ICNet for Real-Time Semantic Segmentation on High-Resolution Images.
Hengshuang ZhaoXiaojuan QiXiaoyong ShenJianping ShiJiaya JiaPublished in: CoRR (2017)
Keyphrases
- high resolution images
- semantic segmentation
- high resolution
- low resolution
- low resolution images
- super resolution
- superpixels
- reconstruction method
- conditional random fields
- weakly supervised
- scene classification
- object categories
- object detection
- three dimensional
- object class
- image set
- pascal voc
- image reconstruction
- object segmentation
- image patches
- light source
- face recognition