Multi-Miner: Object-Adaptive Region Mining for Weakly-Supervised Semantic Segmentation.
Kuangqi ZhouQibin HouZun LiJiashi FengPublished in: CoRR (2020)
Keyphrases
- weakly supervised
- semantic segmentation
- superpixels
- object classes
- bounding box
- object class
- relation extraction
- object detectors
- d objects
- topic models
- object segmentation
- region of interest
- object categories
- semi supervised
- conditional random fields
- object recognition
- scene classification
- multiple objects
- target object
- input image
- multiscale
- pascal voc
- image regions
- object model
- named entities
- text mining
- object detection
- partial occlusion
- domain specific
- co occurrence
- segmentation method
- image features
- active learning