LAKe-Net: Topology-Aware Point Cloud Completion by Localizing Aligned Keypoints.
Junshu TangZhijun GongRan YiYuan XieLizhuang MaPublished in: CVPR (2022)
Keyphrases
- point cloud
- keypoints
- structure from motion
- implicit surfaces
- object recognition
- scale invariant
- image matching
- sift features
- multiscale
- point sets
- laser scanner
- stereo camera
- keypoint detection
- feature detection
- scale space
- image features
- point cloud data
- visual words
- sift descriptors
- machine learning
- keypoint matching
- viewpoint
- high quality
- image segmentation