GSVNET: Guided Spatially-Varying Convolution for Fast Semantic Segmentation on Video.
Shih-Po LeeSi-Cun ChenWen-Hsiao PengPublished in: ICME (2021)
Keyphrases
- spatially varying
- semantic segmentation
- photometric stereo
- superpixels
- conditional random fields
- noisy images
- motion blur
- video sequences
- video data
- blurred images
- video frames
- object categories
- space time
- image processing
- object classes
- shape from shading
- light source
- generative model
- object detection
- feature vectors
- object recognition
- feature extraction