Search papers from top venues in computer vision (CV)
Search 48,741 papers on recognition, detection, segmentation, 3D reconstruction, video understanding and image generation from CVPR, ECCV and ICCV, 2000 to 2026. Describe what you are looking for in your own words, or search with exact keywords. refract.science is free and runs privately in your browser.
Computer vision venues
3 conferences in refract.science. Each has a page with its papers per year and its latest papers.
- CVPR IEEE/CVF Conference on Computer Vision and Pattern Recognition
- ECCV European Conference on Computer Vision
- ICCV IEEE/CVF International Conference on Computer Vision
- Papers
- 48,741
- Venues
- 3
- Years
- 2000–2026
| Year | Papers |
|---|---|
| 2000 | 345 |
| 2001 | 492 |
| 2002 | 226 |
| 2003 | 406 |
| 2004 | 446 |
| 2005 | 606 |
| 2006 | 515 |
| 2007 | 927 |
| 2008 | 750 |
| 2009 | 690 |
| 2010 | 789 |
| 2011 | 778 |
| 2012 | 874 |
| 2013 | 925 |
| 2014 | 902 |
| 2015 | 1128 |
| 2016 | 1058 |
| 2017 | 1404 |
| 2018 | 1755 |
| 2019 | 2369 |
| 2020 | 2824 |
| 2021 | 3272 |
| 2022 | 3719 |
| 2023 | 4509 |
| 2024 | 5103 |
| 2025 | 5572 |
| 2026 | 6357 |
Example searches
Semantic search understands questions and descriptions. Try one of these on the computer vision venues:
- novel view synthesis from a few images
- open-vocabulary object detection
- self-supervised pretraining for visual representations
- diffusion models for image editing
For exact terms, authors and exhaustive lists, use keyword search, e.g. author:"…" AND venue:cvpr.
Latest computer vision papers
-
WildDepth: A Multimodal Dataset for 3D Wildlife Perception and Depth Estimation
-
GhostPoint: Self-Supervised Representation Learning by Hallucinating Occluded LiDAR Structure
-
TopoSlide: Topologically-Informed Histopathology Whole Slide Image Representation Learning
-
Audio-Visual Camera Pose Estimation with Passive Scene Sounds and In-the-Wild Video
-
GeoFlow: Real-Time Fine-Grained Cross-View Geolocalization via Iterative Flow Prediction
-
EoS-FM: Can an Ensemble of Specialist Models Act as a Generalist Feature Extractor?
-
PlaceIt3D: Language-Guided Object Placement in Real 3D Scenes
-
Punching Bag vs. Punching Person: Motion Transferability in Videos
-
Visual Interestingness Decoded: How GPT-4o Mirrors Human Interests
-
NormalLoc: Visual Localization on Textureless 3D Models using Surface Normals
Questions
Which computer vision venues can I search?
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR, 2000–2026), European Conference on Computer Vision (ECCV, 2000–2026) and IEEE/CVF International Conference on Computer Vision (ICCV, 2001–2025). Together they hold 48,741 papers with titles, authors and, for most papers, abstracts.
How do I find computer vision papers on a topic?
Describe the topic in your own words, as a question, a short description or a few keywords, for example "novel view synthesis from a few images". Semantic search ranks the papers of CVPR, ECCV and ICCV by how well their title and abstract match the meaning of your description, so you also find papers that use different terms.
Can I search a single computer vision venue?
Yes. Filter the results to the venues you want, or use keyword search with a venue field, e.g. venue:cvpr AND title:"…". Each venue also has its own page with its papers per year and latest papers.
Can I search computer vision together with other fields?
Yes. refract.science covers 36 venues across computer science, including machine learning, natural language processing, artificial intelligence and information retrieval, web and data mining. Search them all at once to find related work across fields, or open similar papers from any result.
Can my AI agent search computer vision papers?
Yes. Agents that support the Model Context Protocol (MCP), such as Claude, Cursor or VS Code, can search the computer vision venues installed in your browser, follow similar papers and add them to your collections.
Is it free, and are my searches private?
It is free and needs no account. The papers and the search model are downloaded to your browser once, and every search runs there, so your queries stay on your device.