# VIDIT: Virtual Image Dataset for Illumination Transfer

arXiv: https://arxiv.org/abs/2005.05460 · HTML: https://mlsanity.com/paper/2005.05460/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 May 2020.

## Built on by (most cited first, 26 of 26)

- [3D Common Corruptions and Data Augmentation](https://mlsanity.com/paper/2203.01441/) (2022)
- [Making Images Real Again: A Comprehensive Survey on Deep Image Composition](https://mlsanity.com/paper/2106.14490/) (2021)
- [NTIRE 2021 Depth Guided Image Relighting Challenge](https://mlsanity.com/paper/2104.13365/) (2021)
- [S3Net: A Single Stream Structure for Depth Guided Image Relighting](https://mlsanity.com/paper/2105.00681/) (2021)
- [Lasagna: Layered Score Distillation for Disentangled Object Relighting](https://mlsanity.com/paper/2312.00833/) (2023)
- [Multi-modal Bifurcated Network for Depth Guided Image Relighting](https://mlsanity.com/paper/2105.00690/) (2021)
- [Relighting from a Single Image: Datasets and Deep Intrinsic-based Architecture](https://mlsanity.com/paper/2409.18770/) (2024)
- [DSRN: an Efficient Deep Network for Image Relighting](https://mlsanity.com/paper/2102.09242/) (2021)
- [Multi-scale Self-calibrated Network for Image Light Source Transfer](https://mlsanity.com/paper/2104.08838/) (2021)
- [SILT: Self-supervised Lighting Transfer Using Implicit Image Decomposition](https://mlsanity.com/paper/2110.12914/) (2021)
- [Controllable Data Augmentation Through Deep Relighting](https://mlsanity.com/paper/2110.13996/) (2021)
- [Deep Portrait Delighting](https://mlsanity.com/paper/2203.12088/) (2022)
- [MultiPathGAN: Structure Preserving Stain Normalization using Unsupervised Multi-domain Adversarial Network with Perception Loss](https://mlsanity.com/paper/2204.09782/) (2022)
- [Computational Flash Photography through Intrinsics](https://mlsanity.com/paper/2306.06089/) (2023)
- [Physically Inspired Dense Fusion Networks for Relighting](https://mlsanity.com/paper/2105.02209/) (2021)
- [MSR-Net: Multi-Scale Relighting Network for One-to-One Relighting](https://mlsanity.com/paper/2107.06125/) (2021)
- [Designing An Illumination-Aware Network for Deep Image Relighting](https://mlsanity.com/paper/2207.10582/) (2022)
- [Local Relighting of Real Scenes](https://mlsanity.com/paper/2207.02774/) (2022)
- [SIRfyN: Single Image Relighting from your Neighbors](https://mlsanity.com/paper/2112.04497/) (2021)
- [Image color correction, enhancement, and editing](https://mlsanity.com/paper/2107.13117/) (2021)
- [A novel illumination condition varied image dataset-Food Vision Dataset (FVD) for fair and reliable consumer acceptability predictions from food](https://mlsanity.com/paper/2209.06967/) (2022)
- [Joint Shadow Generation and Relighting via Light-Geometry Interaction Maps](https://mlsanity.com/paper/2602.21820/) (2026)
- [PIXLRelight: Controllable Relighting via Intrinsic Conditioning](https://mlsanity.com/paper/2605.18735/) (2026)
- [Hidden-Shot: Towards One-Shot Task Generalization for Low-Level Vision Generalist Models](https://mlsanity.com/paper/2607.01535/) (2026)
- [Consistent Feature Transport for Image Relighting](https://mlsanity.com/paper/2607.17833/) (2026)
- [Geometric Similarity in VLM Low-Level Vision Representations](https://mlsanity.com/paper/2610.00848/) (2026)

## Abstract (arXiv)

Deep image relighting is gaining more interest lately, as it allows photo enhancement through illumination-specific retouching without human effort. Aside from aesthetic enhancement and photo montage, image relighting is valuable for domain adaptation, whether to augment datasets for training or to normalize input test data. Accurate relighting is, however, very challenging for various reasons, such as the difficulty in removing and recasting shadows and the modeling of different surfaces. We present a novel dataset, the Virtual Image Dataset for Illumination Transfer (VIDIT), in an effort to create a reference evaluation benchmark and to push forward the development of illumination manipulation methods. Virtual datasets are not only an important step towards achieving real-image performance but have also proven capable of improving training even when real datasets are possible to acquire and available. VIDIT contains 300 virtual scenes used for training, where every scene is captured 40 times in total: from 8 equally-spaced azimuthal angles, each lit with 5 different illuminants.