# Sub-pixel CNN Super Resolution model

- Source: ONNX Model Zoo `onnxmodelzoo/super-resolution-10`, revision `395472c4b6b92f788357237983d83fd913cfe7e3`.
- Source URL: https://huggingface.co/onnxmodelzoo/super-resolution-10
- Model file: `super-resolution-10.onnx`, 240,078 bytes, renamed here to `model.onnx`.
- SHA-256: `85f36ff88cc504a24af5e0602148bc56a8aa09a58eca8c0da2756f3e8186035e`.
- License: Apache License 2.0; see `LICENSE-APACHE-2.0.txt`.
- Architecture and scope: 3x luminance-only sub-pixel CNN. Input `[1,1,224,224]`, output `[1,1,672,672]`. Chroma is interpolated separately by the client. This predicts detail; it does not recover ground-truth content.