CVPR 2025 Events with Videos

Keynotes

• Exploring the Low Altitude Airspace: From Natural Resource to Economic Engine
• The Llama Herd of Models: System 1, 2, 3 Go!
• Gemini Robotics, Bringing AI to the Physical World

Posters

• ParaHome: Parameterizing Everyday Home Activities Towards 3D Generative Modeling of Human-Object Interactions
• PrEditor3D: Fast and Precise 3D Shape Editing
• VinaBench: Benchmark for Faithful and Consistent Visual Narratives
• Test-Time Fine-Tuning of Image Compression Models for Multi-Task Adaptability
• LT3SD: Latent Trees for 3D Scene Diffusion
• SynthLight: Portrait Relighting with Diffusion Model by Learning to Re-render Synthetic Faces
• Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis
• DexHandDiff: Interaction-aware Diffusion Planning for Adaptive Dexterous Manipulation
• EZSR: Event-based Zero-Shot Recognition
• HiMoR: Monocular Deformable Gaussian Reconstruction with Hierarchical Motion Representation
• Memories of Forgotten Concepts
• Discrete to Continuous: Generating Smooth Transition Poses from Sign Language Observations
• Camouflage Anything: Learning to Hide using Controlled Out-painting and Representation Engineering
• Synthetic Data is an Elegant GIFT for Continual Vision-Language Models
• Polarized Color Screen Matting
• Improve Representation for Imbalanced Regression through Geometric Constraints
• Differentiable Inverse Rendering with Interpretable Basis BRDFs
• Prior-free 3D Object Tracking
• Balanced Rate-Distortion Optimization in Learned Image Compression
• FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering
• Motion Modes: What Could Happen Next?
• Simplification Is All You Need against Out-of-Distribution Overconfidence
• Query Efficient Black-Box Visual Prompting with Subspace Learning
• Context-Aware Multimodal Pretraining
• ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration
• Binarized Neural Network for Multi-spectral Image Fusion
• Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking
• Attention IoU: Examining Biases in CelebA using Attention Maps
• VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation
• DeepLA-Net: Very Deep Local Aggregation Networks for Point Cloud Analysis
• MambaVO: Deep Visual Odometry Based on Sequential Matching Refinement and Training Smoothing
• KVQ: Boosting Video Quality Assessment via Saliency-guided Local Perception
• SceneFactor: Factored Latent 3D Diffusion for Controllable 3D Scene Generation
• Fortifying Federated Learning Towards Trustworthiness via Auditable Data Valuation and Verifiable Client Contribution
• MAD: Memory-Augmented Detection of 3D Objects
• Seeing More with Less: Human-like Representations in Vision Models
• One Diffusion to Generate Them All
• Electromyography-Informed Facial Expression Reconstruction for Physiological-Based Synthesis and Analysis
• iSegMan: Interactive Segment-and-Manipulate 3D Gaussians
• GASP: Gaussian Avatars with Synthetic Priors
• PreciseCam: Precise Camera Control for Text-to-Image Generation
• HumanMM: Global Human Motion Recovery from Multi-shot Videos
• Quaffure: Real-Time Quasi-Static Neural Hair Simulation
• PEACE: Empowering Geologic Map Holistic Understanding with MLLMs
• FinePhys: Fine-grained Human Action Generation by Explicitly Incorporating Physical Laws for Effective Skeletal Guidance
• Illumination Spectrum Estimation for Multispectral Images via Surface Reflectance Modeling and Spatial-Spectral Feature Generation
• Towards Source-Free Machine Unlearning
• RELOCATE: A Simple Training-Free Baseline for Visual Query Localization Using Region-Based Representations
• TexGaussian: Generating High-quality PBR Material via Octree-based 3D Gaussian Splatting
• SeqAfford: Sequential 3D Affordance Reasoning via Multimodal Large Language Model
• EquiPose: Exploiting Permutation Equivariance for Relative Camera Pose Estimation
• PanDA: Towards Panoramic Depth Anything with Unlabeled Panoramas and Mobius Spatial Augmentation
• BlueLM-V-3B: Algorithm and System Co-Design for Multimodal Large Language Models on Mobile Devices
• CaMuViD: Calibration-Free Multi-View Detection
• ChatGarment: Garment Estimation, Generation and Editing via Large Language Models
• Image Generation Diversity Issues and How to Tame Them
• GFlowVLM: Enhancing Multi-step Reasoning in Vision-Language Models with Generative Flow Networks
• Joint Vision-Language Social Bias Removal for CLIP
• AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios
• LinGen: Towards High-Resolution Minute-Length Text-to-Video Generation with Linear Computational Complexity
• AI-Face: A Million-Scale Demographically Annotated AI-Generated Face Dataset and Fairness Benchmark
• Augmenting Perceptual Super-Resolution via Image Quality Predictors
• FDS: Frequency-Aware Denoising Score for Text-Guided Latent Diffusion Image Editing
• Wav2Sem: Plug-and-Play Audio Semantic Decoupling for 3D Speech-Driven Facial Animation
• FLAME: Frozen Large Language Models Enable Data-Efficient Language-Image Pre-training
• CALICO: Part-Focused Semantic Co-Segmentation with Large Vision-Language Models
• Adaptive Non-Uniform Timestep Sampling for Accelerating Diffusion Model Training
• Neural Hierarchical Decomposition for Single Image Plant Modeling
• MeshArt: Generating Articulated Meshes with Structure-Guided Transformers
• Prometheus: 3D-Aware Latent Diffusion Models for Feed-Forward Text-to-3D Scene Generation
• Decoupling Fine Detail and Global Geometry for Compressed Depth Map Super-Resolution
• NTR-Gaussian: Nighttime Dynamic Thermal Reconstruction with 4D Gaussian Splatting Based on Thermodynamics
• Provoking Multi-modal Few-Shot LVLM via Exploration-Exploitation In-Context Learning
• EnvPoser: Environment-aware Realistic Human Motion Estimation from Sparse Observations with Uncertainty Modeling
• Blurry-Edges: Photon-Limited Depth Estimation from Defocused Boundaries
• HybridGS: Decoupling Transients and Statics with 2D and 3D Gaussian Splatting
• Conditional Balance: Improving Multi-Conditioning Trade-Offs in Image Generation
• Stochastic Human Motion Prediction with Memory of Action Transition and Action Characteristic
• SGC-Net: Stratified Granular Comparison Network for Open-Vocabulary HOI Detection
• VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos
• SimLTD: Simple Supervised and Semi-Supervised Long-Tailed Object Detection
• Classifier-guided CLIP Distillation for Unsupervised Multi-label Classification
• AG-VPReID: A Challenging Large-Scale Benchmark for Aerial-Ground Video-based Person Re-Identification
• STPro: Spatial and Temporal Progressive Learning for Weakly Supervised Spatio-Temporal Grounding
• Towards Satellite Image Road Graph Extraction: A Global-Scale Dataset and A Novel Method
• SAMBLE: Shape-Specific Point Cloud Sampling for an Optimal Trade-Off Between Local Detail and Global Uniformity
• Cross-Modal 3D Representation with Multi-View Images and Point Clouds
• GraphMimic: Graph-to-Graphs Generative Modeling from Videos for Policy Learning
• Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models
• Nonisotropic Gaussian Diffusion for Realistic 3D Human Motion Prediction
• VASparse: Towards Efficient Visual Hallucination Mitigation via Visual-Aware Token Sparsification
• High Temporal Consistency through Semantic Similarity Propagation in Semi-Supervised Video Semantic Segmentation for Autonomous Flight
• Deep Fair Multi-View Clustering with Attention KAN
• Learning Extremely High Density Crowds as Active Matters
• RipVIS: Rip Currents Video Instance Segmentation Benchmark for Beach Monitoring and Safety
• Dual Diffusion for Unified Image Generation and Understanding
