# Direct Intrinsics: Learning Albedo-Shading Decomposition by Convolutional Regression

arXiv: https://arxiv.org/abs/1512.02311 · HTML: https://mlsanity.com/paper/1512.02311/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2015.

## Built on by (most cited first, 30 of 33)

- [Intrinsic Image Decomposition via Ordinal Shading](https://mlsanity.com/paper/2311.12792/) (2023)
- [DIB-R++: Learning to Predict Lighting and Material with a Hybrid Differentiable Renderer](https://mlsanity.com/paper/2111.00140/) (2021)
- [A Survey on Intrinsic Images: Delving Deep Into Lambert and Beyond](https://mlsanity.com/paper/2112.03842/) (2021)
- [Material Palette: Extraction of Materials from a Single Image](https://mlsanity.com/paper/2311.17060/) (2023)
- [PIE-Net: Photometric Invariant Edge Guided Network for Intrinsic Image Decomposition](https://mlsanity.com/paper/2203.16670/) (2022)
- [Outdoor inverse rendering from a single image using multiview self-supervision](https://mlsanity.com/paper/2102.06591/) (2021)
- [Estimating Reflectance Layer from A Single Image: Integrating Reflectance Guidance and Shadow/Specular Aware Learning](https://mlsanity.com/paper/2211.14751/) (2022)
- [Measured Albedo in the Wild: Filling the Gap in Intrinsics Evaluation](https://mlsanity.com/paper/2306.15662/) (2023)
- [Weakly-supervised Single-view Image Relighting](https://mlsanity.com/paper/2303.13852/) (2023)
- [Lighting, Reflectance and Geometry Estimation from 360circ Panoramic Stereo](https://mlsanity.com/paper/2104.09886/) (2021)
- [DeRenderNet: Intrinsic Image Decomposition of Urban Scenes with Shape-(In)dependent Shading Rendering](https://mlsanity.com/paper/2104.13602/) (2021)
- [Unsupervised Intrinsic Image Decomposition with LiDAR Intensity](https://mlsanity.com/paper/2303.10820/) (2023)
- [Low-Light Video Enhancement via Spatial-Temporal Consistent Decomposition](https://mlsanity.com/paper/2405.15660/) (2024)
- [Physically Inspired Dense Fusion Networks for Relighting](https://mlsanity.com/paper/2105.02209/) (2021)
- [Joint Learning of Portrait Intrinsic Decomposition and Relighting](https://mlsanity.com/paper/2106.15305/) (2021)
- [Physics-based Indirect Illumination for Inverse Rendering](https://mlsanity.com/paper/2212.04705/) (2022)
- [DPF: Learning Dense Prediction Fields with Weak Supervision](https://mlsanity.com/paper/2303.16890/) (2023)
- [Benchmarking Ultra-High-Definition Image Reflection Removal](https://mlsanity.com/paper/2308.00265/) (2023)
- [Intrinsic Image Transfer for Illumination Manipulation](https://mlsanity.com/paper/2107.00704/) (2021)
- [SIGNet: Intrinsic Image Decomposition by a Semantic and Invariant Gradient Driven Network for Indoor Scenes](https://mlsanity.com/paper/2208.14369/) (2022)
- [Discriminative feature encoding for intrinsic image decomposition](https://mlsanity.com/paper/2209.12155/) (2022)
- [MVID: Feed-Forward Multi-View Intrinsic Image Decomposition](https://mlsanity.com/paper/2512.23667/) (2025)
- [ICTPolarReal: A Polarized Reflection and Material Dataset of Real World Objects](https://mlsanity.com/paper/2603.24912/) (2026)
- [Learning Intrinsic Images for Clothing](https://mlsanity.com/paper/2111.08521/) (2021)
- [Invariant Descriptors for Intrinsic Reflectance Optimization](https://mlsanity.com/paper/2204.04076/) (2022)
- [Intrinsic Image Decomposition Using Point Cloud Representation](https://mlsanity.com/paper/2307.10924/) (2023)
- [Unsupervised Intrinsic Image Decomposition with LiDAR Intensity Enhanced Training](https://mlsanity.com/paper/2403.14089/) (2024)
- [Intrinsic Image Decomposition for Robust Self-supervised Monocular Depth Estimation on Reflective Surfaces](https://mlsanity.com/paper/2503.22209/) (2025)
- [Adaptive Illumination Control for Robot Perception](https://mlsanity.com/paper/2602.15900/) (2026)
- [FreeLit: Paired-Free Indoor Relighting via Physics-Guided Diffusion](https://mlsanity.com/paper/2607.13656/) (2026)

## Abstract (arXiv)

We introduce a new approach to intrinsic image decomposition, the task of decomposing a single image into albedo and shading components. Our strategy, which we term direct intrinsics, is to learn a convolutional neural network (CNN) that directly predicts output albedo and shading channels from an input RGB image patch. Direct intrinsics is a departure from classical techniques for intrinsic image decomposition, which typically rely on physically-motivated priors and graph-based inference algorithms. The large-scale synthetic ground-truth of the MPI Sintel dataset plays a key role in training direct intrinsics. We demonstrate results on both the synthetic images of Sintel and the real images of the classic MIT intrinsic image dataset. On Sintel, direct intrinsics, using only RGB input, outperforms all prior work, including methods that rely on RGB+Depth input. Direct intrinsics also generalizes across modalities; it produces quite reasonable decompositions on the real images of the MIT dataset. Our results indicate that the marriage of CNNs with synthetic training data may be a powerful new technique for tackling classic problems in computer vision.