fernandotonon's picture
Upload README.md with huggingface_hub
ec476fa verified
|
Raw
History Blame Contribute Delete
2.02 kB
---
license: cc0-1.0
tags:
- onnx
- pbr
- normal-map
- roughness
- height-map
- texture
- qtmesheditor
library_name: onnx
pipeline_tag: image-to-image
---
# PBRify_Remix — ONNX export
**ONNX re-exports of the CC0 SPAN models from
[Kim2091/PBRify_Remix](https://github.com/Kim2091/PBRify_Remix)**: predict a
tangent-space **normal map**, **roughness map**, and **height map** from a
single albedo/diffuse texture. All credit for the weights goes to Kim2091.
Exported for **[QtMeshEditor](https://github.com/fernandotonon/QtMeshEditor)**
(issue #404), powering `qtmesh material --generate-pbr`, the Material Editor's
"Generate PBR maps from diffuse" button, the `generate_pbr_maps` MCP tool, and
the PBR stage of the image-to-3D pipeline.
> The files QtMeshEditor downloads at runtime live in the shared
> [`fernandotonon/QtMeshEditor-models`](https://huggingface.co/fernandotonon/QtMeshEditor-models)
> repo (top level). This repo is the standalone model card + mirror.
## Files
| file | output |
|---|---|
| `1x-PBRify_NormalV3.onnx` | tangent-space normal (RGB) |
| `1x-PBRify_RoughnessV2.onnx` | roughness (consume as luminance) |
| `1x-PBRify_Height.onnx` | height/displacement (consume as luminance) |
## I/O
`1×3×H×W` float NCHW in `[0,1]``1×3×H×W` out. Dynamic H/W (tile large
textures with overlapping tiles + feathered seam blending). Opset 18,
~1.6 MB each.
## License & provenance
The upstream repo's LICENSE is **CC0-1.0** and its README states the models
were trained *"exclusively on high quality CC0 content from ambientCG"*.
(OpenModelDB's NormalV3 page lists "ambientCG + UltraSharpV2" as the training
set — UltraSharpV2 is cc-by-nc-sa-4.0 — which contradicts the author's own
statement; we treat the author's explicit repo LICENSE + "exclusively CC0"
statement as authoritative. Documented for transparency.)
## Reproducing
`scripts/export-pbrify-onnx.py` in the QtMeshEditor repo (one-time, offline;
spandrel load → `torch.onnx.export`, opset 18, dynamic H/W).