• Scaling up Image Segmentation across Data and Tasks
• Curriculum Direct Preference Optimization for Diffusion and Consistency Models
• EnliveningGS: Active Locomotion of 3DGS
• HotSpot: Signed Distance Function Optimization with an Asymptotically Sufficient Condition
• Compass Control: Multi Object Orientation Control for Text-to-Image Generation
• SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration
• Rethinking Temporal Fusion with a Unified Gradient Descent View for 3D Semantic Occupancy Prediction
• Associative Transformer
• Zero-Shot Image Restoration Using Few-Step Guidance of Consistency Models (and Beyond)
• MVPaint: Synchronized Multi-View Diffusion for Painting Anything 3D
• PICO: Reconstructing 3D People In Contact with Objects
• MUSt3R: Multi-view Network for Stereo 3D Reconstruction
• SketchFusion: Learning Universal Sketch Features through Fusing Foundation Models
• Continuous Space-Time Video Resampling with Invertible Motion Steganography
• LineArt: A Knowledge-guided Training-free High-quality Appearance Transfer for Design Drawing with Diffusion Model
• Structure-from-Motion with a Non-Parametric Camera Model
• Towards Improved Text-Aligned Codebook Learning: Multi-Hierarchical Codebook-Text Alignment with Long Text
• Accurate Differential Operators for Hybrid Neural Fields
• InsightEdit: Towards Better Instruction Following for Image Editing
• PromptHMR: Promptable Human Mesh Recovery
• A New Statistical Model of Star Speckles for Learning to Detect and Characterize Exoplanets in Direct Imaging Observations
• SAMWISE: Infusing Wisdom in SAM2 for Text-Driven Video Segmentation
• Towards High-fidelity 3D Talking Avatar with Personalized Dynamic Texture
• VideoGuide: Improving Video Diffusion Models without Training Through a Teacher's Guide
• Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather
• Large Self-Supervised Models Bridge the Gap in Domain Adaptive Object Detection
• Gaussian Splatting Feature Fields for (Privacy-Preserving) Visual Localization
• Hardware-Rasterized Ray-Based Gaussian Splatting
• PIAD: Pose and Illumination agnostic Anomaly Detection
• AeSPa : Attention-guided Self-supervised Parallel Imaging for MRI Reconstruction
• Task-aware Cross-modal Feature Refinement Transformer with Large Language Models for Visual Grounding
• SinGS: Animatable Single-Image Human Gaussian Splats with Kinematic Priors
• SoundVista: Novel-View Ambient Sound Synthesis via Visual-Acoustic Binding
• Neuro-Symbolic Evaluation of Text-to-Video Models using Formal Verification
• Efficient Personalization of Quantized Diffusion Model without Backpropagation
• Code-as-Monitor: Constraint-aware Visual Programming for Reactive and Proactive Robotic Failure Detection
• AdaCM^2: On Understanding Extremely Long-Term Video with Adaptive Cross-Modality Memory Reduction
• Multitwine: Multi-Object Compositing with Text and Layout Control
• ProtoDepth: Unsupervised Continual Depth Completion with Prototypes
• Robust Multimodal Survival Prediction with Conditional Latent Differentiation Variational AutoEncoder
• Monocular and Generalizable Gaussian Talking Head Animation
• Enhancing Facial Privacy Protection via Weakening Diffusion Purification
• MonSter: Marry Monodepth to Stereo Unleashes Power
• Open-World Amodal Appearance Completion
• VERA: Explainable Video Anomaly Detection via Verbalized Learning of Vision-Language Models
• HOTFormerLoc: Hierarchical Octree Transformer for Versatile Lidar Place Recognition Across Ground and Aerial Views
• Enhancing Privacy-Utility Trade-offs to Mitigate Memorization in Diffusion Models
• Stable Flow: Vital Layers for Training-Free Image Editing
• VGGT: Visual Geometry Grounded Transformer
• Token Cropr: Faster ViTs for Quite a Few Tasks
• AffordDP: Generalizable Diffusion Policy with Transferable Affordance
• Taming Teacher Forcing for Masked Autoregressive Video Generation
• Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior
• FreqDebias: Towards Generalizable Deepfake Detection via Consistency-Driven Frequency Debiasing
• ArcPro: Architectural Programs for Structured 3D Abstraction of Sparse Points
• PatchDEMUX: A Certifiably Robust Framework for Multi-label Classifiers Against Adversarial Patches
• Multirate Neural Image Compression with Adaptive Lattice Vector Quantization
• EffiDec3D: An Optimized Decoder for High-Performance and Efficient 3D Medical Image Segmentation
• Explainable Saliency: Articulating Reasoning with Contextual Prioritization
• MambaVLT: Time-Evolving Multimodal State Space Model for Vision-Language Tracking
• ERUPT: Efficient Rendering with Unposed Patch Transformer
• SAIST: Segment Any Infrared Small Target Model Guided by Contrastive Language-Image Pretraining
• HOT3D: Hand and Object Tracking in 3D from Egocentric Multi-View Videos
• ScribbleLight: Single Image Indoor Relighting with Scribbles
• HyperGS: Hyperspectral 3D Gaussian Splatting
• From Laboratory to Real World: A New Benchmark Towards Privacy-Preserved Visible-Infrared Person Re-Identification
• Multi-party Collaborative Attention Control for Image Customization
• CrossOver: 3D Scene Cross-Modal Alignment
• Charm: The Missing Piece in ViT Fine-Tuning for Image Aesthetic Assessment
• RASP: Revisiting 3D Anamorphic Art for Shadow-Guided Packing of Irregular Objects
• From Head to Tail: Towards Balanced Representation in Large Vision-Language Models through Adaptive Data Calibration
• 3D-GSW: 3D Gaussian Splatting for Robust Watermarking
• Layered Image Vectorization via Semantic Simplification
• Feat2GS: Probing Visual Foundation Models with Gaussian Splatting
• Doppelgängers and Adversarial Vulnerability
• Gaussian Splatting for Efficient Satellite Image Photogrammetry
• MeshGen: Generating PBR Textured Mesh with Render-Enhanced Auto-Encoder and Generative Data Augmentation
• CholecTrack20: A Multi-Perspective Tracking Dataset for Surgical Tools
• EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild
• Dual Exposure Stereo for Extended Dynamic Range 3D Imaging
• Hypergraph Vision Transformers: Images are More than Nodes, More than Edges
• FG^2: Fine-Grained Cross-View Localization by Fine-Grained Feature Matching
• The Language of Motion: Unifying Verbal and Non-verbal Language of 3D Human Motion
• MVGenMaster: Scaling Multi-View Generation from Any Image via 3D Priors Enhanced Diffusion Model
• LoTUS: Large-Scale Machine Unlearning with a Taste of Uncertainty
• MonoPlace3D: Learning 3D-Aware Object Placement for 3D Monocular Detection
• ConMo: Controllable Motion Disentanglement and Recomposition for Zero-Shot Motion Transfer
• Lifting the Veil on Visual Information Flow in MLLMs: Unlocking Pathways to Faster Inference
• PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting
• Driving by the Rules: A Benchmark for Integrating Traffic Sign Regulations into Vectorized HD Map
• PartGen: Part-level 3D Generation and Reconstruction with Multi-view Diffusion Models
• Resilient Sensor Fusion Under Adverse Sensor Failures via Multi-Modal Expert Fusion
• One-Step Event-Driven High-Speed Autofocus
• Kiss3DGen: Repurposing Image Diffusion Models for 3D Asset Generation
• VISTREAM: Improving Computation Efficiency of Visual Streaming Perception via Law-of-Charge-Conservation Inspired Spiking Neural Network
• Context-Enhanced Memory-Refined Transformer for Online Action Detection
• Pathways on the Image Manifold: Image Editing via Video Generation
• ESCAPE: Equivariant Shape Completion via Anchor Point Encoding
• OSLoPrompt: Bridging Low-Supervision Challenges and Open-Set Domain Generalization in CLIP
• Efficient Depth Estimation for Unstable Stereo Camera Systems on AR Glasses
