# Learning-Based Dequantization For Image Restoration Against Extremely Poor Illumination

arXiv: https://arxiv.org/abs/1803.01532 · HTML: https://mlsanity.com/paper/1803.01532/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Mar 2018.

## Built on by (most cited first, 1 of 1)

- [HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization](https://mlsanity.com/paper/2105.13084/) (2021)

## Abstract (arXiv)

All existing image enhancement methods, such as HDR tone mapping, cannot recover A/D quantization losses due to insufficient or excessive lighting, (underflow and overflow problems). The loss of image details due to A/D quantization is complete and it cannot be recovered by traditional image processing methods, but the modern data-driven machine learning approach offers a much needed cure to the problem. In this work we propose a novel approach to restore and enhance images acquired in low and uneven lighting. First, the ill illumination is algorithmically compensated by emulating the effects of artificial supplementary lighting. Then a DCNN trained using only synthetic data recovers the missing detail caused by quantization.