Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation.
Fatemehsadat SalehMohammad Sadegh Ali AkbarianMathieu SalzmannLars PeterssonJose M. AlvarezPublished in: CoRR (2017)
Keyphrases
- weakly supervised
- semantic segmentation
- foreground objects
- superpixels
- bounding box
- video sequences
- object segmentation
- object class
- object classes
- video frames
- topic models
- multiscale
- object categories
- relation extraction
- semi supervised
- object detection
- segmentation algorithm
- long range
- scene classification
- object detectors
- input image
- named entities
- object recognition
- conditional random fields
- image segmentation
- graph cuts
- image classification
- higher order
- image features
- active learning
- image sequences