Static NeRFs¶
Performance Overview¶
Methods |
Dataset |
Training Time \(\downarrow\) |
PSNR \(\uparrow\) |
LPIPS \(\downarrow\) |
|---|---|---|---|---|
TensoRF [1] |
Tanks&Temple |
19min |
28.11 |
0.167 |
+nerfacc (occgrid) |
14min |
28.06 |
0.174 |
|
TensoRF [1] |
NeRF-Synthetic |
10.3min |
32.73 |
0.049 |
+nerfacc (occgrid) |
7.1min |
32.52 |
0.054 |
|
NeRF [2] |
NeRF-Synthetic |
days |
31.00 |
0.047 |
+nerfacc (occgrid) |
1hr |
31.55 |
0.072 |
|
Instant-NGP [3] |
NeRF-Synthetic |
4.4min |
32.35 |
|
+nerfacc (occgrid) |
4.5min |
33.11 |
0.053 |
|
+nerfacc (propnet) |
4.0min |
31.76 |
0.062 |
|
Instant-NGP [3] |
Mip-NeRF 360 |
5.3min |
25.93 |
|
+nerfacc (occgrid) |
5.0min |
26.41 |
0.353 |
|
+nerfacc (propnet) |
4.9min |
27.58 |
0.292 |
Implementation Details¶
3rd-Party Use Cases¶
nerfstudio: A collaboration friendly studio for NeRFs.
modelscope: A collection of deep-learning algorithms.
sdfstudio: A Unified Framework for Surface Reconstruction.
instant-nsr-pl: Train NeuS in 10min.