NAFNet

Install

NAFNet needs no optional extra. Everything it imports is in the base install.

pip install libreyolo

Predict

Weights download from Hugging Face on first use and are cached locally.

from libreyolo import LibreYOLO model = LibreYOLO("LibreNAFNetl-restore-sidd.pt")result = model("noisy.jpg", save=True) restored = result.restoredprint(restored.array.shape)

libreyolo predict model=LibreNAFNetl-restore-sidd.pt source=noisy.jpg save=True

from libreyolo import LibreYOLO model = LibreYOLO("LibreNAFNetl-restore-sidd.pt")result = model.predict("noisy.jpg") result.restored.save("denoised.png")

The returned Results object carries one field for this family, restored, a dense HWC uint8 RGB image on the original canvas; there are no boxes to iterate. save=True writes that restored image straight to disk rather than drawing an annotation over the input. conf, iou and max_det are accepted for signature parity with every other family but have no effect, since restoration produces no detections to filter. See prediction for sources, streaming and result handling.

Variants

Two widths share this architecture: s (width 32) and l (width 64), both built around a 256 px training patch. Predict and validate run at native image resolution regardless of size, padding only to the network's downsample factor. Only the l width is currently published, as a real-image denoising checkpoint trained on SIDD.

Train

NAFNet fine-tunes on your own paired degraded/clean images: a dataset YAML pointing at an inputs/<split>/ folder of degraded images and a targets/<split>/ folder of clean targets, matched by filename stem. degradation and dataset are optional strings recorded on the saved checkpoint for provenance; they take no part in training.

from libreyolo import LibreYOLO model = LibreYOLO("LibreNAFNetl-restore-sidd.pt")model.train(data="my-dataset.yaml", epochs=100, imgsz=256, batch=16, lr0=1e-3)

libreyolo train model=LibreNAFNetl-restore-sidd.pt data=my-dataset.yaml \  epochs=100 imgsz=256 batch=16 lr0=1e-3

from libreyolo import LibreYOLO # degradation and dataset are recorded on the saved checkpoint; they# don't change what is trained.model = LibreYOLO("LibreNAFNetl-restore-sidd.pt")model.train(    data="my-dataset.yaml",    epochs=100,    degradation="denoise",    dataset="MyDataset",)

libreyolo train model=LibreNAFNetl-restore-sidd.pt data=my-dataset.yaml \  epochs=100 device=0,1 batch=32

Left alone, the trainer runs 100 epochs with AdamW at lr0=1e-3, a batch of 16, 256 px crops, and early stopping after 50 epochs without PSNR improvement. There is no LoRA path for this family: lora=True raises an error rather than running, since NAFNetTrainer never opts in to adapter fine-tuning.

During training the network runs with plain global-average pooling. NAFNet's inference-only windowed local pooling (Test-time Local Converter) is detached before the first epoch and reattached once training finishes, since backpropagating through a fixed-window local pool would not match how the checkpoint is used at inference.

See training for datasets, augmentation, multi-GPU and loggers.

Validate

val() returns a dictionary with metrics/PSNR and metrics/SSIM, computed in RGB over the full valid canvas: SSIM uses an 11x11 Gaussian window with sigma 1.5, and fitness for best-checkpoint selection is the PSNR value. data points at the same paired-image dataset format used for training.

from libreyolo import LibreYOLO model = LibreYOLO("LibreNAFNetl-restore-sidd.pt") # val() returns a plain dict, not an objectmetrics = model.val(data="my-dataset.yaml") print(metrics["metrics/PSNR"])print(metrics["metrics/SSIM"])

libreyolo val model=LibreNAFNetl-restore-sidd.pt data=my-dataset.yaml

Export

An exported artifact loads back through LibreYOLO() on its file suffix, so a .onnx or .engine file behaves like a checkpoint and returns the same Results, with restored carrying the output image. NAFNet exports at a fixed spatial resolution: imgsz must be divisible by the network's downsample factor (16 for both architecture widths), and only the batch dimension is dynamic when dynamic=True; height and width are fixed at export time.

from libreyolo import LibreYOLO model = LibreYOLO("LibreNAFNetl-restore-sidd.pt")model.export(format="onnx", imgsz=256)model.export(format="tensorrt", imgsz=256, half=True)

libreyolo export model=LibreNAFNetl-restore-sidd.pt format=onnx imgsz=256libreyolo export model=LibreNAFNetl-restore-sidd.pt format=tensorrt imgsz=256 half=True

from libreyolo import LibreYOLO # The factory routes on the file suffix, so an exported artifact loads# like any checkpoint and returns the same Results object.model = LibreYOLO("LibreNAFNetl-restore-sidd.onnx")result = model("noisy.jpg") result.restored.save("denoised.png")

Checkpoints

Every published weight file for this family.

Licensing

Citation

@article{chen2022simple,
  title={Simple Baselines for Image Restoration},
  author={Chen, Liangyu and Chu, Xiaojie and Zhang, Xiangyu and Sun, Jian},
  journal={arXiv preprint arXiv:2204.04676},
  year={2022}
}