• Social EgoMesh Estimation
• MENTOR: Human Perception-Guided Pretraining for Increased Generalization
• A Conic Transformation Approach for Solving the Perspective-Three-Point Problem
• OT-VP: Optimal Transport-Guided Visual Prompting for Test-Time Adaptation
• Per-Pixel Solution of Multispectral Photometric Stereo
• PositiveCoOp: Rethinking Prompting Strategies for Multi-Label Recognition with Partial Annotations
• NAT: Learning to Attack Neurons for Enhanced Adversarial Transferability
• VideoGameBunny: Towards Vision Assistants for Video Games
• Diffusion-Based Generative Regularization for Supervised Discriminative Learning
• CLIPArTT: Adaptation of CLIP to New Domains at Test Time
• BIV-Priv-Seg: Locating Private Content in Images Taken by People with Visual Impairments在视障人士拍摄的图像中定位私人内容
• ROSA: Reconstructing Object Shape and Appearance Textures by Adaptive Detail Transfer重建物体形状和外观纹理
• LatteCLIP: Unsupervised CLIP Fine-Tuning via LMM-Synthetic Texts
• Diffusion-Based Particle-DETR for BEV Perception
• DisCo: Discovering Common Affordance from Large Models for Actionable Part Perception
• WARLearn: Weather-Adaptive Representation Learning
• Cross-Modal Feature Alignment and MMD Improve Robustness of Prompt Tuning
• ReinDiffuse: Crafting Physically Plausible Motions with Reinforced Diffusion Model用强化扩散模型制作物理上合理的运动
• Learning Deep Illumination-Robust Features from Multispectral Filter Array Images
• Label Augmented Dataset Distillation
• Towards Privacy-Preserving Split Learning for ControlNet
• Evaluating Sensitivity Consistency of Explanations
• Stable Autofocus with Focal Consistency Loss
• Semantically Conditioned Prompts for Visual Recognition under Missing Modality Scenarios
• RD-DPP: Rate-Distortion Theory Meets Determinantal Point Process to Diversify Learning Data Samples
• Towards Robust Training via Gradient-Diversified Backpropagation
• Benchmarking VLMs' Reasoning About Persuasive Atypical Images
• Non-Cross Diffusion for Semantic Consistency
• Optimizing Neural Network Effectiveness via Non-Monotonicity Refinement
• [FlashMix: Fast Map-Free LiDAR Localization via Feature Mixing and Contrastive-Constrained Accelerated Traini