Combining clustering and active learning for the detection and learning of new image classes.
Luiz F. S. ColettaMoacir PontiEduardo R. HruschkaAyan AcharyaJoydeep GhoshPublished in: Neurocomputing (2019)
Keyphrases
- active learning
- learning algorithm
- image data
- learning process
- input image
- unsupervised learning
- image features
- supervised learning
- multiscale
- single image
- image retrieval
- image analysis
- image content
- image classification
- detection algorithm
- segmentation method
- segmentation algorithm
- labeled data
- edge detection
- semi supervised
- low level
- reinforcement learning
- image regions
- bounding box
- keypoints
- positive examples
- learning strategies
- clustering method
- image representation
- k means
- pairwise
- clustering algorithm