CI-Net: Contextual Information for Joint Semantic Segmentation and Depth Estimation.
Tianxiao GaoWu WeiZhongbin CaiZhun FanShane S. Q. XieXinmei WangQiuda YuPublished in: CoRR (2021)
Keyphrases
- contextual information
- semantic segmentation
- depth estimation
- depth map
- stereo vision
- context aware
- superpixels
- conditional random fields
- stereo matching
- dynamic scenes
- depth information
- scene classification
- weakly supervised
- real scenes
- object categories
- high level
- scene understanding
- multi view
- stereo pair
- object classes
- object recognition
- disparity map
- feature matching
- three dimensional
- multiscale
- space time
- input image
- viewpoint
- object detection
- higher order
- graphical models
- super resolution