Low light & Unpair training

• BIMEF(北大深研院)：A Bio-Inspired Multi-Exposure Fusion Framework for Low-light Image Enhancement Project

• Ying：A New Image Contrast Enhancement Algorithm Using Exposure Fusion Framework Project

• LIME：Low-light IMage Enhancement via Illumination Map Estimation Project

• NPE：Naturalness Preserved Enhancement Algorithm for Non-Uniform Illumination Images Project

• SRIE：A weighted variational model for simultaneous reflectance and illumination estimation Project

• MF：A Fusion-based Enhancing Method for Weakly Illuminated Images Projetc

• Deep Retinex Decomposition for Low Light Enhancement (BMVC18) Paper

• MBLLEN: Low Light Image/Video Enhancement using CNNs (BMVC18) Paper

• Learning to see in the dark (CVPR18) Paper Code

• Deep Photo Enhancer (CVPR18) Paper

• CycleGAN : Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (CVPR18) Paper Code

• AugCGAN: Augmented CycleGAN Paper

• WISPE: Weakly Supervised Photo Enhancer for Digital Cameras (CVPR18) Paper

• Image Blind Denoising With Generative Adversarial Network Based Noise Modeling (CVPR18) Paper

• NEXET vehicle dataset

  • link
  • quantity: Training 50K, Testing 41K
  • resolution: 1280x720
  • ratio: Day - Night (about 50% each)
  • scene: road, car

  Night	Day

• KAIST dataset

  • link
  • quantity: Training 50K, Testing 45K
  • ratio: Day - Night (about 60% - 40%)
  • resolution: 640x512
  • feature: with extra thermal annotation
  • scene: road, campus, downtown

  Night	Thermal	Day

• Adobe 5K

  • link
  • quantity: 5K
  • ratio: Day(100%)
  • resolution: 4000x2000
  • feature: high resolution, high quality day time pictures

  Day	Day

• Exclusively Dark Image Dataset

  • link
  • quantity: 7.3k
  • ratio: Night(100%)
  • feature: low light
  • scene: indoor, outdoor