VET: Visual Error Tomography for Point Cloud Completion and High-Quality Neural Rendering.
Linus FrankeDarius RückertLaura FinkMatthias InnmannMarc StammingerPublished in: SIGGRAPH Asia (2023)
Keyphrases
- point cloud
- high quality
- photorealistic
- surface reconstruction
- structure from motion
- real world objects
- point sets
- stereo camera
- implicit surfaces
- high fidelity
- image reconstruction
- laser scanner
- neural network
- visual processing
- real time
- point cloud data
- ground truth
- triangular mesh
- high resolution
- highly accurate
- computer vision
- computer graphics
- depth map
- semi supervised
- machine learning
- lidar data
- iterative closest point