Domain Adaptive Semantic Segmentation through Photorealistic Enhancement of Video Game.
Kaito NakajimaTakafumi KatayamaTian SongXiantao JiangTakashi ShimamotoPublished in: ICCE (2022)
Keyphrases
- video games
- photorealistic
- semantic segmentation
- computer graphics
- hand held
- virtual reality
- highly accurate
- learning experience
- complex scenes
- range images
- conditional random fields
- d scene
- superpixels
- computer games
- object categories
- object classes
- vision system
- point cloud
- weakly supervised
- scene classification
- three dimensional
- semi supervised
- active learning