Video Generative Models Benchmarks

More workshops at Conference on Computer Vision and Pattern Recognition 2026

• 3D Multi-View Supervision Workshop — Explores learning-based approaches to 3D scene reconstruction and understanding using multi-view image supervision.
• 3D Scene Understanding for Vision and Graphics — Covers semantic 3D scene reconstruction, object detection, and layout estimation for robotics and graphics applications....
• 4D Vision Workshop — Focuses on modeling dynamic 3D worlds over time, covering video understanding, point cloud sequences, and...
• Affective Behavior Analysis in-the-wild — Addresses facial expression, action unit, and emotion recognition from real-world unconstrained video and images.
• Adversarial Machine Learning on Computer Vision — Examines adversarial attacks and defenses in computer vision systems, with a focus on the safety...
• Agriculture-Vision Workshop — Advances aerial and ground-based vision for precision agriculture, crop monitoring, and yield estimation.
• Artificial Intelligence for Space — Applies vision and AI to space exploration tasks including terrain mapping, spacecraft navigation, and satellite...
• AI-Generated Media and Security — Addresses detection of deepfakes, synthetic media forensics, and the security implications of generative vision models....
• Authenticity and Provenance in Generative AI — Explores content authentication, watermarking, and provenance tracking for AI-generated images and videos.
• Computational Cameras and Displays — Bridges optics and computation for next-generation imaging and display systems, from HDR capture to neural...
• Computer Vision for Animal Behavior — Applies computer vision to animal tracking, behavior recognition, and ecological monitoring in the wild.
• Computer Vision in the Wild — Advances vision models that generalize beyond curated benchmarks to open-world, real-world, and multimodal settings.
• Computer Vision for Microscopy Image Analysis — Bridges computer vision and biomedical imaging for analysis of microscopy, pathology slides, and biological imagery....
• Computer Vision in Sports — Applies computer vision to sports analysis including player tracking, action recognition, and performance analytics.
• Embodied Artificial Intelligence Workshop — Advances embodied agents that perceive, reason, and act in simulated and real environments through vision...
• EarthVision: CV for Remote Sensing — Applies computer vision to satellite and aerial imagery for earth observation, environmental monitoring, and geospatial...
• Efficient Deep Learning for Computer Vision — Focuses on model compression, quantization, pruning, and hardware-efficient inference for real-world vision deployment.
• Efficient On-Device Generation — Covers efficient generative models for on-device inference, including mobile diffusion and compressed vision transformers.
• Egocentric Vision Workshop — Focuses on first-person video understanding, activity recognition, and interaction detection from wearable cameras.
• Embodied Reasoning in Action — Focuses on robotic manipulation through visual reasoning, with benchmark challenges for perception-to-action pipelines.
• Fine-Grained Visual Categorization — Addresses recognition of subtle visual differences across species, products, and object subtypes with large-scale benchmarks....
• Foundation Models for Medical Vision — Explores large pre-trained models adapted for medical imaging tasks including segmentation, detection, and report generation....
• Eye and Gaze in Computer Vision — Advances gaze estimation, eye tracking, and gaze-based interaction for AR/VR, accessibility, and human-computer interaction.
• Generative Models for Computer Vision — Covers generative image and video models including diffusion, GANs, and flow-based methods applied to vision...
• Human Motion Generation Workshop — Covers generation and synthesis of realistic human body motion for animation, avatars, and embodied agents....
• Image Matching Workshop — Covers local feature detection, descriptor matching, and 3D reconstruction for image correspondence tasks.
• Long-form Video Understanding — Advances understanding of long-duration videos covering temporal reasoning, summarization, and action recognition.
• Mobile AI Workshop — Benchmarks and advances AI efficiency on mobile devices including image processing, super-resolution, and neural architecture...
• Multimodal Algorithmic Reasoning — Explores structured reasoning in vision-language models, including mathematical, spatial, and logical problem solving from visual...
• Multimodal Foundation Models Workshop — Explores the frontier of multimodal large models combining vision, language, audio, and action.
• Foundation Models for Remote Sensing — Investigates large vision models for remote sensing applications including change detection, segmentation, and scene classification....
• Multimodal Spatial Intelligence — Advances spatial understanding in multimodal systems, connecting vision, language, and 3D scene reasoning.
• NTIRE Image Restoration Workshop — Hosts challenges and research on image super-resolution, denoising, deblurring, and low-light enhancement.
• Open-World 3D Scene Understanding — Challenges open-vocabulary 3D scene understanding using foundation models for indoor and outdoor environments.
• Precognition Workshop — Focuses on anticipatory vision - predicting future states of people, vehicles, and environments from visual...
• Simulation for Autonomous Driving — Explores simulation environments, synthetic data generation, and sim-to-real transfer for autonomous driving research.
• Synthetic and Adversarial Forensics — Addresses detection and forensic analysis of synthetic and adversarially manipulated visual media.
• ScaleBot: Scalable Robot Learning Systems — Examines how robot learning systems can scale via data, compute, and foundation model pre-training.
• Synthetic Data for Computer Vision — Explores synthetic and procedurally generated training data for improving model robustness and reducing annotation costs....
• Transformers for Vision — Explores transformer architectures for image, video, and multimodal tasks including vision-language models.
• UG2+ Workshop — Challenges participants to improve vision systems under adverse conditions including fog, rain, and low light....
• VizWiz Grand Challenge — Advances AI assistance for blind and low-vision users through visual question answering and image description...
• Workshop on Autonomous Driving — Covers perception, prediction, and planning for self-driving vehicles, including benchmark challenges and real-world deployment.
• Explainable AI for Computer Vision — Develops interpretability and explainability methods for deep vision models, addressing transparency in safety-critical applications.