PFENet++: Boosting Few-shot Semantic Segmentation with the Noise-filtered Context-aware Prior Mask.
Xiaoliu LuoZhuotao TianTaiping ZhangBei YuYuan Yan TangJiaya JiaPublished in: CoRR (2021)
Keyphrases
- context aware
- semantic segmentation
- contextual information
- context awareness
- superpixels
- conditional random fields
- ubiquitous computing
- mobile devices
- weakly supervised
- scene classification
- context aware systems
- object categories
- video sequences
- object class
- bounding box
- feature selection
- prior knowledge
- video data
- multiscale
- current context
- image set
- key frames
- graphical models
- semi supervised
- computer vision