International Conference on Computer Vision is a massive conference. In 2025 alone, 11,239 papers were submitted, and 2,699 were accepted. resulting in an acceptance rate of 24%. Oral Presentation — Top 0.56% (64/11239) of all submissions. I created this repository to help you search for crème de la crème of ICCV publications. If the paper you are looking for is not on my short list, take a peek at the full list of accepted papers. Some of the papers’ authors have not posted their work on arXiv or made a GitHub repository available, so the list here may be incomplete or contain errors. We are updating it, and your patience is greatly appreciated. If you want a complete version, please refer to the official repositories Oral list.
| topic | title | repository / paper |
|---|---|---|
| 3D from multi-view and sensors | Multi-View 3D Point Tracking |   |
| 3D from multi-view and sensors | SceneSplat: Gaussian Splatting-based Scene Understanding with Vision-Language Pretraining |   |
| 3D from multi-view and sensors | WIR3D: Visually-Informed and Geometry-Aware 3D Shape Abstraction |   |
| 3D from multi-view and sensors | SparseFlex: High-Resolution and Arbitrary-Topology 3D Shape Modeling |   |
| 3D from multi-view and sensors | Uncalibrated Structure from Motion on a Sphere |  |
| Geometric Computer Vision | Forecasting Continuous Non-Conservative Dynamical Systems in SO(3) |   |
| Geometric Computer Vision | Diving into the Fusion of Monocular Priors for Generalized Stereo Matching |   |
| Geometric Computer Vision | RePoseD: Efficient Relative Pose Estimation With Known Depth Information |   |
| Geometric Computer Vision | Deterministic Object Pose Confidence Region Estimation |  |
| 3D Human Pose Estimation | DPoser-X: Diffusion Model as Robust 3D Whole-body Human Pose Prior |   |
| 3D head avatar modeling | HairCUP: Hair Compositional Universal Prior for 3D Gaussian Avatars |  |
| 3D head avatar modeling | FixTalk: Taming Identity Leakage for High-Quality Talking Head Generation in Extreme Cases |   |
| 3D Scene Understanding | Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion |   |
| 3D Scene Understanding | SuperDec: 3D Scene Decomposition with Superquadric Primitives |   |
| 3D Scene Understanding | ForestFormer3D: A Unified Framework for End-to-End Segmentation of Forest LiDAR s3D Point Clouds |   |
| 3D Scene Understanding | GT-Loc: Unifying When and Where in Images Through a Joint Embedding Space |   |
| NeRF | Self-Ensembling Gaussian Splatting for Few-Shot Novel View Synthesis |   |
| NeRF | RayZer: A Self-supervised Large View Synthesis Model |   |
| Gaussian Splatting | "EVER: Exact Volumetric Ellipsoid Rendering for Real-time View Synthesis" |   |
| Embodied AI | Learning Streaming Video Representation via Multitask Training |   |
| Generative AI | TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models |   |
| Generative AI | MaskControl: Spatio-Temporal Control for Masked Motion Synthesis |   |
| Generative AI | Dynamic Typography: Bringing Text to Life via Video Diffusion Prior |   |
| Generative AI | Generating Physically Stable and Buildable Brick Structures from Text |   |
| Generative AI | LaRender: Training-Free Occlusion Control in Image Generation via Latent Rendering |   |
| Generative AI | ReCamMaster: Camera-Controlled Generative Rendering from A Single Video |   |
| Generative AI | Diffusion Image Prior |  |
| Generative AI | LayerTracer: Cognitive-Aligned Layered SVG Synthesis via Diffusion Transformer |   |
| Generative AI | LOTS of Fashion! Multi-Conditioning for Image Generation via Sketch-Text Pairing |   |
| Generative AI | MikuDance: Animating Character Art with Mixed Motion Dynamics |   |
| Segmentation | CorrCLIP: Reconstructing Patch Correlations in CLIP for Open-Vocabulary Semantic Segmentation |   |
| Segmentation | Easy3D: A Simple Yet Effective Method for 3D Interactive Segmentation |  |
| Segmentation | E-SAM: Training-Free Segment Every Entity Model |  |
| Robotics | Back on Track: Bundle Adjustment for Dynamic Scene Reconstruction |  |
| Robotics | Image as an IMU: Estimating Camera Motion from a Single Motion-Blurred Image |   |
| Robotics | Certifiably Optimal Anisotropic Rotation Averaging |   |
| Robotics | Moto: Latent Motion Token as the Bridging Language for Learning Robot Manipulation from Videos |   |
| Vision Foundation Models | LoftUp: Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models |   |
| Vision Foundation Models | Towards a Unified Copernicus Foundation Model for Earth Vision |   |
| Vision Foundation Models | RS-vHeat: Heat Conduction Guided Efficient Remote Sensing Foundation Model |  |
| Vision Language Models | Online Reasoning Video Segmentation with Just-in-Time Digital Twins |  |
| Multimodal LLMs | Token Activation Map to Visually Explain Multimodal LLMs |   |
| Multimodal LLMs | Scaling Laws for Native Multimodal Models |  |
| Multimodal Learning | Differential-informed Sample Selection Accelerates Multimodal Contrastive Learning |   |
| Multimodal Learning | Understanding Co-speech Gestures in-the-wild |   |
| Multimodal Learning | Differentiable Room Acoustic Rendering with Multi-View Vision Priors |  |
| Object Detection | Counting Stacked Objects | |
| Object Detection | Automated Model Evaluation for Object Detection via Prediction Consistency and Reliability | |
| Image Editing | FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models |   |
| Video Restoration & Enhancement | MIORe & VAR-MIORe: Benchmarks to Push the Boundaries of Restoration |   |
| Trustworthy AI | Soft Local Completeness: Rethinking Completeness in Model Explainability |   |
| Image Retrieval | Learning Visual Hierarchies in Hyperbolic Space for Image Retrieval |  |
| Model Acceleration | Variance-Based Pruning for Accelerating and Compressing Trained Networks |  |
| Model Acceleration | Importance-Based Token Merging for Efficient Image and Video Generation |  |
We would love your help in making this repository even better! If you know of an amazing paper that isn't listed here, or if you have any suggestions for improvement, feel free to open an issue or submit a pull request.