Importance of Self-Consistency in Active Learning for Semantic Segmentation.
S. Alireza GolestanehKris M. KitaniPublished in: CoRR (2020)
Keyphrases
- semantic segmentation
- active learning
- street scenes
- conditional random fields
- label transfer
- superpixels
- object categories
- scene classification
- training examples
- weakly supervised
- object class
- semi supervised
- object classes
- machine learning
- training set
- pascal voc
- image understanding
- object segmentation
- viewpoint
- decision trees
- labeled data
- pixel level
- learning process
- image set