Optical Flow Training Under Limited Label Budget via Active Learning.
Shuai YuanXian SunHannah Halin KimShuzhi YuCarlo TomasiPublished in: ECCV (22) (2022)
Keyphrases
- active learning
- optical flow
- active learner
- training examples
- labeling effort
- training set
- label noise
- supervised learning
- unlabeled instances
- annotation effort
- semi supervised
- online learning
- random sampling
- batch mode
- selective sampling
- computer vision
- active learning framework
- least squares
- machine learning
- image sequences
- pool based active learning
- class labels
- training phase
- motion field
- learning strategies
- labeled data
- motion estimation
- learning algorithm
- limited memory
- d scene
- sample selection
- motion model
- test set
- learning environment