Importance of Self-Consistency in Active Learning for Semantic Segmentation.
S. Alireza GolestanehKris KitaniPublished in: BMVC (2020)
Keyphrases
- semantic segmentation
- active learning
- street scenes
- conditional random fields
- label transfer
- weakly supervised
- superpixels
- scene classification
- object classes
- semi supervised
- object categories
- learning algorithm
- object class
- image set
- long range
- image understanding
- learning process
- machine learning
- training examples
- segmentation algorithm
- pascal voc
- training set
- computer vision