• Benchmarking Object Detectors under Real-World Distribution Shifts in Satellite Imagery
• JiSAM: Alleviate Labeling Burden and Corner Case Problems in Autonomous Driving via Minimal Real-World Data
• Revisiting Audio-Visual Segmentation with Vision-Centric Transformer
• Unboxed: Geometrically and Temporally Consistent Video Outpainting
• Binarized Mamba-Transformer for Lightweight Quad Bayer HybridEVS Demosaicing
• Gazing Into Missteps: Leveraging Eye-Gaze for Unsupervised Mistake Detection in Egocentric Videos of Skilled Human Activities
• Generative Photomontage
• ANNEXE: Unified Analyzing, Answering, and Pixel Grounding for Egocentric Interaction
• Solving Instance Detection from an Open-World Perspective
• Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networks
• HuMoCon: Concept Discovery for Human Motion Understanding
• Minding Fuzzy Regions: A Data-driven Alternating Learning Paradigm for Stable Lesion Segmentation
• Gradient Inversion Attacks on Parameter-Efficient Fine-Tuning
• Variance-Based Membership Inference Attacks Against Large-Scale Image Captioning Models
• CLIP Under the Microscope: A Fine-Grained Analysis of Multi-Object Representation
• PSA-SSL: Pose and Size-aware Self-Supervised Learning on LiDAR Point Clouds
• MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM
• MAtCha Gaussians: Atlas of Charts for High-Quality Geometry and Photorealism From Sparse Views
• Rethinking Spiking Self-Attention Mechanism: Implementing α-XNOR Similarity Calculation in Spiking Transformers
• Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering
• Disentangling Safe and Unsafe Image Corruptions via Anisotropy and Locality
• HVI: A New Color Space for Low-light Image Enhancement
• KeyFace: Expressive Audio-Driven Facial Animation for Long Sequences via KeyFrame Interpolation
• BioX-CPath: Biologically-driven Explainable Diagnostics for Multistain IHC Computational Pathology
• Towards Explainable and Unprecedented Accuracy in Matching Challenging Finger Crease Patterns
• Generating Multimodal Driving Scenes via Next-Scene Prediction
• 4Deform: Neural Surface Deformation for Robust Shape Interpolation
• A Tale of Two Classes: Adapting Supervised Contrastive Learning to Binary Imbalanced Datasets
• HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video Understanding
• Rethinking Noisy Video-Text Retrieval via Relation-aware Alignment
• UNEM: UNrolled Generalized EM for Transductive Few-Shot Learning
• Lost in Translation, Found in Context: Sign Language Translation with Contextual Cues
• LOCORE: Image Re-ranking with Long-Context Sequence Modeling
• Self-Expansion of Pre-trained Models with Mixture of Adapters for Continual Learning
• Multi-subject Open-set Personalization in Video Generation
• Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset
• Geometry Field Splatting with Gaussian Surfels
• RoboPEPP: Vision-Based Robot Pose and Joint Angle Estimation through Embedding Predictive Pre-Training
• RobSense: A Robust Multi-modal Foundation Model for Remote Sensing with Static, Temporal, and Incomplete Data Adaptability
• SATA: Spatial Autocorrelation Token Analysis for Enhancing the Robustness of Vision Transformers
• Video Depth without Video Models
• UMotion: Uncertainty-driven Human Motion Estimation from Inertial and Ultra-wideband Units
• Dissecting and Mitigating Diffusion Bias via Mechanistic Interpretability
• Do Your Best and Get Enough Rest for Continual Learning
• Mind the Gap: Detecting Black-box Adversarial Attacks in the Making through Query Update Analysis
• HOIGPT: Learning Long-Sequence Hand-Object Interaction with Language Models
• EDM: Equirectangular Projection-Oriented Dense Kernelized Feature Matching
• Descriptor-In-Pixel : Point-Feature Tracking For Pixel Processor Arrays
• VerbDiff: Text-Only Diffusion Models with Enhanced Interaction Awareness
• T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation
• Towards Cost-Effective Learning: A Synergy of Semi-Supervised and Active Learning
• Advancing Manga Analysis: Comprehensive Segmentation Annotations for the Manga109 Dataset
• Reducing Class-wise Confusion for Incremental Learning with Disentangled Manifolds
• Vision-Language Model IP Protection via Prompt-based Learning
• Decompositional Neural Scene Reconstruction with Generative Diffusion Prior
• LLaVA-ST: A Multimodal Large Language Model for Fine-Grained Spatial-Temporal Understanding
• How Do I Do That? Synthesizing 3D Hand Motion and Contacts for Everyday Interactions
• Temporal Alignment-Free Video Matching for Few-shot Action Recognition
• Graph Neural Network Combining Event Stream and Periodic Aggregation for Low-Latency Event-based Vision
• Exploiting Deblurring Networks for Radiance Fields
• EBS-EKF: Accurate and High Frequency Event-based Star Tracking
• GAF: Gaussian Avatar Reconstruction from Monocular Videos via Multi-view Diffusion
• CH3Depth: Efficient and Flexible Depth Foundation Model with Flow Matching
• Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space
• RigGS: Rigging of 3D Gaussians for Modeling Articulated Objects in Videos
• OSMamba: Omnidirectional Spectral Mamba with Dual-Domain Prior Generator for Exposure Correction
• ForestLPR: LiDAR Place Recognition in Forests Attentioning Multiple BEV Density Images
• Distraction is All You Need for Multimodal Large Language Model Jailbreaking
• Generative Gaussian Splatting for Unbounded 3D City Generation
• GarmentPile: Point-Level Visual Affordance Guided Retrieval and Adaptation for Cluttered Garments Manipulation
• Seeing the Abstract: Translating the Abstract Language for Vision Language Models
• Tiled Diffusion
• Dual-Agent Optimization framework for Cross-Domain Few-Shot Segmentation
• PACT: Pruning and Clustering-Based Token Reduction for Faster Visual Language Models
• BiomedCoOp: Learning to Prompt for Biomedical Vision-Language Models
• UnCommon Objects in 3D
• EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting
• CaricatureBooth: Data-Free Interactive Caricature Generation in a Photo Booth
• SemanticDraw: Towards Real-Time Interactive Content Creation from Image Diffusion Models
• Robotic Visual Instruction
• An Image-like Diffusion Method for Human-Object Interaction Detection
• PolarFree: Polarization-based Reflection-Free Imaging
• Harnessing Global-Local Collaborative Adversarial Perturbation for Anti-Customization
• Generative Omnimatte: Learning to Decompose Video into Layers
• Recovering Dynamic 3D Sketches from Videos
• Towards Long-Horizon Vision-Language Navigation: Platform, Benchmark and Method
• Pixel-aligned RGB-NIR Stereo Imaging and Dataset for Robot Vision
• FSBench: A Figure Skating Benchmark for Advancing Artistic Sports Understanding
• Hyperbolic Uncertainty-Aware Few-Shot Incremental Point Cloud Segmentation
• Physical Plausibility-aware Trajectory Prediction via Locomotion Embodiment
• FedCS: Coreset Selection for Federated Learning
• EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialogues
• TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Splatting
• Automatic Joint Structured Pruning and Quantization for Efficient Neural Network Training and Compression
• Joint Optimization of Neural Radiance Fields and Continuous Camera Motion from a Monocular Video
• Make-It-Animatable: An Efficient Framework for Authoring Animation-Ready 3D Characters
• Efficient Test-time Adaptive Object Detection via Sensitivity-Guided Pruning
• Conformal Prediction and MLLM aided Uncertainty Quantification in Scene Graph Generation
• Complementary Advantages: Exploiting Cross-Field Frequency Correlation for NIR-Assisted Image Denoising
