Shortcuts

Unbounded Scene

See code examples/train_ngp_nerf.py at our github repository for details.

Benchmarks

updated on 2022-11-07

Here we trained a Instant-NGP Nerf on the MipNerf360 dataset. We used train split for training and test split for evaluation. Our experiments are conducted on a single NVIDIA TITAN RTX GPU. The training memory footprint is about 6-9GB.

The main difference between working with unbounded scenes and bounded scenes, is that a contraction method is needed to map the infinite space to a finite Occupancy Grid. We have difference options provided for this (see Occupancy Grid). The experiments here is basically the Instant-NGP experiments (see Instant-NGP) with a contraction method that takes from MipNerf360.

Note

Even though we are comparing with Nerf++ and MipNerf360, the model and everything are totally different with them. There are plenty of ideas from those papers that would be very helpful for the performance, but we didn’t adopt them. As this is just a simple example to show how to use the library, we didn’t want to make it too complicated.

PSNR

Garden

Bicycle

Bonsai

Counter

Kitchen

Room

Stump

MEAN

Nerf++ (~days)

24.32

22.64

29.15

26.38

27.80

28.87

24.34

26.21

MipNerf360 (~days)

26.98

24.37

33.46

29.55

32.23

31.63

26.40

29.23

Ours (~20 mins)

25.41

22.97

30.71

27.34

30.32

31.00

23.43

27.31

Ours (Training time)

25min

17min

19min

23min

28min

20min

17min

21min