Andrew Zisserman
Graph View
Backlinks
- A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames
- Efficiently Reconstructing Dynamic Scenes One D4RT at a Time
- Learning from Streaming Video with Orthogonal Gradients
- Quo Vadis, Action Recognition A New Model and the Kinetics Dataset
- Recurrent Video Masked Autoencoders
- Scaling 4D Representations
- Seeing without Pixels - Perception from Camera Trajectories
- Self-Supervised Learning of Structured Dynamics from Videos
- Unique Lives, Shared World - Learning from Single-Life Videos
- Video Generation Models are General-Purpose Vision Learners