CVPR 2026

The IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) is the premier annual computer vision event comprising the main conference and several co-located workshops and short courses.   With its high quality and low cost, it provides an exceptional value for students, academics and industry researchers.

重要信息

CCF推荐：A（人工智能）

录用率：22.1%（2878/13008，2025年）

时间地点：2026年6月3日-丹佛·美国

截稿时间：2025年11月6日

Call for Papers

3D from multi-view and sensors  

3D from single images  

Adversarial attack and defense

Autonomous driving

Biometrics  

Computational imaging  

Computer vision for social good  

Computer vision theory  

Datasets and evaluation  

Deep learning architectures and techniques  

Document analysis and understanding  

Efficient and scalable vision  

Embodied vision: Active agents, simulation

Event-based cameras

Explainable computer vision  

Humans: Face, body, pose, gesture, movement

Image and video synthesis and generation  

Low-level vision  

Machine learning (other than deep learning)

Medical and biological vision, cell microscopy

Multimodal learning

Optimization methods (other than deep learning)

Photogrammetry and remote sensing  

Physics-based vision and shape-from-X  

Recognition: Categorization, detection, retrieval  

Representation learning  

Computer Vision for Robotics  

Scene analysis and understanding  

Segmentation, grouping and shape analysis  

Self-, semi-, meta- and unsupervised learning

Transfer/ low-shot/ continual/ long-tail learning  

Transparency, fairness, accountability, privacy and ethics in vision  

Video: Action and event understanding

Video: Low-level analysis, motion, and tracking  

Vision + graphics  

Vision, language, and reasoning  

Vision applications and systems

Publication and Patent Disclosure

All accepted papers will be publicly released by the Computer Vision Foundation (CVF) two weeks before the conference. Authors considering patent applications should note that the paper's official public disclosure date is whichever occurs first:

·Two weeks before the conference, or

·When authors make the paper publicly available themselves