mlsanityData to 7 Oct 2026

See Blue Sky: Deep Image Dehaze Using Paired and Unpaired Training Images

arXiv 2210.07594 ↗First posted 14 Oct 2022cs.CV cs.LG cs.MMcited by 0

Not on the map. Why?

Abstract (arXiv)

The issue of image haze removal has attracted wide attention in recent years. However, most existing haze removal methods cannot restore the scene with clear blue sky, since the color and texture information of the object in the original haze image is insufficient. To remedy this, we propose a cycle generative adversarial network to construct a novel end-to-end image dehaze model. We adopt outdoor image datasets to train our model, which includes a set of real-world unpaired image dataset and a set of paired image dataset to ensure that the generated images are close to the real scene. Based on the cycle structure, our model adds four different kinds of loss function to constrain the effect including adversarial loss, cycle consistency loss, photorealism loss and paired L1 loss. These four constraints can improve the overall quality of such degraded images for better visual appeal and ensure reconstruction of images to keep from distortion. The proposed model could remove the haze of images and also restore the sky of images to be clean and blue (like captured in a sunny weather).

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 60 references that are in the map. First posted on arXiv 14 Oct 2022.

Builds on

25 of 60 references.

References · 25

  1. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  2. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  3. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,678
  4. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  5. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  6. 2016Context Encoders: Feature Learning by Inpaintingcited by 1,424
  7. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,187
  8. 2016Least Squares Generative Adversarial Networkscited by 1,000
  9. 2017Benchmarking Single Image Dehazing and Beyondcited by 445
  10. 2016DehazeNet: An End-to-End System for Single Image Haze Removalcited by 313
  11. 2017DualGAN: Unsupervised Dual Learning for Image-to-Image Translationcited by 310
  12. 2016Learning from Simulated and Unsupervised Images through Adversarial Trainingcited by 287
13 more
  1. 2017Learning to Discover Cross-Domain Relations with Generative Adversarial Networkscited by 278
  2. 2017Image De-raining Using a Conditional Generative Adversarial Networkcited by 197
  3. 2017Deep Photo Style Transfercited by 168
  4. 2018Densely Connected Pyramid Dehazing Networkcited by 145
  5. 2016Photo Aesthetics Ranking Network with Attributes and Content Adaptationcited by 122
  6. 2018O-HAZE: a dehazing benchmark with real hazy and haze-free outdoor imagescited by 95
  7. 2018Video Inpainting by Jointly Learning Temporal Structure and Spatial Detailscited by 57
  8. 2017Face Aging With Conditional Generative Adversarial Networkscited by 51
  9. 2018Fast End-to-End Trainable Guided Filtercited by 50
  10. 2017Unsupervised Diverse Colorization via Generative Adversarial Networkscited by 30
  11. 2017Joint Transmission Map Estimation and Dehazing using Deep Networkscited by 16
  12. 2020Single Image Dehazing Using Ranking Convolutional Neural Networkcited by 12
  13. 2017Style Transfer Generative Adversarial Networks: Learning to Play Chess Differentlycited by 2

Built on by

Not cited yet.