Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation.
Guolei SunWenguan WangJifeng DaiLuc Van GoolPublished in: CoRR (2020)
Keyphrases
- semantic segmentation
- weakly supervised
- superpixels
- scene classification
- conditional random fields
- object class
- topic models
- object categories
- multiscale
- relation extraction
- input image
- pascal voc
- object detection
- image classification
- semi supervised
- image retrieval
- named entities
- image set
- image features
- object classes
- computer vision
- single image
- image representation
- natural images
- natural language processing
- segmentation method
- object segmentation
- segmentation algorithm
- object detectors
- probabilistic model
- co occurrence
- text mining