NVlabs/intrinsic3d
A 2017 ICCV paper implementing joint optimization of 3D geometry, surface albedo, camera poses, and scene lighting from RGB-D sensor data.
Not currently ranked — collecting fresh signals.
star history
Intrinsic3D recovers high-quality 3D reconstructions from low-cost RGB-D sensors by simultaneously optimizing reconstructed geometry, surface albedos, camera poses, and spatially-varying lighting modeled via spherical harmonics. The method uses SDF-based surface representation and classical optimization rather than deep learning. Published by NVIDIA Research and TU Munich.
Frequently asked
- What is NVlabs/intrinsic3d?
- A 2017 ICCV paper implementing joint optimization of 3D geometry, surface albedo, camera poses, and scene lighting from RGB-D sensor data.
- Is intrinsic3d open source?
- Yes — NVlabs/intrinsic3d is open source, released under the BSD-3-Clause license.
- What language is intrinsic3d written in?
- NVlabs/intrinsic3d is primarily written in C++.
- How popular is intrinsic3d?
- NVlabs/intrinsic3d has 458 stars on GitHub.
- Where can I find intrinsic3d?
- NVlabs/intrinsic3d is on GitHub at https://github.com/NVlabs/intrinsic3d.