A Review of Illumination Estimation in Augmented Reality

DOI: https://doi.org/10.62517/jbdc.202601304

Author(s)

Yang Lu1,*, Cong Lv2

Affiliation(s)

1Lecturer at Army Arms University, Nanjing, Jiangsu, China 2Staff Officer at Army Arms University, Nanjing, Jiangsu, China *Corresponding Author

Abstract

Illumination estimation is a core component of augmented reality and the key to achieving illumination consistency in virtual-real fusion; its implementation directly governs the visual credibility of virtual objects in real scenes. Most existing studies address illumination estimation in a single dimension, and no prior survey has systematically categorized methods across the three dimensions of light intensity, shadow, and fog concentration. This survey follows three main threads—light intensity estimation, shadow estimation, and fog concentration estimation—to classify and summarize the technical routes and representative works within each dimension. Our review identifies three notable gaps: hierarchical adaptation of illumination modes, decoupled processing of shadow direction and intensity, and the rendering integration of multi-dimensional parameters. A systematic understanding of these advances and limitations provides a comprehensive technical reference for future research on illumination estimation in virtual-real fusion scenarios.

Keywords

Augmented Reality; Illumination Estimation; Light Intensity Estimation; Shadow Estimation; Fog Concentration Estimation

References

[1] Ershov E, Savchik A, Semenkov I, et al. Illumination estimation challenge: The experience of the first 2 years. Color Research & Application, 2021, 46(4):705-718. [2] Büyükarıkan B, Ülker E. Using convolutional neural network models illumination estimation according to light colors. Optik, 2022, 271: 170002. [3] Einabadi F, Guillemaut J Y, Hilton A. Deep Neural Models for Illumination Estimation and Relighting: A Survey. Computer Graphics Forum, 2021, 40(6):315-336. [4] Zhang Z F, Huang C H, Huang R J, et al. Illu-NASNet: unsupervised illumination estimation based on dense spatio-temporal smoothness. The Visual Computer, 2023, 29(3):1413-1424. [5] Zhang Z F. Research on Deep Illumination Estimation Based on Image Intrinsic Property Decomposition. Guizhou University, 2023. [6] Zhu Y J. Illumination Estimation Based on Natural Scenes. Beijing University of Posts and Telecommunications, 2022. [7] Krawez M, Caselitz T, Sundram J, et al. Real-Time Outdoor Illumination Estimation for Camera Tracking in Indoor Environments. IEEE Robotics and Automation Letters, 2021, 6(3):6084-6091. [8] Wang M R. Research on Outdoor Scene Illumination Estimation Based on Neural Networks. Shandong University of Finance and Economics, 2022. [9] Ma F Y. Outdoor Illumination Estimation Algorithms and Their Application on Mobile Devices. Guangdong University of Technology, 2022. [10] Zhang K, Li X, Jin X, et al. Outdoor illumination estimation via all convolutional neural networks. Displays, 2021, 70:102089. [11] Zhao P, Su B, Chai R, et al. ReCC-GAN: scene global illuminant estimation network based on Retinex and generative adversarial network. Journal of Electronic Imaging, 2024, 33(2):023012. [12] Yu L J, Liu H J. Illumination Estimation Method Based on Linear Transformation Cosine Spherical Distribution. Journal of Northeastern University (Natural Science), 2025, 46(09):34-40. [13] DomislovićI, Vršnjak D, SubašićM,et al. Color constancy for non-uniform illumination estimation with variable number of illuminants. Neural Computing and Applications, 2023, 35(20):14573-14587. [14] Ai S C. Design and Implementation of a Mobile Augmented Reality System Based on Deep Learning Illumination Estimation. Southeast University, 2022. [15] Feng X, Li J, Hua Z, et al. Low-light image enhancement based on multi-illumination estimation. Applied Intelligence, 2021, 51(7):5211-5225. [16] Singh K, Parihar A S. Illumination estimation for nature preserving low-light image enhancement. The Visual Computer, 2024, 40(1):444-459. [17] Zhao Z, Lin H, Shi D, et al. A non-regularization self-supervised Retinex approach to low-light image enhancement with parameterized illumination estimation. Pattern Recognition, 2024, 146:110023. [18] Shen L, Wei W, Chen T. Low illumination image enhancement based on attention mechanism and global illumination estimation. Journal of Physics: Conference Series, 2023, 2589(1):012015. [19] Zhao R, Wang Z, Guo W, et al. Multi-scene image enhancement based on multi-channel illumination estimation. Displays, 2023, 226:104670. [20] Cao T C, Li X S, Li D, et al. Illumination Estimation Algorithm Based on Image Decomposition. Computer Engineering & Science, 2021, 43(08):1422-1428. [21] Xiao Z P. Research on Outdoor Scene Light Source Direction Estimation Method Based on Shadows. Hunan University, 2012. [22] Wen W, Wu X, Wan Y. Single-image shadow removal using detail extraction and illumination estimation. The Visual Computer, 2021, 38(5):1685-1696. [23] Wang W, Gong C, Huang S, et al. Dual-module uncalibrated photometric stereo with illumination estimation and robust normal regression. IOP Publishing, 2026, 8(8). [24] Wang L L, Yang Y. Image Dehazing Algorithm Combining Fog Concentration Segmentation and Atmospheric Light Veil Mapping. Chinese Journal of Liquid Crystals and Displays, 2025, 40(05):761-772. [25] Guo H. Fog Concentration Estimation and Inter-frame Smoothness Constrained Video Dehazing Algorithm. Beijing Forestry University, 2022. [26] Li Z C. Research on Texture Restoration Enhancement Assisted Smoke Concentration Estimation Network. Jiangxi University of Finance and Economics, 2025. [27] Xu C J. Research and Application of Fog Concentration Detection Algorithm for Waterway Channels in Scenic Areas. Xidian University, 2023. [28] Li C M, Jiang Y T, Song H P, et al. Fog Concentration Estimation and Image Dehazing Based on Optical Thickness Surrogate Model. Acta Armamentarii, 2019, 40(07):1425-1433. [29] Li J L, Qiu R X, Zhang J F, et al. Research on Image Clarity Enhancement Algorithm for Coal Mine Underground Mining Face Based on Dust-Fog Concentration Estimation. Henan Science and Technology, 2026, 53(02):13-20. [30] Gong X, Guan Y, Yang Y, et al. Zero-reference illumination estimation model for image enhancement in underground mines. Neural Networks, 2026, 174:106238.