K-Planes

K-Planes is a NeRF-based method representing d-dimensional space using 2 planes allowing for a seamless way to go from static (d=3) to dynamic (d=4) scenes.

Web: https://sarafridov.github.io/K-Planes/
Paper: K-Planes: Explicit Radiance Fields in Space, Time, and Appearance
Authors: Sara Fridovich-Keil, Giacomo Meanti, Frederik Warburg, Benjamin Recht, Angjoo Kanazawa

Blender

Blender (nerf-synthetic) is a synthetic dataset used to benchmark NeRF methods. It consists of 8 scenes of an object placed on a white background. Cameras are placed on a semi-sphere around the object. Scenes are licensed under various CC licenses.

Scene PSNR SSIM LPIPS (VGG) Time GPU mem.
lego 35.73 0.981 0.047 23m 54s 4.64 GB
drums 25.68 0.938 0.096 24m 10s 4.64 GB
ficus 31.31 0.974 0.052 23m 37s 4.64 GB
hotdog 36.46 0.981 0.033 23m 38s 4.64 GB
materials 29.41 0.949 0.070 24m 1s 4.64 GB
mic 33.95 0.988 0.019 24m 6s 4.64 GB
ship 30.72 0.897 0.141 24m 8s 4.64 GB
chair 34.85 0.983 0.036 24m 9s 4.64 GB
Average 32.27 0.961 0.062 23m 58s 4.64 GB

Photo Tourism

Photo Tourism is a dataset of images of famous landmarks, such as the Sacre Coeur, the Trevi Fountain, and the Brandenburg Gate. The images were captured by tourist at different times of the day and year, images have varying lighting conditions and occlusions. The evaluation protocol is based on NeRF-W, where the image appearance embeddings are optimized on the left side of the image and the metrics are computed on the right side of the image.

Scene PSNR SSIM LPIPS Time GPU mem.
Sacre Coeur 19.96 0.762 0.299 24m 26s 3.62 GB
Trevi Fountain 19.70 0.662 0.388 24m 44s 3.59 GB
Brandenburg Gate 23.65 0.859 0.253 24m 40s 3.56 GB
Average 21.10 0.761 0.313 24m 37s 3.59 GB