README.md

---
license: cc0-1.0
task_categories:
  - image-to-3d
  - depth-estimation
tags:
  - nerf
  - 3d-gaussian-splatting
  - 3dgs
  - nerfstudio
  - multi-view
  - depth-maps
  - normal-maps
  - point-cloud
  - computer-vision
  - 3d-reconstruction
pretty_name: "DX.GL Multi-View Datasets"
size_categories:
  - 1K<n<10K
---

# DX.GL Multi-View Datasets for NeRF & 3D Gaussian Splatting

Multi-view training datasets rendered from CC0 3D models via [DX.GL](https://dx.gl). Each dataset includes calibrated camera poses, depth maps, normal maps, binary masks, and point clouds — ready for [nerfstudio](https://docs.nerf.studio/) out of the box.

**10 objects × 196 views × 1024×1024 resolution × full sphere coverage.**

## Quick Start

```bash
# Download a dataset (Apple, 196 views, 1024x1024)
wget https://dx.gl/api/v/EJbs8npt2RVM/vCHDLxjWG65d/dataset -O apple.zip
unzip apple.zip -d apple

# Train with nerfstudio
pip install nerfstudio
ns-train splatfacto --data ./apple \
  --max-num-iterations 20000 \
  --pipeline.model.sh-degree 3 \
  --pipeline.model.background-color white
```

Or use the download script:

```bash
pip install requests
python download_all.py
```

## What's in Each Dataset ZIP

```
dataset/
├── images/           # RGB frames (PNG, transparent background)
│   ├── frame_00000.png
│   └── ...
├── depth/            # 8-bit grayscale depth maps
├── depth_16bit/      # 16-bit grayscale depth maps (higher precision)
├── normals/          # World-space normal maps
├── masks/            # Binary alpha masks
├── transforms.json   # Camera poses (nerfstudio / instant-ngp format)
└── points3D.ply      # Sparse point cloud for initialization
```

### transforms.json Format

Compatible with both **nerfstudio** and **instant-ngp**:

```json
{
  "camera_angle_x": 0.857,
  "camera_angle_y": 0.857,
  "fl_x": 693.5,
  "fl_y": 693.5,
  "cx": 400,
  "cy": 400,
  "w": 800,
  "h": 800,
  "depth_near": 0.85,
  "depth_far": 2.35,
  "ply_file_path": "points3D.ply",
  "frames": [
    {
      "file_path": "images/frame_00000.png",
      "depth_file_path": "depth/frame_00000.png",
      "normal_file_path": "normals/frame_00000.png",
      "mask_file_path": "masks/frame_00000.png",
      "transform_matrix": [[...], [...], [...], [0, 0, 0, 1]]
    }
  ]
}
```

## Specs

| Property | Value |
|---|---|
| **Views** | 196 per object |
| **Resolution** | 1024×1024 |
| **Coverage** | Full sphere (±89° elevation) |
| **Point cloud** | ~200k points |
| **Camera distribution** | Fibonacci golden-angle spiral |
| **Background** | Transparent (RGBA) |
| **Lighting** | Studio HDRI + directional lights |

## Camera Distribution

Views are distributed on a full sphere (±89° elevation) using a golden-angle Fibonacci spiral. The distribution is uniform in solid angle — more views near the equator, fewer near the poles — optimized for NeRF/3DGS training.

![Camera Distribution](camera-distribution.png)

## Objects

| # | Object | Category | Download | Browse |
|---|---|---|---|---|
| 1 | Apple | organic | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/vCHDLxjWG65d/dataset) | [View](https://dx.gl/datasets/vCHDLxjWG65d) |
| 2 | Cash Register | electronics | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/JfjLRexr6J7z/dataset) | [View](https://dx.gl/datasets/JfjLRexr6J7z) |
| 3 | Drill | tool | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/A0dcsk7HHgAg/dataset) | [View](https://dx.gl/datasets/A0dcsk7HHgAg) |
| 4 | Fire Extinguisher | metallic | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/cLgyqM5mhQoq/dataset) | [View](https://dx.gl/datasets/cLgyqM5mhQoq) |
| 5 | LED Lightbulb | glass | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/ZuYmv3K9xN7u/dataset) | [View](https://dx.gl/datasets/ZuYmv3K9xN7u) |
| 6 | Measuring Tape | tool | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/qqvDYx7RtHZd/dataset) | [View](https://dx.gl/datasets/qqvDYx7RtHZd) |
| 7 | Modern Arm Chair | furniture | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/KLBJAuie9JaB/dataset) | [View](https://dx.gl/datasets/KLBJAuie9JaB) |
| 8 | Multi Cleaner 5L | product | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/79gDW15Gw9Ft/dataset) | [View](https://dx.gl/datasets/79gDW15Gw9Ft) |
| 9 | Potted Plant | organic | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/o4c5zRyGuT7W/dataset) | [View](https://dx.gl/datasets/o4c5zRyGuT7W) |
| 10 | Wet Floor Sign | plastic | [ZIP](https://dx.gl/api/v/EJbs8npt2RVM/tHdRul1GzzoU/dataset) | [View](https://dx.gl/datasets/tHdRul1GzzoU) |

