Add RealPLKSR ONNX models, source checkpoints, and model card

README.md

@@ -1,3 +1,53 @@
 ---
 license: mit
+library_name: onnx
+pipeline_tag: image-to-image
+tags:
+  - super-resolution
+  - image-restoration
+  - photography
+  - darktable
 ---
+
+# RealPLKSR (ONNX)
+
+ONNX conversion of [RealPLKSR](https://github.com/dslisleedh/PLKSR) – the Real-world variant of Partial Large Kernel CNNs for Super-Resolution. Includes 2x and 4x scale factors, MSSIM-pretrain stage (no GAN finetune).
+
+Packaged for use with [darktable](https://darktable.org)'s neural restore module via the [darktable-ai](https://github.com/darktable-org/darktable-ai) pipeline.
+
+## Files
+
+| Path                                          | Purpose                                              |
+|-----------------------------------------------|------------------------------------------------------|
+| `onnx/model_x2.onnx`                          | 2x upscaler, static input 512×512 → output 1024×1024 |
+| `onnx/model_x4.onnx`                          | 4x upscaler, static input 256×256 → output 1024×1024 |
+| `checkpoints/2x_realplksr_mssim_pretrain.pth` | Original PyTorch weights for the x2 model            |
+| `checkpoints/4x_realplksr_mssim_pretrain.pth` | Original PyTorch weights for the x4 model            |
+| `config.json`                                 | HF model metadata                                    |
+
+The `checkpoints/` directory holds the original `.pth` files used to produce the ONNX exports. They are kept here so the conversion can be reproduced even if the upstream Google Drive links stop being reachable.
+
+## Source
+
+- Architecture and weights: [dslisleedh/PLKSR](https://github.com/dslisleedh/PLKSR) (MIT) – Dongheon Lee et al., Machine Intelligence Laboratory, University of Seoul
+- Paper: [Partial Large Kernel CNNs for Efficient Super-Resolution](https://arxiv.org/abs/2404.11848) (2024)
+- Checkpoint release thread: [dslisleedh/PLKSR#4](https://github.com/dslisleedh/PLKSR/issues/4) ("Real-world PLKSR")
+- Trained via the [neosr](https://github.com/neosr-project/neosr) framework with the RealESRGAN degradation pipeline
+- Conversion code: [darktable-org/darktable-ai](https://github.com/darktable-org/darktable-ai) under GPL-3.0+
+- Original Google Drive links (subject to availability):
+  - x2: <https://drive.google.com/file/d/1GAdf5VOqYa5ntswT9sYsKKZ2Z7OQp7gO/view>
+  - x4: <https://drive.google.com/file/d/12ek1vitEporWc5qqaYo6AMy0-RYlRqu8/view>
+
+## Usage
+
+Inference inputs are RGB images in the [0, 1] range. The graphs have static input dimensions, so callers must tile at exactly the declared size. See the [darktable-ai conversion script](https://github.com/darktable-org/darktable-ai/blob/master/models/upscale-realplksr/convert.py) for the full export configuration, and the [demo script](https://github.com/darktable-org/darktable-ai/blob/master/models/upscale-realplksr/demo.py) for an ONNX Runtime example that handles tiling, mirror-padding, and overlap stitching.
+
+## License
+
+- **Model weights** (in `onnx/` and `checkpoints/`): MIT, from upstream [dslisleedh/PLKSR](https://github.com/dslisleedh/PLKSR).
+- **Conversion script and packaging**: GPL-3.0+, see [darktable-org/darktable-ai](https://github.com/darktable-org/darktable-ai).
+
+## Notes
+
+- Training dataset is not documented by the weights author; assumed to be DF2K (DIV2K + Flickr2K) per neosr common practice. Flickr2K does not carry an explicit open-source license.
+- MSSIM-pretrain checkpoints only (no GAN finetune) – conservative output, no hallucinated detail.

checkpoints/2x_realplksr_mssim_pretrain.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0214b82d20a2b7d524c9f70828c7a9b0113a3058ff3eb3cabe2b11783645c381
+size 29595266

checkpoints/4x_realplksr_mssim_pretrain.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37a1a6245ea29dc17655b1df64ba5090b0d5ed9d4b50525a722c53e378d552f1
+size 29678402

config.json

@@ -0,0 +1,7 @@
+{
+  "model_type": "onnx",
+  "library_name": "onnx",
+  "architectures": ["RealPLKSR"],
+  "task": "image-to-image",
+  "pipeline_tag": "image-to-image"
+}

onnx/model_x2.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7abb65092f3808d3aa255ffdab42b1d672883915d90adbdd321204168b9f293
+size 29627920

onnx/model_x4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ce692652494e2e5c31e1f7c7ab61eb084e9a450cde727a374f5d6746cfe054b
+size 29711010