Occlusion handling using semantic segmentation and visibility-based rendering for mixed reality.
Menandro RoxasTomoki HoriTaiki FukiageYasuhide OkamotoTakeshi OishiPublished in: VRST (2018)
Keyphrases
- mixed reality
- occlusion handling
- semantic segmentation
- stereo matching
- superpixels
- multiple objects
- pedestrian detection
- augmented reality
- object tracking
- conditional random fields
- object categories
- human body
- real time
- high quality
- computer graphics
- stereo images
- ground truth
- multiple images
- dynamic programming
- pairwise
- machine learning
- natural images
- multi view
- multiscale