Marc Pollefeys
- EPFL-Smart-Kitchen: An Ego-Exo Multi-Modal Dataset for Challenging Action and Motion Understanding in Video-Language Models
- Object-X: Learning to Reconstruct Multi-Modal 3D Object Representations
- Scaling Image Geo-Localization to Continent Level
- Situat3DChange: Situated 3D Change Understanding Dataset for Multimodal Large Language Model
- Video Perception Models for 3D Scene Synthesis