ST++: Make Self-training Work Better for Semi-supervised Semantic Segmentation.
Lihe YangWei ZhuoLei QiYinghuan ShiYang GaoPublished in: CoRR (2021)
Keyphrases
- semantic segmentation
- semi supervised
- weakly supervised
- semi supervised learning
- co training
- street scenes
- conditional random fields
- supervised methods
- label transfer
- superpixels
- unlabeled data
- labeled data
- scene classification
- pairwise
- multi view
- supervised learning
- unsupervised learning
- generative model
- active learning
- object categories
- object classes
- pascal voc
- image understanding
- image set
- object segmentation
- long range
- training set
- object recognition
- segmentation algorithm
- object class
- machine learning
- graph cuts
- text classification
- object detection
- multiscale