SSPU-Net: Self-Supervised Point Cloud Upsampling via Differentiable Rendering.
Yifan ZhaoLe HuiJin XiePublished in: CoRR (2021)
Keyphrases
- point cloud
- photorealistic
- free viewpoint
- implicit surfaces
- structure from motion
- ray tracing
- surface reconstruction
- point sets
- low resolution
- point cloud data
- laser scanner
- computer graphics
- depth map
- stereo camera
- high fidelity
- high quality
- image based rendering
- triangular mesh
- real time
- multi view
- real world objects
- feature vectors
- image synthesis
- machine learning
- computer vision
- nearest neighbor
- depth information
- single image
- point correspondences
- d scene