Semantic Segmentation for Point Cloud Scenes via Dilated Graph Feature Aggregation and Pyramid Decoders.
Yongqiang MaoXian SunWenhui DiaoKaiqiang ChenZonghao GuoXiaonan LuKun FuPublished in: CoRR (2022)
Keyphrases
- point cloud
- semantic segmentation
- superpixels
- structure from motion
- conditional random fields
- weakly supervised
- point sets
- graph structure
- multiscale
- scene classification
- object categories
- image features
- input image
- object classes
- natural scenes
- computer vision
- object class
- three dimensional
- machine learning
- pairwise
- viewpoint
- pascal voc
- high dimensional
- learning algorithm