# An Optical physics inspired CNN approach for intrinsic image decomposition

arXiv: https://arxiv.org/abs/2105.10076 · HTML: https://mlsanity.com/paper/2105.10076/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 19 references that are in the map.
- First posted on arXiv 21 May 2021.

## Builds on (most cited first, 9 of 9)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Deep Retinex Decomposition for Low-Light Enhancement](https://mlsanity.com/paper/1808.04560/) (2018)
- [Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement](https://mlsanity.com/paper/2001.06826/) (2020)
- [CGIntrinsics: Better Intrinsic Image Decomposition through Physically-Based Rendering](https://mlsanity.com/paper/1808.08601/) (2018)
- [Attention Guided Low-light Image Enhancement with a Large Scale Low-light Simulation Dataset](https://mlsanity.com/paper/1908.00682/) (2019)
- [Self-Supervised Intrinsic Image Decomposition](https://mlsanity.com/paper/1711.03678/) (2017)
- [STAR: A Structure and Texture Aware Retinex Model](https://mlsanity.com/paper/1906.06690/) (2019)
- [Unsupervised Deep Single-Image Intrinsic Decomposition using Illumination-Varying Image Sequences](https://mlsanity.com/paper/1803.00805/) (2018)
- [A Retinex based GAN Pipeline to Utilize Paired and Unpaired Datasets for Enhancing Low Light Images](https://mlsanity.com/paper/2006.15304/) (2020)

## Abstract (arXiv)

Intrinsic Image Decomposition is an open problem of generating the constituents of an image. Generating reflectance and shading from a single image is a challenging task specifically when there is no ground truth. There is a lack of unsupervised learning approaches for decomposing an image into reflectance and shading using a single image. We propose a neural network architecture capable of this decomposition using physics-based parameters derived from the image. Through experimental results, we show that (a) the proposed methodology outperforms the existing deep learning-based IID techniques and (b) the derived parameters improve the efficacy significantly. We conclude with a closer analysis of the results (numerical and example images) showing several avenues for improvement.