Unsupervised Learning for Intrinsic Image Decomposition from a Single Image

TL;DR

This paper introduces an unsupervised framework for intrinsic image decomposition that learns reflectance and shading without labeled data or handcrafted priors, demonstrating superior results on synthetic and real datasets.

Contribution

It presents a novel unsupervised approach that leverages independence and physical constraints to decompose images into reflectance and shading without supervision.

Findings

Outperforms existing methods on synthetic datasets

Achieves superior results on real-world images

Effectively learns latent features of reflectance and shading

Abstract

Intrinsic image decomposition, which is an essential task in computer vision, aims to infer the reflectance and shading of the scene. It is challenging since it needs to separate one image into two components. To tackle this, conventional methods introduce various priors to constrain the solution, yet with limited performance. Meanwhile, the problem is typically solved by supervised learning methods, which is actually not an ideal solution since obtaining ground truth reflectance and shading for massive general natural scenes is challenging and even impossible. In this paper, we propose a novel unsupervised intrinsic image decomposition framework, which relies on neither labeled training data nor hand-crafted priors. Instead, it directly learns the latent feature of reflectance and shading from unsupervised and uncorrelated data. To enable this, we explore the independence between reflectance and shading, the domain invariant content constraint and the physical constraint. Extensive experiments on both synthetic and real image datasets demonstrate consistently superior performance of the proposed method.