Learning Deep Priors for Image Dehazing


Yang Liu          Jinshan Pan          Jimmy Ren          Zhixun Su

Abstract


Image dehazing is a well-known ill-posed problem, which usually requires some image priors to make the problem well-posed. We propose an effective iteration algorithm with deep CNNs to learn haze-relevant priors for image dehazing. We formulate the image dehazing problem as the minimization of a variational model with favorable data fidelity terms and prior terms to regularize the model. We solve the variational model based on the classical gradient descent method with built-in deep CNNs so that iteration-wise image priors for the atmospheric light, transmission map and clear image can be well estimated. Our method combines the properties of both the physical formation of image dehazing as well as deep learning approaches. We show that it is able to generate clear images as well as accurate atmospheric light and transmission maps. Extensive experimental results demonstrate that the proposed algorithm performs favorably against state-of-the-art methods in both benchmark datasets and real-world images.


An overview of the proposed method. Our method takes an iterative optimization scheme with an effective neural network-based learning manner.

Technical Papers, Codes, and Datasets

Yang Liu, Jinshan Pan, Jimmy Ren, and Zhixun Su, "Learning Deep Priors for Image Dehazing", IEEE Conference on Computer Vision (ICCV), 2019.
[Project] [Paper] [Code] [Datasets] 

          @InProceedings{Liu_2019_ICCV,
          author = {Liu, Yang and Pan, Jinshan and Ren, Jimmy and Su, Zhixun},
          title = {Learning Deep Priors for Image Dehazing},
          booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
          month = {October},
          year = {2019}
          }

Experimental Results

1. Image Dehazing on the Proposed Synthetic Dataset


2. Image Dehazing on the SOTS Dataset


3. Image Dehazing on the HazeRD Dataset


4. Image Dehazing on Real-world Images


5. Image Deraining on Real-world Images


Contact

If you have any question, please contact Yang Liu at lewisyangliu@gmail.com.

References

[1] Ancuti C, Ancuti CO, Timofte R, De Vleeschouwer C (2018a) I-HAZE - A Dehazing Benchmark with Real Hazy and Haze-Free Indoor Images. ACIVS 11182(8):620–631

[2] Ancuti CO, Ancuti C, Timofte R, De Vleeschouwer C (2018b) O- HAZE - A Dehazing Benchmark With Real Hazy and Haze-Free Outdoor Images. In: CVPR Workshops pp 867–8678

[3] Berman D, Treibitz T, Avidan S (2016) Non-local Image Dehazing. In: CVPR, pp 1628–1636

[4] Cai B, Xu X, Jia K, Qing C, Tao D (2016) DehazeNet - An End-to-End System for Single Image Haze Removal. IEEE TIP 25(11):5187– 5198

[5] Fattal R (2014) Dehazing Using Color-Lines. ACM Trans Graph 34(1):1–14

[6] Fu X, Huang J, Ding X, Liao Y, Paisley JW (2017a) Clearing the Skies - A Deep Network Architecture for Single-Image Rain Removal. IEEE Trans Image Process 26(6):2944–2956

[7] Fu X, Huang J, Zeng D, Huang Y, Ding X, Paisley JW (2017b) Remov- ing Rain from Single Images via a Deep Detail Network. In: CVPR pp 1715–1723

[8] He K, Sun J, Tang X (2009) Single image haze removal using dark channel prior. In: CVPR, pp 1956–1963

[9] Li B, Peng X, Wang Z, Xu J, Feng D (2017a) AOD-Net - All-in-One Dehazing Network. In: ICCV, pp 4780–4788

[10] Li B, Ren W, Fu D, Tao D, Feng D, Zeng W, Wang Z (2017b) Bench- marking Single Image Dehazing and Beyond. arXiv 1712.04143

[11] Li Y, Tan RT, Guo X, Lu J, Brown MS (2016) Rain Streak Removal Using Layer Priors. In: CVPR, pp 2736–2744

[12] Ren D, Zuo W, Hu Q, Zhu P, Meng D (2019) Progressive Image De- raining Networks: A Better and Simpler Baseline. In: CVPR, pp 3937–3946

[13] Ren W, Liu S, Zhang H, Pan J, Cao X, Yang MH (2016) Single Image Dehazing via Multi-scale Convolutional Neural Networks. In: ECCV, pp 154–169

[14] Ren W, Ma L, Zhang J, Pan J, Cao X, Liu W, Yang MH (2018) Gated Fusion Network for Single Image Dehazing. In: CVPR, pp 3253–3261

[15] Saxena A, Chung SH, Ng AY (2005) Learning Depth from Single Monocular Images. In: NIPS, pp 1161–1168

[16] Saxena A, Chung SH, Ng AY (2008) 3-D Depth Reconstruction from a Single Still Image. IJCV 76(1):53–69

[17] Saxena A, Sun M, Ng AY (2009) Make3D: Learning 3D Scene Struc- ture from a Single Still Image. IEEE TPAMI 31(5):824–840

[18] Silberman N, Hoiem D, Kohli P, Fergus R (2012) Indoor Segmentation and Support Inference from RGBD Images. In: ECCV, pp 746–760

[19] Wang H, Xie Q, Zhao Q, Meng D (2020) A Model-driven Deep Neural Network for Single Image Rain Removal. CVPR pp 3103–3112

[20] Yang D, Sun J (2018) Proximal Dehaze-Net: A Prior Learning-Based Deep Network for Single Image Dehazing. In: ECCV, pp 702–717

[21] Yang W, Tan RT, Feng J, Liu J, Guo Z, Yan S (2017) Deep Joint Rain Detection and Removal from a Single Image. CVPR pp 1685–1694

[22] Yasarla R, Patel VM (2019) Uncertainty Guided Multi-Scale Residual Learning-using a Cycle Spinning CNN for Single Image DeRaining. In: CVPR, pp 8405–8414

[23] Zhang H, Patel VM (2018a) Densely Connected Pyramid Dehazing Network. In: CVPR, pp 3194–3203

[24] Zhang H, Patel VM (2018b) Density-aware Single Image De-raining using a Multi-stream Dense Network. In: CVPR, pp 695–704

[25] Zhang Y, Ding L, Sharma G (2017) HazeRD - An outdoor scene dataset and benchmark for single image dehazing. In: ICIP, pp 3205–3209

[26] Zhu Q, Mai J, Shao L (2015) A Fast Single Image Haze Removal Algo- rithm Using Color Attenuation Prior. IEEE TIP 24(11):3522–3533