otthorn/nerf_plus_plus
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

Update README.md

Browse source
This commit is contained in:
Kai Zhang 2020-10-15 15:32:40 -04:00 committed by GitHub
parent e79995e664
commit 496b91679f
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
README.md
View file

@ -1,5 +1,5 @@
# NeRF++ # NeRF++
Codebase for arXiv preprint: Codebase for arXiv preprint "NeRF++: Analyzing and Improving Neural Radiance Fields"
* Work with 360 capture of large-scale unbounded scenes. * Work with 360 capture of large-scale unbounded scenes.
* Support multi-gpu training and inference with PyTorch DistributedDataParallel (DDP). * Support multi-gpu training and inference with PyTorch DistributedDataParallel (DDP).
* Optimize per-image autoexposure (**experimental feature**). * Optimize per-image autoexposure (**experimental feature**).
@ -10,6 +10,7 @@ Codebase for arXiv preprint:
* Data format. * Data format.
* Each scene consists of 3 splits: train/test/validation. * Each scene consists of 3 splits: train/test/validation.
* Intrinsics and poses are stored as flattened 4x4 matrices (row-major). * Intrinsics and poses are stored as flattened 4x4 matrices (row-major).
* Pixel coordinate of an image's upper-left corner is (column, row)=(0, 0).
* Poses are camera-to-world, not world-to-camera transformations. * Poses are camera-to-world, not world-to-camera transformations.
* Opencv camera coordinate system is adopted, i.e., x--->right, y--->down, z--->scene. * Opencv camera coordinate system is adopted, i.e., x--->right, y--->down, z--->scene.
* To convert camera poses between Opencv and Opengl conventions, the following snippet can be used for both Opengl2Opencv and Opencv2Opengl. * To convert camera poses between Opencv and Opengl conventions, the following snippet can be used for both Opengl2Opencv and Opencv2Opengl.