tink/nerfstudio
4
0 Fork 0
mirror of https://github.com/nerfstudio-project/nerfstudio.git synced 2026-11-13 03:14:00 +08:00
Code Activity

Update README.md

Browse Source
This commit is contained in:
Justin Kerr authored and GitHub committed 2023-04-03 12:51:05 -07:00
1 parent 71d235d673
commit 7ee1e31281
1 file changed
+11 -5
Download Patch File Download Diff File Expand all files Collapse all files
projects/lerf/README.md
+11 -5
View File
@@ -45,15 +45,21 @@ The images below show the rgb, raw, centered, and normalized output views for th
<div align='center'>
<img src="readme_images/lily_rgb.jpg" width="50px">
<img src="readme_images/lily_raw.jpg" width="50px">
<img src="readme_images/lily_centered.jpg" width="50px">
<img src="readme_images/lily_normalized.jpg" width="50px">
<img src="readme_images/lily_rgb.jpg" width="150px">
<img src="readme_images/lily_raw.jpg" width="150px">
<img src="readme_images/lily_centered.jpg" width="150px">
<img src="readme_images/lily_normalized.jpg" width="150px">
</div>
## Resolution
The Nerfstudio viewer dynamically changes resolution to achieve a desired training throughput.
**To increase resolution, pause training**. Rendering at high resolution (512 or above) can take a second or two, so we recommend rendering at 256px
## `lerf-big` and `lerf-lite`
If your GPU is struggling on memory, we provide a `lerf-lite` implementation that reduces the LERF network capacity and number of samples along rays. If you find you still need to reduce memory footprint, the most impactful parameters for memory are `num_lerf_samples` and
`lerf-big` provides a larger model that uses ViT-L/14 instead of ViT-B/16 for those with large memory GPUs.
# Extending LERF
TODO