RECNet

用于混合曝光校正的区域感知曝光一致性网络 (AAAI 2024)

---

一、简介

该存储库是 RECNet 的官方实现，其中提供了更多实现细节。

2. 要求

 python=3.8.16
pytorch=2.0.1
torchvision=0.8
cuda=11.7
opencv-python

3. 数据集准备

详情请参阅 ENC。

4. 测试

 bash scripts/test.sh

5. 培训

 bash scripts/train.sh

---

引文

如果您发现这项工作对您的研究有用，请考虑引用：

 @inproceedings{DBLP:conf/aaai/LiuFWM24,
  author       = {Jin Liu and
                  Huiyuan Fu and
                  Chuanming Wang and
                  Huadong Ma},
  editor       = {Michael J. Wooldridge and
                  Jennifer G. Dy and
                  Sriraam Natarajan},
  title        = {Region-Aware Exposure Consistency Network for Mixed Exposure Correction},
  booktitle    = {Thirty-Eighth {AAAI} Conference on Artificial Intelligence, {AAAI}
                  2024, Thirty-Sixth Conference on Innovative Applications of Artificial
                  Intelligence, {IAAI} 2024, Fourteenth Symposium on Educational Advances
                  in Artificial Intelligence, {EAAI} 2014, February 20-27, 2024, Vancouver,
                  Canada},
  pages        = {3648--3656},
  publisher    = {{AAAI} Press},
  year         = {2024},
  url          = {https://doi.org/10.1609/aaai.v38i4.28154},
  doi          = {10.1609/AAAI.V38I4.28154}
}

致谢

该存储库基于 ENC - 特别感谢他们的代码！