Generate calibrated multi-view training data from any GLB model. 100–400 views with RGB, depth, normals, masks, point clouds, and camera poses. No GPU required.
10 free credits on signup — no watermarks, no credit card required.
This interactive 3DGS was trained from a DX.GL dataset in 10 minutes. Drag to orbit.
10 CC0 models from Polyhaven · Trained with nerfstudio splatfacto · View collection · Use ← → to browse
Polyhaven-10 on HuggingFace →1,026 curated Objaverse objects × 196 views × 1024×1024 × 6 modalities. Full sphere coverage with RGB, depth, normals, masks, point clouds, and nerfstudio-compatible camera poses. CC0 and CC-BY licensed with full attribution.
Quick start:
pip install requests python download_all.py --object "chair" ns-train splatfacto --data ./dxgl-objaverse-1k/chair
Compared to existing Objaverse render sets (Zero123: 12 views at 256px, GObjaverse: 38 views at 512px), Objaverse-1K provides 5–16× denser views with 6 modalities instead of 1–4.
Each dataset is a ZIP containing calibrated multi-view renders with full modality coverage.
100–400 PNG images with transparent backgrounds. Fibonacci hemisphere or full sphere sampling for uniform coverage.
Both 8-bit and 16-bit grayscale depth maps. Tight near/far planes computed from actual model geometry — not screen-space approximations.
Per-pixel surface normal maps in camera space. Useful for surface reconstruction, relighting, and material estimation.
Clean object/background segmentation masks. Transparent backgrounds ensure zero contamination in training data.
PLY point cloud generated alongside views. Ready for initialization or evaluation in your reconstruction pipeline.
nerfstudio-compatible transforms.json with per-frame intrinsics, extrinsics, and depth_near/depth_far. Works with nerfacto, splatfacto, and custom loaders.
Choose the tier that fits your research. All tiers include every modality.
| Tier | Views | Resolution | Credits |
|---|---|---|---|
| 100×800 | 100 | 800 × 800 | 4 |
| 196×1024 | 196 | 1024 × 1024 | 16 |
| 400×2048 | 400 | 2048 × 2048 | 64 |
Credits: $9.90 × 10 · $79 × 100 · $490 × 1,000. All prices USD, excl. VAT.
Every tier includes RGB + depth (8+16-bit) + normals + masks + PLY + transforms.json.
Skip the Blender scripts and GPU provisioning. Focus on your research.
Upload a GLB, get a dataset. No Blender, no CUDA setup, no local rendering. We handle the compute.
PBR rendering on controlled GPU hardware. Same lighting, same camera model, same output quality — every time.
8-bit and 16-bit depth maps with tight near/far planes from actual model geometry. Not screen-space approximations.
Fibonacci hemisphere or sphere sampling. nerfstudio transforms.json with intrinsics + extrinsics per frame.
Render datasets for hundreds of models via REST API. No manual Blender scripting or GPU provisioning.
RGBA output with no background contamination. Clean masks, clean depth — your training data stays clean.
Three steps from 3D model to training data.
Upload a GLB via drag-and-drop, URL, or API. Files up to 100 MB. OBJ scans in ZIP are converted automatically.
Select views (100 or 400), resolution (800–2048), and coverage (hemisphere or full sphere). Click render.
Get a ZIP with images/, depth/, depth_16bit/, normals/, masks/, points3D.ply, transforms.json, and overview.webp.
Render datasets programmatically for batch pipelines.
# Upload a model curl -X POST https://dx.gl/v1/models \ -H "Authorization: Bearer dxgl_sk_..." \ -F "[email protected]" # Render a dataset curl -X POST https://dx.gl/v1/renders \ -H "Authorization: Bearer dxgl_sk_..." \ -H "Content-Type: application/json" \ -d '{"modelId":"abc123","output":"dataset","datasetQuality":"196x1024","coverage":"hemisphere"}'
Use the DX.GL MCP server to let AI agents in Windsurf, Claude Desktop, or Cursor generate datasets across your entire model library. Describe what you want in plain English — the agent handles the rest.
“Generate 196×1024 sphere datasets for every model tagged validation-set. Quote the cost first.”
Read the MCP for Datasets guide →Create an account, upload your first model, and get a dataset in minutes. Every new account gets 10 free credits on signup — no watermarks, no credit card required.