A Self-Training Framework Based on Multi-Scale Attention Fusion for Weakly Supervised Semantic Segmentation.
Guoqing YangChuang ZhuYu ZhangPublished in: CoRR (2023)
Keyphrases
- weakly supervised
- semantic segmentation
- superpixels
- multiscale
- semi supervised
- object class
- topic models
- relation extraction
- machine learning
- pairwise
- probabilistic model
- object detectors
- image representation
- pascal voc
- object classes
- segmentation algorithm
- natural images
- d objects
- feature space
- image segmentation
- computer vision