Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulation.
Hongtao WuYa JingChilam CheangGuangzeng ChenJiafeng XuXinghang LiMinghuan LiuHang LiTao KongPublished in: CoRR (2023)
Keyphrases
- visual data
- real time
- mobile robot
- visual information
- visual cues
- visual input
- video content
- video data
- landmark recognition
- visual analysis
- manipulation tasks
- training set
- vision system
- video streams
- video sequences
- discriminative training
- video search
- human robot interaction
- multimedia
- training process
- robot navigation
- visual feedback
- autonomous robots
- humanoid robot
- video analysis
- video retrieval
- video surveillance
- multimedia data
- path planning
- space time
- visual features
- generative model
- real world
- video database
- video clips
- news video
- discriminative classifiers
- classifier training