MIM4D: Masked Modeling with Multi-View Video for Autonomous Driving Representation Learning.
Jialv ZouBencheng LiaoQian ZhangWenyu LiuXinggang WangPublished in: CoRR (2024)
Keyphrases
- multi view
- single view
- multi view learning
- multiple views
- dynamic scenes
- reinforcement learning
- co training
- depth map
- real time
- depth video
- semi supervised
- learning tasks
- autonomous driving
- multi view images
- learning process
- video sequences
- three dimensional
- multiple viewpoints
- free viewpoint
- computer vision
- multi view reconstruction
- learning algorithm