Twin Deformable Point Convolutions for Point Cloud Semantic Segmentation in Remote Sensing Scenes.
Yongqiang MaoHanbo BiXuexue LiKaiqiang ChenZhirui WangXian SunKun FuPublished in: CoRR (2024)
Keyphrases
- image sequences
- optical flow
- remote sensing
- point cloud
- semantic segmentation
- structure from motion
- computer vision
- high resolution
- conditional random fields
- image analysis
- image processing
- moving objects
- scene classification
- superpixels
- image fusion
- natural scenes
- point sets
- pattern recognition
- object categories
- viewpoint
- depth map
- pairwise
- object classes
- training data
- image understanding
- multiscale
- deformable models
- object recognition
- natural images