GraphNet: Learning Image Pseudo Annotations for Weakly-Supervised Semantic Segmentation.
Mengyang PuYaping HuangQingji GuanQi ZouPublished in: ACM Multimedia (2018)
Keyphrases
- weakly supervised
- semantic segmentation
- superpixels
- object class
- input image
- relation extraction
- semi supervised
- named entities
- multiscale
- object detectors
- topic models
- image content
- learning algorithm
- conditional random fields
- scene classification
- segmentation algorithm
- image retrieval
- object detection
- image representation
- segmentation method
- image classification
- hidden markov models
- test images
- viewpoint
- automatic extraction
- information extraction
- object segmentation
- higher order
- spatial information
- super resolution