All source models from [Polyhaven](https://polyhaven.com) (CC0).

## Pre-trained 3DGS Splats

We include pre-trained Gaussian Splat `.ply` files (nerfstudio splatfacto, 20k iterations, SH degree 3) for each object. Download them with:

```bash
python download_all.py --splats
```

Or view them directly:

- [DX.GL Splat Viewer](https://dx.gl/splat/index.html) (all 10 models, use ← → to browse)
- [SuperSplat Editor](https://superspl.at/editor) (drag-drop the .ply)
- nerfstudio viewer: `ns-viewer --load-config outputs/*/config.yml`

### Training Parameters

```bash
ns-train splatfacto --data ./dataset \
  --max-num-iterations 20000 \
  --pipeline.model.sh-degree 3 \
  --pipeline.model.background-color white \
  --pipeline.model.cull-alpha-thresh 0.2 \
  --pipeline.model.densify-size-thresh 0.005 \
  --pipeline.model.use-scale-regularization True \
  --pipeline.model.max-gauss-ratio 5.0
```

Training time: ~10 minutes on RTX 4000 Pro Ada (70W) at the 196×1024 tier.

## Rendering Pipeline

Datasets are rendered using [DX.GL](https://dx.gl)'s cloud GPU rendering pipeline:

- **Lighting**: Studio HDRI environment with PBR materials
- **Camera**: Fibonacci golden-angle sphere distribution
- **Depth**: Tight near/far planes from model bounding sphere for maximum precision
- **Point cloud**: Back-projected from depth maps, ~1000 points per view
- **Background**: Transparent (RGBA)

## Modalities

| Modality | Format | Notes |
|---|---|---|
| **RGB** | PNG, RGBA | Transparent background, PBR-lit |
| **Depth (8-bit)** | PNG, grayscale | Normalized to near/far range |
| **Depth (16-bit)** | PNG, grayscale | RG-encoded, higher precision |
| **Normals** | PNG, RGB | World-space, MeshNormalMaterial |
| **Masks** | PNG, grayscale | Binary alpha from RGB alpha channel |
| **Point Cloud** | PLY, binary | XYZ + RGB, ~100k points |
| **Camera Poses** | JSON | 4×4 camera-to-world matrices |

## License

All source 3D models are **CC0** (public domain) from [Polyhaven](https://polyhaven.com). The rendered datasets inherit this license — use them for anything, no attribution required.

## Citation

```bibtex
@misc{dxgl_multiview_2026,
  title  = {DX.GL Multi-View Datasets for NeRF and 3D Gaussian Splatting},
  author = {DXGL},
  year   = {2026},
  url    = {https://huggingface.co/datasets/dxgl/multiview-datasets},
  note   = {Multi-view datasets with depth, normals, masks, and point clouds. Rendered via DX.GL.}
}
```

## Links

- **This collection**: [dx.gl/datasets/polyhaven-10](https://dx.gl/datasets/polyhaven-10)
- **Browse all datasets**: [dx.gl/datasets](https://dx.gl/datasets)
- **Pipeline details**: [dx.gl/for-research](https://dx.gl/for-research)
- **API documentation**: [dx.gl/portal/docs](https://dx.gl/portal/docs)
- **Generate your own**: [dx.gl/signup](https://dx.gl/signup) (2 free renders included)

## Feedback

We're actively improving the rendering pipeline. If you find issues with depth accuracy, mask quality, camera calibration, or view distribution — please open a Discussion on this repo. Specific feedback we're looking for:

- Depth map accuracy at object edges
- Mask quality for transparent/reflective materials
- Point cloud alignment with RGB views
- View distribution quality for your training method
- Missing modalities or metadata
- Any other issues or suggestions?