Update README.md
Browse files
README.md
CHANGED
|
@@ -50,6 +50,49 @@ model-index:
|
|
| 50 |
|
| 51 |
---
|
| 52 |
|
| 53 |
-
|
| 54 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
Code: https://github.com/albrateanu/LYT-Net
|
|
|
|
| 50 |
|
| 51 |
---
|
| 52 |
|
| 53 |
+
# [SPL 2025] LYT-Net: Lightweight YUV Transformer-based Network for Low-Light Image Enhancement
|
| 54 |
|
| 55 |
+
<div align="center">
|
| 56 |
+
|
| 57 |
+
[](https://arxiv.org/abs/2401.15204)
|
| 58 |
+
[](https://ieeexplore.ieee.org/abstract/document/10972228)
|
| 59 |
+
|
| 60 |
+
Ranked #1 on FLOPS(G) (3.49 GFLOPS) and Params(M) (0.045M = 45k Params)
|
| 61 |
+
</div>
|
| 62 |
+
|
| 63 |
+
## π Updates
|
| 64 |
+
- `27.07.2025` π€ LYT-Net now has a new HuggingFace page! Check it out [here](https://huggingface.co/albrateanu/LYT-Net)! **HF Demo coming soon!**
|
| 65 |
+
- `09.05.2025` π’ Check out our other works on [Low-light Image Enhancement](https://github.com/albrateanu/KANT) and [Image Denoising](https://github.com/albrateanu/AKDT)!
|
| 66 |
+
- `21.04.2025` π LYT-Net is published as a IEEE Signal Processing Letters paper. [Link to paper](https://ieeexplore.ieee.org/abstract/document/10972228).
|
| 67 |
+
- `17.07.2024` π§ͺ Released rudimentary PyTorch implementation.
|
| 68 |
+
- `03.04.2024` π§ Training code re-added and adjusted.
|
| 69 |
+
- `30.01.2024` π arXiv pre-print available.
|
| 70 |
+
- `10.01.2024` π Pre-trained model weights and code for training and testing are released.
|
| 71 |
+
|
| 72 |
+
## π§ͺ Experiment
|
| 73 |
+
Please check the [GitHub](https://github.com/albrateanu/LYT-Net) for PyTorch and TensorFlow implementations.
|
| 74 |
+
|
| 75 |
+
## π Citation
|
| 76 |
+
```
|
| 77 |
+
@article{brateanu2025lyt,
|
| 78 |
+
author={Brateanu, Alexandru and Balmez, Raul and Avram, Adrian and Orhei, Ciprian and Ancuti, Cosmin},
|
| 79 |
+
journal={IEEE Signal Processing Letters},
|
| 80 |
+
title={LYT-NET: Lightweight YUV Transformer-based Network for Low-light Image Enhancement},
|
| 81 |
+
year={2025},
|
| 82 |
+
volume={},
|
| 83 |
+
number={},
|
| 84 |
+
pages={1-5},
|
| 85 |
+
doi={10.1109/LSP.2025.3563125}}
|
| 86 |
+
|
| 87 |
+
|
| 88 |
+
@article{brateanu2024lyt,
|
| 89 |
+
title={LYT-Net: Lightweight YUV Transformer-based Network for Low-Light Image Enhancement},
|
| 90 |
+
author={Brateanu, Alexandru and Balmez, Raul and Avram, Adrian and Orhei, Ciprian and Cosmin, Ancuti},
|
| 91 |
+
journal={arXiv preprint arXiv:2401.15204},
|
| 92 |
+
year={2024}
|
| 93 |
+
}
|
| 94 |
+
```
|
| 95 |
+
|
| 96 |
+
|
| 97 |
+
Paper: arxiv.org/abs/2401.15204
|
| 98 |
Code: https://github.com/albrateanu/LYT-Net
|