MVSSC: Meta-reinforcement learning based visual indoor navigation using multi-view semantic spatial context.
Wanruo ZhangHong LiuJianbing WuYidi LiPublished in: Pattern Recognit. Lett. (2024)
Keyphrases
- multi view
- spatial context
- indoor navigation
- reinforcement learning
- single view
- spatial and temporal
- multiple views
- contextual information
- three dimensional
- depth map
- d objects
- high level
- spatial information
- spatial relationships
- multi view images
- image regions
- machine learning
- visual information
- low level
- semi supervised
- high quality
- learning algorithm
- multi view clustering
- co occurrence
- supervised learning
- spatial databases
- object recognition
- multiscale
- robot control
- structured environments
- real time