Self-supervised Scale Equivariant Network for Weakly Supervised Semantic Segmentation.
Yude WangJie ZhangMeina KanShiguang ShanXilin ChenPublished in: CoRR (2019)
Keyphrases
- weakly supervised
- semantic segmentation
- superpixels
- object class
- topic models
- relation extraction
- semi supervised
- conditional random fields
- scene classification
- named entities
- object categories
- natural images
- information retrieval
- object classes
- object segmentation
- object detectors
- small number
- multiple images
- image understanding
- input image
- multiscale
- computer vision