LinkNet: 2D-3D linked multi-modal network for online semantic segmentation of RGB-D videos.

Junxiong Cai Tai-Jiang Mu Yu-Kun Lai Shimin Hu

Published in: Comput. Graph. (2021)

Keyphrases

multi modal
semantic segmentation
semantic concepts
high dimensional
conditional random fields
superpixels
image annotation
humanoid robot
weakly supervised
machine learning
video sequences
probabilistic model
video data
object categories
object classes