Ranked #1 on FLOPS(G) (3.49 GFLOPS) and Params(M) (0.045M = 45k Params)
17.07.2024Released rudimentary PyTorch implementation.03.04.2024Training code re-added and adjusted.30.01.2024arXiv pre-print available.10.01.2024Pre-trained model weights and code for training and testing are released.
Please check the TensorFlow and PyTorch folders for library-specific implementations.
| Dataset | TensorFlow | PyTorch | ||
|---|---|---|---|---|
| PSNR | SSIM | PSNR | SSIM | |
| LOLv1 | 27.23 | 0.853 | 26.63 | 0.836 |
| LOLv2-R | 27.80 | 0.873 | 28.41 | 0.878 |
| LOLv2-S | 29.39 | 0.939 | 26.72 | 0.928 |
Preprint Citation
@article{brateanu2024,
title={LYT-Net: Lightweight YUV Transformer-based Network for Low-Light Image Enhancement},
author={Brateanu, Alexandru and Balmez, Raul and Avram, Adrian and Orhei, Ciprian},
journal={arXiv preprint arXiv:2401.15204},
year={2024}
}