Improving LiDAR Semantic Segmentation on Minority Classes and Generalization Capability Using U-Net++ for Self-Driving Scenes.
Chiao-Hua TsengYu-Ting LinWen-Chieh LinChieh-Chih WangPublished in: J. Inf. Sci. Eng. (2024)
Keyphrases
- semantic segmentation
- street scenes
- scene classification
- conditional random fields
- superpixels
- label transfer
- natural scenes
- weakly supervised
- pascal voc
- object categories
- bag of words
- object classes
- active learning
- object detection
- image understanding
- long range
- multiple objects
- image segmentation
- visual attention
- visual words
- natural images
- text classification
- image classification
- graphical models
- multiscale