• R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization
• T-CIL: Temperature Scaling using Adversarial Perturbation for Calibration in Class-Incremental Learning
• DSPNet: Dual-vision Scene Perception for Robust 3D Question Answering
• PerLA: Perceptive 3D Language Assistant
• Compositional Caching for Training-free Open-vocabulary Attribute Detection
• PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splatting
• ClearSight: Visual Signal Enhancement for Object Hallucination Mitigation in Multimodal Large Language Models
• SCSA: A Plug-and-Play Semantic Continuous-Sparse Attention for Arbitrary Semantic Style Transfer
• GENMANIP: LLM-driven Simulation for Generalizable Instruction-Following Manipulation
• VinTAGe: Joint Video and Text Conditioning for Holistic Audio Generation
• AvatarArtist: Open-Domain 4D Avatarization
• Few-shot Personalized Scanpath Prediction
• SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Images
• Masked Point-Entity Contrast for Open-Vocabulary 3D Scene Understanding
• Towards Universal Dataset Distillation via Task-Driven Diffusion
• Efficient Event-Based Object Detection: A Hybrid Neural Network with Spatial and Temporal Attention
• SimLingo: Vision-Only Closed-Loop Autonomous Driving with Language-Action Alignment
• WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments
• Spotting the Unexpected (STU): A 3D LiDAR Dataset for Anomaly Segmentation in Autonomous Driving
• Spectral Informed Mamba for Robust Point Cloud Processing
• Co-op: Correspondence-based Novel Object Pose Estimation
• RePerformer: Immersive Human-centric Volumetric Videos from Playback to Photoreal Reperformance
• Enhancing 3D Gaze Estimation in the Wild using Weak Supervision with Gaze Following Labels
• Seeing is Not Believing: Adversarial Natural Object Optimization for Hard-Label 3D Scene Attacks
• Full-DoF Egomotion Estimation for Event Cameras Using Geometric Solvers
• Dynamic Camera Poses and Where to Find Them
• LoRACLR: Contrastive Adaptation for Customization of Diffusion Models
• No Pains, More Gains: Recycling Sub-Salient Patches for Efficient High-Resolution Image Recognition
• Feature4X: Bridging Any Monocular Video to 4D Agentic AI with Versatile Gaussian Feature Fields
• Arc2Avatar: Generating Expressive 3D Avatars from a Single Image via ID Guidance
• EVolSplat: Efficient Volume-based Gaussian Splatting for Urban View Synthesis
• PoseBH: Prototypical Multi-Dataset Training Beyond Human Pose Estimation
• QuCOOP: A Versatile Framework for Solving Composite and Binary-Parametrised Problems on Quantum Annealers
• Noise-Consistent Siamese-Diffusion for Medical Image Synthesis and Segmentation
• Classifier-Free Guidance Inside the Attraction Basin May Cause Memorization
• Overcoming Shortcut Problem in VLM for Robust Out-of-Distribution Detection
• TacoDepth: Towards Efficient Radar-Camera Depth Estimation with One-stage Fusion
• SOGS: Second-Order Anchor for Advanced 3D Gaussian Splatting
• Volumetrically Consistent 3D Gaussian Rasterization
• Mitigating Hallucinations in Large Vision-Language Models via DPO: On-Policy Data Hold the Key
• Masked Scene Modeling: Narrowing the Gap Between Supervised and Self-Supervised Learning in 3D Scene Understanding
• Coherent 3D Portrait Video Reconstruction via Triplane Fusion
• DiffLocks: Generating 3D Hair from a Single Image using Diffusion Models
• Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration
• Seeing A 3D World in A Grain of Sand
• MarkushGrapher: Joint Visual and Textual Recognition of Markush Structures
• CRISP: Object Pose and Shape Estimation with Test-Time Adaptation
• Vision-Guided Action: Enhancing 3D Human Motion Prediction with Gaze-informed Affordance in 3D Scenes
• HumanDreamer: Generating Controllable Human-Motion Videos via Decoupled Generation
• UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection
• InsTaG: Learning Personalized 3D Talking Head from Few-Second Video
• ArtiFade: Learning to Generate High-quality Subject from Blemished Images
• DTOS: Dynamic Time Object Sensing with Large Multimodal Model
• Geometry-guided Online 3D Video Synthesis with Multi-View Temporal Consistency
• Question-Aware Gaussian Experts for Audio-Visual Question Answering
• Efficient Decoupled Feature 3D Gaussian Splatting via Hierarchical Compression
• FALCON: Fairness Learning via Contrastive Attention Approach to Continual Semantic Scene Understanding
• MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 3D Gaussian Splatting
• Focusing on Tracks for Online Multi-Object Tracking
• DIFFER: Disentangling Identity Features via Semantic Cues for Clothes-Changing Person Re-ID
• FruitNinja: 3D Object Interior Texture Generation with Gaussian Splatting
• Is `Right' Right? Enhancing Object Orientation Understanding in Multimodal Large Language Models through Egocentric Instruction Tuning
• DyMO: Training-Free Diffusion Model Alignment with Dynamic Multi-Objective Scheduling
• CASP: Consistency-aware Audio-induced Saliency Prediction Model for Omnidirectional Video
• Parallelized Autoregressive Visual Generation
• CAP-Net: A Unified Network for 6D Pose and Size Estimation of Categorical Articulated Parts from a Single RGB-D Image
• Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation
• STiL: Semi-supervised Tabular-Image Learning for Comprehensive Task-Relevant Information Exploration in Multimodal Classification
• MammAlps: A Multi-view Video Behavior Monitoring Dataset of Wild Mammals in the Swiss Alps
• Decoupled Motion Expression Video Segmentation
• MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation
• FineLIP: Extending CLIP’s Reach via Fine-Grained Alignment with Longer Text Inputs
• DecoupledGaussian: Object-Scene Decoupling for Physics-Based Interaction
• Identifying and Mitigating Position Bias of Multi-image Vision-Language Models
• Thin-Shell-SfT: Fine-Grained Monocular Non-rigid 3D Surface Tracking with Neural Deformation Fields
• GCC: Generative Color Constancy via Diffusing a Color Checker
• Semantic-guided Cross-Modal Prompt Learning for Skeleton-based Zero-shot Action Recognition
• DeCLIP: Decoupled Learning for Open-Vocabulary Dense Perception
• A Unified Framework for Heterogeneous Semi-supervised Learning
• ViKIENet: Towards Efficient 3D Object Detection with Virtual Key Instance Enhanced Network
• FluxSpace: Disentangled Semantic Editing in Rectified Flow Models
• Image Reconstruction from Readout-Multiplexed Single-Photon Detector Arrays
• MEAT: Multiview Diffusion Model for Human Generation on Megapixels with Mesh Attention
• Generative Map Priors for Collaborative BEV Semantic Segmentation
• Dynamic Stereotype Theory Induced Micro-expression Recognition with Oriented Deformation
• Gazing at Rewards: Eye Movements as a Lens into Human and AI Decision-Making in Hybrid Visual Foraging
• InterMimic: Towards Universal Whole-Body Control for Physics-Based Human-Object Interactions
• MPDrive: Improving Spatial Understanding with Marker-Based Prompt Learning for Autonomous Driving
• Explaining in Diffusion: Explaining a Classifier with Diffusion Semantics
• DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation
• ClimbingCap: Multi-Modal Dataset and Method for Rock Climbing in World Coordinate
• Synthetic Prior for Few-Shot Drivable Head Avatar Inversion
• Olympus: A Universal Task Router for Computer Vision Tasks
• Identity-Preserving Text-to-Video Generation by Frequency Decomposition
• ILIAS: Instance-Level Image retrieval At Scale
• VI^3NR: Variance Informed Initialization for Implicit Neural Representations
• PhysicsGen: Can Generative Models Learn from Images to Predict Complex Physical Relations?
• NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction
• HyperPose: Hypernetwork-Infused Camera Pose Localization and an Extended Cambridge Landmarks Dataset
• HyperLoRA: Parameter-Efficient Adaptive Generation for Portrait Synthesis
• FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors
• Parametric Point Cloud Completion for Polygonal Surface Reconstruction
• Classifier-to-Bias: Toward Unsupervised Automatic Bias Detection for Visual Classifiers
• FSboard: Over 3 Million Characters of ASL Fingerspelling Collected via Smartphones
• Insightful Instance Features for 3D Instance Segmentation
• WF-VAE: Enhancing Video VAE by Wavelet-Driven Energy Flow for Latent Video Diffusion Model
• A Bias-Free Training Paradigm for More General AI-generated Image Detection
• R2C: Mapping Room to Chessboard to Unlock LLM As Low-Level Action Planner
• Apply Hierarchical-Chain-of-Generation to Complex Attributes Text-to-3D Generation
• Cheb-GR: Rethinking K-nearest Neighbor Search in Re-ranking for Person Re-identification
• Multi-Modal Contrastive Masked Autoencoders: A Two-Stage Progressive Pre-training Approach for RGBD Datasets
• MagicArticulate: Make Your 3D Models Articulation-Ready
• MDP: Multidimensional Vision Model Pruning with Latency Constraint
• Dora: Sampling and Benchmarking for 3D Shape Variational Auto-Encoders
• Mask^2DiT: Dual Mask-based Diffusion Transformer for Multi-Scene Long Video Generation
• PromptHash: Affinity-Prompted Collaborative Cross-Modal Learning for Adaptive Hashing Retrieval
• MExD: An Expert-Infused Diffusion Model for Whole-Slide Image Classification
• ObjectMover: Generative Object Movement with Video Prior
• Circumventing Shortcuts in Audio-visual Deepfake Detection Datasets with Unsupervised Learning
• AnyCam: Learning to Recover Camera Poses and Intrinsics from Casual Videos
• HandOS: 3D Hand Reconstruction in One Stage
• Do Computer Vision Foundation Models Learn the Low-level Characteristics of the Human Visual System?
• VoteFlow: Enforcing Local Rigidity in Self-Supervised Scene Flow
• NoiseCtrl: A Sampling-Algorithm-Agnostic Conditional Generation Method for Diffusion Models
• SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images
• Auto Cherry-Picker: Learning from High-quality Generative Data Driven by Language
• Mesh Mamba: A Unified State Space Model for Saliency Prediction in Non-Textured and Textured Meshes
• VELOCITI: Benchmarking Video-Language Compositional Reasoning with Strict Entailment
• VideoHandles: Editing 3D Object Compositions in Videos Using Video Generative Priors
• MicroVQA: A Multimodal Reasoning Benchmark for Microscopy-Based Scientific Research
• UltraFusion: Ultra High Dynamic Imaging using Exposure Fusion
• Retrieving Semantics from the Deep: an RAG Solution for Gesture Synthesis
• Hybrid Concept Bottleneck Models
• Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models
• PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomalies
• Auto-Encoded Supervision for Perceptual Image Super-Resolution
• Dynamic Content Prediction with Motion-aware Priors for Blind Face Video Restoration
• FoundHand: Large-Scale Domain-Specific Learning for Controllable Hand Image Generation
• STING-BEE: Towards Vision-Language Model for Real-World X-ray Baggage Security Inspection
• BADGR: Bundle Adjustment Diffusion Conditioned by Gradients for Wide-Baseline Floor Plan Reconstruction
• Diffusion-based Realistic Listening Head Generation via Hybrid Motion Modeling
• Less is More: Efficient Image Vectorization with Adaptive Parameterization
• PS-Diffusion: Photorealistic Subject-Driven Image Editing with Disentangled Control and Attention
• Toward Real-world BEV Perception: Depth Uncertainty Estimation via Gaussian Splatting
• Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization
• GauSTAR: Gaussian Surface Tracking and Reconstruction
• ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos
• Mamba-Adaptor: State Space Model Adaptor for Visual Recognition
• Spatial Transport Optimization by Repositioning Attention Map for Training-Free Text-to-Image Synthesis
• Saliuitl: Ensemble Salience Guided Recovery of Adversarial Patches against CNNs
• Paint by Inpaint: Learning to Add Image Objects by Removing Them First
• Plug-and-Play Versatile Compressed Video Enhancement
• 3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation
• Task Singular Vectors: Reducing Task Interference in Model Merging
• Crab: A Unified Audio-Visual Scene Understanding Model with Explicit Cooperation
• Localized Concept Erasure for Text-to-Image Diffusion Models Using Training-Free Gated Low-Rank Adaptation
• Generative Video Propagation
• PERSE: Personalized 3D Generative Avatars from A Single Portrait
• Any-Resolution AI-Generated Image Detection by Spectral Learning
• SAT-HMR: Real-Time Multi-Person 3D Mesh Estimation via Scale-Adaptive Tokens
• Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Models
• Leveraging SD Map to Augment HD Map-based Trajectory Prediction
• Re-HOLD: Video Hand Object Interaction Reenactment via adaptive Layout-instructed Diffusion Model
• Sim-to-Real Causal Transfer: A Metric Learning Approach to Causally-Aware Interaction Representations
• LongDiff: Training-Free Long Video Generation in One Go
• SMTPD: A New Benchmark for Temporal Prediction of Social Media Popularity
• Relative Pose Estimation through Affine Corrections of Monocular Depth Priors
• Articulated Kinematics Distillation from Video Diffusion Models
• SKE-Layout: Spatial Knowledge Enhanced Layout Generation with LLMs
• ONDA-Pose: Occlusion-Aware Neural Domain Adaptation for Self-Supervised 6D Object Pose Estimation
• RoadSocial: A Diverse VideoQA Dataset and Benchmark for Road Event Understanding from Social Video Narratives
• vesselFM: A Foundation Model for Universal 3D Blood Vessel Segmentation
• VisionZip: Longer is Better but Not Necessary in Vision Language Models
• ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping
• Video-ColBERT: Contextualized Late Interaction for Text-to-Video Retrieval
• ActiveGAMER: Active GAussian Mapping through Efficient Rendering
• Scenario Dreamer: Vectorized Latent Diffusion for Generating Driving Simulation Environments
• COBRA: COmBinatorial Retrieval Augmentation for Few-Shot Adaptation
• Consistent Normal Orientation for 3D Point Clouds via Least Squares on Delaunay Graph
• Gaussian Eigen Models for Human Heads
• Sketchtopia: A Dataset and Foundational Agents for Benchmarking Asynchronous Multimodal Communication with Iconic Feedback
• ImViD: Immersive Volumetric Videos for Enhanced VR Engagement
• UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Prior
• Dense Dispersed Structured Light for Hyperspectral 3D Imaging of Dynamic Scenes
• Sea-ing in Low-light
• MM-OR: A Large Multimodal Operating Room Dataset for Semantic Understanding of High-Intensity Surgical Environments
• O-TPT: Orthogonality Constraints for Calibrating Test-time Prompt Tuning in Vision-Language Models
• Self-Supervised Cross-View Correspondence with Predictive Cycle Consistency
• Towards Scalable Human-aligned Benchmark for Text-guided Image Editing
• Chapter-Llama: Efficient Chaptering in Hour-Long Videos with LLMs
• GigaHands: A Massive Annotated Dataset of Bimanual Hand Activities
• Believing is Seeing: Unobserved Object Detection using Generative Models
• SkillMimic: Learning Basketball Interaction Skills from Demonstrations
• Video-Bench: Human-Aligned Video Generation Benchmark
• SF3D: Stable Fast 3D Mesh Reconstruction with UV-unwrapping and Illumination Disentanglement
• VideoGLaMM : A Large Multimodal Model for Pixel-Level Visual Grounding in Videos
• SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videos
• DepthCues: Evaluating Monocular Depth Perception in Large Vision Models
• How to Merge Your Multimodal Models Over Time?
• Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detection
• One-shot 3D Object Canonicalization based on Geometric and Semantic Consistency
• A Flag Decomposition for Hierarchical Datasets
• Occlusion-aware Text-Image-Point Cloud Pretraining for Open-World 3D Object Recognition
• Zero-1-to-A: Zero-Shot One Image to Animatable Head Avatars Using Video Diffusion
• ProHOC: Probabilistic Hierarchical Out-of-Distribution Classification via Multi-Depth Networks
• Creating Your Editable 3D Photorealistic Avatar with Tetrahedron-constrained Gaussian Splatting
• LongVALE: Vision-Audio-Language-Event Benchmark Towards Time-Aware Omni-Modal Perception of Long Videos
• 3DEnhancer: Consistent Multi-View Diffusion for 3D Enhancement
• Detecting Open World Objects via Partial Attribute Assignment
• VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling
• Structure from Collision
• All Languages Matter: Evaluating LMMs on Culturally Diverse 100 Languages
• Free360: Layered Gaussian Splatting for Unbounded 360-Degree View Synthesis from Extremely Sparse and Unposed Views
• Scaling Down Text Encoders of Text-to-Image Diffusion Models
• Through-The-Mask: Mask-based Motion Trajectories for Image-to-Video Generation
• Shape Abstraction via Marching Differentiable Support Functions
• AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360° Unbounded Scene Inpainting
• Visual Agentic AI for Spatial Reasoning with a Dynamic API
• Learning Occlusion-Robust Vision Transformers for Real-Time UAV Tracking
• Optimizing for the Shortest Path in Denoising Diffusion Model
• Attention Distillation: A Unified Approach to Visual Characteristics Transfer
• RainyGS: Efficient Rain Synthesis with Physically-Based Gaussian Splatting
• GPVK-VL: Geometry-Preserving Virtual Keyframes for Visual Localization under Large Viewpoint Changes
• GuardSplat: Efficient and Robust Watermarking for 3D Gaussian Splatting
• Gen3DEval: Using vLLMs for Automatic Evaluation of Generated 3D Objects
• VideoScene: Distilling Video Diffusion Model to Generate 3D Scenes in One Step
• Exploring Contextual Attribute Density in Referring Expression Counting
• BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature
• BIGS: Bimanual Category-agnostic Interaction Reconstruction from Monocular Videos via 3D Gaussian Splatting
• Unleashing In-context Learning of Autoregressive Models for Few-shot Image Manipulation
• ZeroVO: Visual Odometry with Minimal Assumptions
• GazeGene: Large-scale Synthetic Gaze Dataset with 3D Eyeball Annotations
• BG-Triangle: Bézier Gaussian Triangle for 3D Vectorization and Rendering
• Generalized Few-shot 3D Point Cloud Segmentation with Vision-Language Model
• Video Summarization with Large Language Models
• Towards Human-Understandable Multi-Dimensional Concept Discovery
• Pos3R: 6D Pose Estimation for Unseen Objects Made Easy
• WildAvatar: Learning In-the-wild 3D Avatars from the Web
• SemGeoMo: Dynamic Contextual Human Motion Generation with Semantic and Geometric Guidance
• LC-Mamba: Local and Continuous Mamba with Shifted Windows for Frame Interpolation
• Focus-N-Fix: Region-Aware Fine-Tuning for Text-to-Image Generation
• SSHNet: Unsupervised Cross-modal Homography Estimation via Problem Reformulation and Split Optimization
• Cross-modal Information Flow in Multimodal Large Language Models
• CDI: Copyrighted Data Identification in Diffusion Models
• SyncSDE: A Probabilistic Framework for Diffusion Synchronization
• SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation
• Towards Precise Scaling Laws for Video Diffusion Transformers
• MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors
• FreeCloth: Free-form Generation Enhances Challenging Clothed Human Modeling
• The Impact Label Noise and Choice of Threshold has on Cross-Entropy and Soft-Dice in Image Segmentation
• GRAPHGPT-O: Synergistic Multimodal Comprehension and Generation on Graphs
• MetaWriter: Personalized Handwritten Text Recognition Using Meta-Learned Prompt Tuning
• A Semantic Knowledge Complementarity based Decoupling Framework for Semi-supervised Class-imbalanced Medical Image Segmentation
• MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion
• Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects
• DiverseFlow: Sample-Efficient Diverse Mode Coverage in Flows
• HD-EPIC: A Highly-Detailed Egocentric Video Dataset
• Inversion Circle Interpolation: Diffusion-based Image Augmentation for Data-scarce Classification
• SpectroMotion: Dynamic 3D Reconstruction of Specular Scenes
• LoRA Recycle: Unlocking Tuning-Free Few-Shot Adaptability in Visual Foundation Models by Recycling Pre-Tuned LoRAs
• HuPerFlow: A Comprehensive Benchmark for Human vs. Machine Motion Estimation Comparison
• MIMO: Controllable Character Video Synthesis with Spatial Decomposed Modeling
• Volumetric Surfaces: Representing Fuzzy Geometries with Layered Meshes
• DTGBrepGen: A Novel B-rep Generative Model through Decoupling Topology and Geometry
• InteractVLM: 3D Interaction Reasoning from 2D Foundational Models
• Finsler Multi-Dimensional Scaling: Manifold Learning for Asymmetric Dimensionality Reduction and Embedding
• Not Only Text: Exploring Compositionality of Visual Representations in Vision-Language Models
• Unsupervised Discovery of Facial Landmarks and Head Pose
• Decoupling Training-Free Guided Diffusion by ADMM
• Joint Out-of-Distribution Filtering and Data Discovery Active Learning
• High-fidelity 3D Object Generation from Single Image with RGBN-Volume Gaussian Reconstruction Model
• HOIGen-1M: A Large-scale Dataset for Human-Object Interaction Video Generation
• Preserve or Modify? Context-Aware Evaluation for Balancing Preservation and Modification in Text-Guided Image Editing
• IM-Portrait: Learning 3D-aware Video Diffusion for Photorealistic Talking Heads from Monocular VideosC
• Unveiling the Mist over 3D Vision-Language Understanding: Object-centric Evaluation with Chain-of-Analysis
• Directional Label Diffusion Model for Learning from Noisy Labels
• Improving Transferable Targeted Attacks with Feature Tuning Mixup
• Attribute-Missing Multi-view Graph Clustering
• A Unified Model for Compressed Sensing MRI Across Undersampling Patterns
• DynaMoDe-NeRF: Motion-aware Deblurring Neural Radiance Field for Dynamic Scenes
• It’s a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data
• GeoAvatar: Geometrically-Consistent Multi-Person Avatar Reconstruction from Sparse Multi-View Videos
• Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images
• Dynamic Neural Surfaces for Elastic 4D Shape Representation and Analysis
• RICCARDO: Radar Hit Prediction and Convolution for Camera-Radar 3D Object Detection
• GliaNet: Adaptive Neural Network Structure Learning with Glia-Driven
• GenAssets: Generating in-the-wild 3D Assets in Latent Space
• HomoGen: Enhanced Video Inpainting via Homography Propagation and Diffusion
• RelationField: Relate Anything in Radiance Fields
• Time of the Flight of the Gaussians: Optimizing Depth Indirectly in Dynamic Radiance Fields
• 4D LangSplat: 4D Language Gaussian Splatting via Multimodal Large Language Models
• Bringing CLIP to the Clinic: Dynamic Soft Labels and Negation-Aware Learning for Medical Analysis
• FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Views
• HybridMQA: Exploring Geometry-Texture Interactions for Colored Mesh Quality Assessment
• Multi-Resolution Pathology-Language Pre-training Model with Text-Guided Visual Representation
• ShiftwiseConv: Small Convolutional Kernel with Large Kernel Effect
• SUM Parts: Benchmarking Part-Level Semantic Segmentation of Urban Meshes
• Perceptually Accurate 3D Talking Head Generation: New Definitions, Speech-Mesh Representation, and Evaluation Metrics
• Parameterized Blur Kernel Prior Learning for Local Motion Deblurring
• Dynamic Motion Blending for Versatile Motion Editing
• StableAnimator: High-Quality Identity-Preserving Human Image Animation
• Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives
• Category-Agnostic Neural Object Rigging
• Mitigating the Human-Robot Domain Discrepancy in Visual Pre-training for Robotic Manipulation
• AnyDressing: Customizable Multi-Garment Virtual Dressing via Latent Diffusion Models
• Floxels: Fast Unsupervised Voxel Based Scene Flow Estimation
• Understanding Multi-layered Transmission Matrices
• Attraction Diminishing and Distributing for Few-Shot Class-Incremental Learning
• Protecting Your Video Content: Disrupting Automated Video-based LLM Annotations
• GEM: A Generalizable Ego-Vision Multimodal World Model for Fine-Grained Ego-Motion, Object Dynamics, and Scene Composition Control
• Meta-Learning Hyperparameters for Parameter Efficient Fine-Tuning
• Boosting the Dual-Stream Architecture in Ultra-High Resolution Segmentation with Resolution-Biased Uncertainty Estimation
• DOF-GS: Adjustable Depth-of-Field 3D Gaussian Splatting for Post-Capture Refocusing, Defocus Rendering and Blur Removal
• HyperNVD: Accelerating Neural Video Decomposition via Hypernetworks
• Good, Cheap, and Fast: Overfitted Image Compression with Wasserstein Distortion
• Spectral State Space Model for Rotation-Invariant Visual Representation Learning
• Towards Enhanced Image Inpainting: Mitigating Unwanted Object Insertion and Preserving Color Consistency
• MonoInstance: Enhancing Monocular Priors via Multi-view Instance Alignment for Neural Rendering and Reconstruction
• DINOv2 Meets Text: A Unified Framework for Image- and Pixel-Level Vision-Language Alignment
• Advancing Generalizable Tumor Segmentation with Anomaly-Aware Open-Vocabulary Attention Maps and Frozen Foundation Diffusion Models
• Can Text-to-Video Generation help Video-Language Alignment?
