Update README.md

README.md

@@ -1,48 +1,94 @@
-# Exclusively Dark (ExDark) Image Dataset (Official Site)
-[ArXiv](http://cs-chan.com/doc/cviu.pdf)
+---
+pretty_name: DGFF (Detection-Guided Feature Feedback)
+license: mit
+language:
+  - en
+task_categories:
+  - object-detection
+  - image-to-image
+tags:
+  - low-light
+  - image-enhancement
+  - object-detection
+  - feature-feedback
+  - yolov8
+  - unet
+size_categories:
+  - n<1K
+---
 
-Updated on Sept 02, 2022 (Updated the dataset link)
+# DGFF: Detection-Guided Feature Feedback
 
-Updated on June 02, 2019 (Code for low-light image enhancement is released)
+Companion repository for **"End-to-End Task-Oriented Low-Light Image Enhancement via Detection-Guided Feature Feedback"** (ICCA 2026).
 
-Updated on Oct. 31, 2018 (Accepted for publication in CVIU)
+> The detector teaches the enhancer what to preserve, and the gated adapters learn to pass only the useful lessons.
 
-Released on May 29, 2018
+| Resource | Link |
+|---|---|
+| Code | <https://github.com/ShuvroSankar/DGFF> |
+| Data (LOL + ExDark) | <https://huggingface.co/datasets/abbaab/DGFF-dataset> |
 
-## Description
+## About DGFF
 
-In order to facilitate a new object detection and image enhancement research particularly in the low-light environment, we introduce the Exclusively Dark (ExDark) dataset [(CVIU2019)](http://cs-chan.com/doc/cviu.pdf). The Exclusively Dark (ExDARK) dataset is a collection of 7,363 low-light images from very low-light environments to twilight (i.e 10 different conditions) with 12 object classes (similar to PASCAL VOC) annotated on both image class level and local object bounding boxes. 
+DGFF is a training framework that routes intermediate **YOLOv8n** backbone features (P3, P4, P5) back into the decoder of a U-Net low-light enhancer (**LLEN**) through lightweight **gated 1x1 adapters**. The detector's view of what matters shapes the reconstruction, so enhancement preserves edges and textures that detectors rely on. The adapters are discarded at inference, so DGFF adds **zero overhead** over a plain LLEN + detector pipeline.
 
-![demo](Exdark.gif)
+| Component | Parameters |
+|---|---|
+| LLEN (U-Net) | 4,844,803 |
+| DGFF adapters (training only) | 86,464 |
+| YOLOv8n backbone (frozen) | 1,272,656 |
 
-## Source Code 
-The source code for our work on low-light image enhancement is now available. Please refer to the [SPIC](https://github.com/cs-chan/Exclusively-Dark-Image-Dataset/tree/master/SPIC) folder.
+## Headline results
 
-<img src="SPIC/2015_00003.gif" height="150" > <img src="SPIC/2015_02446.gif" height="150" > <img src="SPIC/2015_06400.gif" height="150" >
+| Benchmark | Metric | Result |
+|---|---|---|
+| LOL (validation) | PSNR / SSIM | 19.83 dB / 0.9048 |
+| ExDark (domain-matched) | mAP@0.5 | 0.474 (raw baseline 0.408, LLEN-only 0.469) |
+| ExDark (domain-matched) | mAP@0.5:0.95 | 0.280 (raw baseline 0.232, LLEN-only 0.276) |
 
+The +0.5-point mAP@0.5 gain over LLEN-only is directionally consistent but not statistically significant (paired bootstrap, 95% CI [-0.0065, 0.0112]). The +6.6-point gain over the unenhanced baseline is clear. Full details are in the paper and the GitHub README.
+
+## Contents
+
+<!-- TODO: list the files actually stored in this repository (the repo currently holds only ~2.5 kB). Example table below. -->
+
+| File | Description |
+|---|---|
+| `README.md` | This card |
+| *(to add)* | LLEN checkpoint after Phase 1 (LOL, best epoch 185) |
+| *(to add)* | DGFF adapter weights after Phase 2 (ExDark fine-tuning) |
+
+## Access
+
+This repository is **gated**: you must agree to share your contact information to access its files. It is intended for research use.
+
+## How to use
+
+1. Clone the code: `git clone https://github.com/ShuvroSankar/DGFF.git`
+2. Download the data from [abbaab/DGFF-dataset](https://huggingface.co/datasets/abbaab/DGFF-dataset).
+3. Follow the training and evaluation steps in the GitHub README.
+
+At inference only LLEN is needed: enhance the image, then run any detector (the paper uses YOLOv8n).
+
+## Limitations
+
+- Trained on LOL (indoor, 485 pairs) and fine-tuned on ExDark, so an indoor/outdoor domain gap remains.
+- Evaluated with YOLOv8n only; other detectors are untested.
+- Classes with small, occluded or homogeneous-texture objects (Chair, Motorbike, People, Table) did not improve over LLEN-only.
 
 ## Citation
-If you find this dataset useful for your research, please cite
+
 ```bibtex
-@article{Exdark,
-  title = {Getting to Know Low-light Images with The Exclusively Dark Dataset},
-  author = {Loh, Yuen Peng and Chan, Chee Seng},
-  journal = {Computer Vision and Image Understanding},
-  volume = {178},
-  pages = {30-42},
-  year = {2019},
-  doi = {https://doi.org/10.1016/j.cviu.2018.10.010}
+@inproceedings{sen2026dgff,
+  title     = {End-to-End Task-Oriented Low-Light Image Enhancement via Detection-Guided Feature Feedback},
+  author    = {Sen, Shuvro Sankar and Mia, MD. Maruf and Hasan, Naim and Chayon, Muhammad Hasibur Rashid},
+  booktitle = {Proceedings of The 4th International Conference on Computing Advancements (ICCA 2026)},
+  year      = {2026},
+  address   = {Dhaka, Bangladesh},
+  publisher = {ACM}
 }
 ```
 
-## Feedback
-Suggestions and opinions on this dataset (both positive and negative) are greatly welcomed. Please contact the authors by sending an email to
-`lexloh2009 at hotmail.com` or `cs.chan at um.edu.my`.
-
-## License and Copyright
-The project is open source under BSD-3 license (see the ``` LICENSE ``` file). 
-
-For commercial purpose usage, please contact Dr. Chee Seng Chan at `cs.chan at um.edu.my`
+## Authors
 
-&#169;2018-2022 Center of Image and Signal Processing, Faculty of Computer Science and Information Technology, Universiti Malaya.
-# DGFF
+Shuvro Sankar Sen, MD. Maruf Mia, Naim Hasan, Muhammad Hasibur Rashid Chayon, American International University - Bangladesh (AIUB).