Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation.
Jiaqi GuHyoukjun KwonDilin WangWei YeMeng LiYu-Hsin ChenLiangzhen LaiVikas ChandraDavid Z. PanPublished in: CoRR (2021)
Keyphrases
- semantic segmentation
- multiscale
- superpixels
- high resolution
- image processing
- street scenes
- low resolution
- label transfer
- weakly supervised
- natural images
- conditional random fields
- scene classification
- computer vision
- remote sensing
- scale space
- pixel level
- object class
- super resolution
- image representation
- pascal voc
- object categories
- object recognition
- image segmentation
- image fusion
- high quality
- denoising
- image understanding
- visual words
- image patches
- probabilistic model
- test images
- keypoints
- active learning
- face images
- object detection