L2G Auto-encoder: Understanding Point Clouds by Local-to-Global Reconstruction with Hierarchical Self-Attention.
Xinhai LiuZhizhong HanXin WenYu-Shen LiuMatthias ZwickerPublished in: CoRR (2019)
Keyphrases
- point cloud
- architectural scenes
- urban scenes
- surface reconstruction
- structure from motion
- point sets
- multi view stereo
- stereo camera
- laser scanner
- reconstructed surface
- image reconstruction
- cad model
- bit rate
- triangular mesh
- lidar data
- three dimensional
- visual hull
- multi view
- icp algorithm
- motion estimation
- reconstruction method
- uncalibrated images
- high resolution