• Scene-Centric Unsupervised Panoptic Segmentation
• GraphI2P: Image-to-Point Cloud Registration with Exploring Pattern of Correspondence via Graph Learning
• 3D Student Splatting and Scooping
• Continuous Locomotive Crowd Behavior Generation
• DexGrasp Anything: Towards Universal Robotic Dexterous Grasping with Physics Awareness
• 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes
• Data Distributional Properties As Inductive Bias for Systematic Generalization
• Beyond Sight: Towards Cognitive Alignment in LVLM via Enriched Visual Knowledge
• Closest Neighbors are Harmful for Lightweight Masked Auto-encoders
• Exploiting Temporal State Space Sharing for Video Semantic Segmentation
• PRaDA: Projective Radial Distortion Averaging
• A Simple Data Augmentation for Feature Distribution Skewed Federated Learning
• Functionality Understanding and Segmentation in 3D Scenes
• Potential Field Based Deep Metric Learning
• Arbitrary-steps Image Super-resolution via Diffusion Inversion
• Improving Adversarial Transferability on Vision Transformers via Forward Propagation Refinement
• Ego4o: Egocentric Human Motion Capture and Understanding from Multi-Modal Input
• BLADE: Single-view Body Mesh Estimation through Accurate Depth Estimation
• Text Augmented Correlation Transformer For Few-shot Classification & Segmentation
• Prof. Robot: Differentiable Robot Rendering Without Static and Self-Collisions
• Enhancing Vision-Language Compositional Understanding with Multimodal Synthetic Data
• SketchVideo: Sketch-based Video Generation and Editing
• F-LMM: Grounding Frozen Large Multimodal Models
• Pursuing Temporal-Consistent Video Virtual Try-On via Dynamic Pose Interaction
• TopoCellGen: Generating Histopathology Cell Topology with a Diffusion Model
• Unified Uncertainty-Aware Diffusion for Multi-Agent Trajectory Modeling
• Embodied Scene Understanding for Vision Language Models via MetaVQA
• Robust 3D Shape Reconstruction in Zero-Shot from a Single Image in the Wild
• Cross-Modal Distillation for 2D/3D Multi-Object Discovery from 2D Motion
• Cross-modal Causal Relation Alignment for Video Question Grounding
• Omni-Scene: Omni-Gaussian Representation for Ego-Centric Sparse-View Scene Reconstruction
• RoboGround: Robotic Manipulation with Grounded Vision-Language Priors
• SwiftEdit: Lightning Fast Text-Guided Image Editing via One-Step Diffusion
• Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis
• Learning from Synchronization: Self-Supervised Uncalibrated Multi-View Person Association in Challenging Scenes
• UniAP: Unifying Inter- and Intra-Layer Automatic Parallelism by Mixed Integer Quadratic Programming
• PGC: Physics-Based Gaussian Cloth from a Single Pose
• Learning Temporally Consistent Video Depth from Video Diffusion Priors
• Low-Rank Adaptation in Multilinear Operator Networks for Security-Preserving Incremental Learning
• Change3D: Revisiting Change Detection and Captioning from A Video Modeling Perspective
• Hyperdimensional Uncertainty Quantification for Multimodal Uncertainty Fusion in Autonomous Vehicles Perception
• Language-Assisted Debiasing and Smoothing for Foundation Model-Based Semi-Supervised Learning
• Let Humanoids Hike! Integrative Skill Development on Complex Trails
• Towards More General Video-based Deepfake Detection through Facial Component Guided Adaptation for Foundation Model
• PersonaBooth: Personalized Text-to-Motion Generation
• DropGaussian: Structural Regularization for Sparse-view Gaussian Splatting
• Using Diffusion Priors for Video Amodal Segmentation
• Consistency Posterior Sampling for Diverse Image Synthesis
• RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins
• Satellite Observations Guided Diffusion Model for Accurate Meteorological States at Arbitrary Resolution
• HRAvatar: High-Quality and Relightable Gaussian Head Avatar
• DeepCompress-ViT: Rethinking Model Compression to Enhance Efficiency of Vision Transformers at the Edge
• DNF: Unconditional 4D Generation with Dictionary-based Neural Fields
• VidBot: Learning Generalizable 3D Actions from In-the-Wild 2D Human Videos for Zero-Shot Robotic Manipulation
• APT: Adaptive Personalized Training for Diffusion Models with Limited Data
• FlipSketch: Flipping Static Drawings to Text-Guided Sketch Animations
• TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing
• NexusGS: Sparse View Synthesis with Epipolar Depth Priors in 3D Gaussian Splatting
• 4DTAM: Non-Rigid Tracking and Mapping via Dynamic Surface Gaussians
• Material Anything: Generating Materials for Any 3D Object via Diffusion
• From Sparse to Dense: Camera Relocalization with Scene-Specific Detector from Feature Gaussian Splatting
• SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model
• Towards Realistic Example-based Modeling via 3D Gaussian Stitching
• MotionPRO: Exploring the Role of Pressure in Human MoCap and Beyond
• LesionLocator: Zero-Shot Universal Tumor Segmentation and Tracking in 3D Whole-Body Imaging
• BIMBA: Selective-Scan Compression for Long-Range Video Question Answering
• DV-Matcher: Deformation-based Non-rigid Point Cloud Matching Guided by Pre-trained Visual Features
• DyCON: Dynamic Uncertainty-aware Consistency and Contrastive Learning for Semi-supervised Medical Image Segmentation
• MP-GUI: Modality Perception with MLLMs for GUI Understanding
• Bias for Action: Video Implicit Neural Representations with Bias Modulation
• Homogeneous Dynamics Space for Heterogeneous Humans
• Doppelgangers++: Improved Visual Disambiguation with Geometric 3D Features
• Neural LightRig: Unlocking Accurate Object Normal and Material Estimation with Multi-Light Diffusion
• h-Edit: Effective and Flexible Diffusion-Based Editing via Doob's h-Transform
• Generalized Recorrupted-to-Recorrupted: Self-Supervised Learning Beyond Gaussian Noise
• Adversarial Diffusion Compression for Real-World Image Super-Resolution
• Probability Density Geodesics in Image Diffusion Latent Space
• RNG: Relightable Neural Gaussians
• Instruct-CLIP: Improving Instruction-Guided Image Editing with Automated Data Refinement Using Contrastive Learning
• Temporal Action Detection Model Compression by Progressive Block Drop
• HistoFS: Non-IID Histopathologic Whole Slide Image Classification via Federated Style Transfer with RoI-Preserving
• GCE-Pose: Global Context Enhancement for Category-level Object Pose Estimation
• Hybrid Global-Local Representation with Augmented Spatial Guidance for Zero-Shot Referring Image Segmentation
• StageDesigner: Artistic Stage Generation for Scenography via Theater Scripts
• Vision-Language Models Do Not Understand Negation
