FsaNet: Frequency Self-attention for Semantic Segmentation.
Fengyu ZhangAshkan PanahiGuangjun GaoPublished in: CoRR (2022)
Keyphrases
- semantic segmentation
- street scenes
- conditional random fields
- superpixels
- label transfer
- scene classification
- object categories
- weakly supervised
- pascal voc
- hidden markov models
- bag of words
- object class
- visual attention
- image understanding
- natural images
- object classes
- image patches
- bounding box
- object segmentation
- image segmentation
- segmentation method
- segmentation algorithm
- image representation
- generative model
- object detection
- probabilistic model
- active learning