Skip to content

Dataset · rishitdagli

Scene collection for NeRF and 3D Gaussian Splatting

rishitdagli/nerf-gs-datasets

A collection of image datasets ranging from 1K to 10K objects with pre-trained Gaussian Splat models for 3D reconstruction training.

What's inside

This is a compilation of existing image datasets from various sources, prepared for training NeRF and 3D Gaussian Splatting reconstruction methods. The collection size ranges from 1k to 10k items. Each individual scene in the set includes a pre-trained Gaussian Splat model.

What you can do with it

  • Training and testing NeRF and 3D Gaussian Splatting algorithms
  • Comparing new rendering methods using ready-made 3D scenes
Size
1K–10K rows
Downloads this month
36.5K
Likes
3

How to load it

from datasets import load_dataset

ds = load_dataset("rishitdagli/nerf-gs-datasets")