• PCDreamer: Point Cloud Completion Through Multi-view Diffusion Priors
• V2V3D: View-to-View Denoised 3D Reconstruction for Light Field Microscopy
• Birth and Death of a Rose
• DoF-Gaussian: Controllable Depth-of-Field for 3D Gaussian Splatting
• Depth Any Camera: Zero-Shot Metric Depth Estimation from Any Camera
• Latent Space Imaging
• Unlocking the Potential of Unlabeled Data in Semi-Supervised Domain Generalization
• 3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion
• Rethinking Few-Shot Adaptation of Vision-Language Models in Two Stages
• Open Set Label Shift with Test Time Out-of-Distribution Reference
• BimArt: A Unified Approach for the Synthesis of 3D Bimanual Interaction with Articulated Objects
• GLASS: Guided Latent Slot Diffusion for Object-Centric Learning
• Prompt2Perturb (P2P): Text-Guided Diffusion-Based Adversarial Attack on Breast Ultrasound Images
• PICD: Versatile Perceptual Image Compression with Diffusion Rendering
• Task-Specific Gradient Adaptation for Few-Shot One-Class Classification
• Rethinking Decoder Design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention
• HalLoc: Token-level Localization of Hallucinations for Vision Language Models
• HyperGLM: HyperGraph for Video Scene Graph Generation and Anticipation
• DiffPortrait360: Consistent Portrait Diffusion for 360 View Synthesis
• ArticulatedGS: Self-supervised Digital Twin Modeling of Articulated Objects using 3D Gaussian Splatting
• Subnet-Aware Dynamic Supernet Training for Neural Architecture Search
• LiveCC: Learning Video LLM with Streaming Speech Transcription at Scale
• Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy
• Distilling Long-tailed Datasets
• Unsupervised Foundation Model-Agnostic Slide-Level Representation Learning
• SplineGS: Robust Motion-Adaptive Spline for Real-Time Dynamic 3D Gaussians from Monocular Video
• CL-LoRA: Continual Low-Rank Adaptation for Rehearsal-Free Class-Incremental Learning
• Disco4D: Disentangled 4D Human Generation and Animation from a Single Image
• HoGS: Unified Near and Far Object Reconstruction via Homogeneous Gaussian Splatting
• Exploring the Deep Fusion of Large Language Models and Diffusion Transformers for Text-to-Image Synthesis
• Enhancing Creative Generation on Stable Diffusion-based Models
• DIO: Decomposable Implicit 4D Occupancy-Flow World Model
• 3D Gaussian Inpainting with Depth-Guided Cross-View Consistency
• GLane3D: Detecting Lanes with Graph of 3D Keypoints
• Deep Change Monitoring: A Hyperbolic Representative Learning Framework and a Dataset for Long-term Fine-grained Tree Change Detection
• Learning Endogenous Attention for Incremental Object Detection
• CoMapGS: Covisibility Map-based Gaussian Splatting for Sparse Novel View Synthesis
• Blurred LiDAR for Sharper 3D: Robust Handheld 3D Scanning with Diffuse LiDAR and RGB
• Real-time High-fidelity Gaussian Human Avatars with Position-based Interpolation of Spatially Distributed MLPs
• Precise, Fast, and Low-cost Concept Erasure in Value Space: Orthogonal Complement Matters
• ETAP: Event-based Tracking of Any Point
• Segment Anything, Even Occluded
• ReSpec: Relevance and Specificity Grounded Online Filtering for Learning on Video-Text Data Streams
• Which Viewpoint Shows it Best? Language for Weakly Supervising View Selection in Multi-view Instructional Videos
• 3D-SLNR: A Super Lightweight Neural Representation for Large-scale 3D Mapping
• DAMM-Diffusion: Learning Divergence-Aware Multi-Modal Diffusion Model for Nanoparticles Distribution Prediction
• Dyn-HaMR: Recovering 4D Interacting Hand Motion from a Dynamic Camera
• Harnessing Frozen Unimodal Encoders for Flexible Multimodal Alignment
• SPARS3R: Semantic Prior Alignment and Regularization for Sparse 3D Reconstruction
• Automated Generation of Challenging Multiple-Choice Questions for Vision Language Model Evaluation
• Interactive Medical Image Analysis with Concept-based Similarity Reasoning
• Exploration-Driven Generative Interactive Environments
• PQPP: A Joint Benchmark for Text-to-Image Prompt and Query Performance Prediction
• GenVDM: Generating Vector Displacement Maps From a Single Image
• MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis
• Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch
• FactCheXcker: Mitigating Measurement Hallucinations in Chest X-ray Report Generation Models
• DiffSensei: Bridging Multi-Modal LLMs and Diffusion Models for Customized Manga Generation
• Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining
• BASKET: A Large-Scale Video Dataset for Fine-Grained Skill Estimation
• Scene-agnostic Pose Regression for Visual Localization
• BF-STVSR: B-Splines and Fourier---Best Friends for High Fidelity Spatial-Temporal Video Super-Resolution
• Geometry in Style: 3D Stylization via Surface Normal Deformation
• Nearly Zero-Cost Protection Against Mimicry by Personalized Diffusion Models
• SAR3D: Autoregressive 3D Object Generation and Understanding via Multi-scale 3D VQVAE
• Octopus: Alleviating Hallucination via Dynamic Contrastive Decoding
• 3DGUT: Enabling Distorted Cameras and Secondary Rays in Gaussian Splatting
• ROLL: Robust Noisy Pseudo-label Learning for Multi-View Clustering with Noisy Correspondence
• Learning Physics-Based Full-Body Human Reaching and Grasping from Brief Walking References

Tutorials

• Foundations of Interpretable AI
• The 2nd Point Cloud Tutorial: All You Need To Know About 3D Point Cloud
• Scalable Generative Models in Computer Vision
• From Video Generation to World Model
• Volumetric Video in the Real World
• Cognitive AI for the Future: Agentic Multimodal Models and RAG for Vision Language Applications, from Training to Deployment
• Evaluating Large Multi-modal Models: Challenges and Methods
• Multi-Modal Computer Vision and Foundation Models In Agriculture in conjunction with IEEE CVPR 2025
• Robotics 101: An Odyssey from A Vision Perspective
• Animal re-identification
• Computer Vision over Homomorphically Encrypted Data
• Continuous Data Cycle via Foundation Models
• Edge AI in Action: Technologies and Applications
• Identifying Structure in Data: All you need to know about Dimensionality Reduction, Clustering and more
• Multimodal Mathematical Reasoning: Frontiers in Integrating Vision, Language, and Symbolic Representations
• Full-Stack, GPU-based Acceleration of Deep Learning and Foundation Models
• Power-efficient neural networks using low-precision data types and quantization
• Intelligent Healthcare based on Cameras and Wireless Sensors
• Recent Advances in Vision Foundation Models

Report issues here.