Computer Graphics and Visualization • Publications

Computer Graphics Forum

A Survey on Gradient-Domain Rendering

B. S. Hua, A Gruson, Victor Petitjean, M Zwicker, D Nowrouzezahrai, Elmar Eisemann, and Toshiya Hachisuka

: Comparison between gradient-domain reconstruction and Monte Carlo denoising. For surface rendering, gradient-domain rendering is less efficient than Monte Carlo denoisers that use auxiliary buffers (NFOR [BRM∗ 16]) or histograms of path samples (BCD [BB17]). NFOR could also be applied to address noisy regions remaining in gradient-domain path tracing by using the reconstructed image as guiding features, leading to improved image quality (see G-PT + NFOR in the KITCHEN scene). For volume rendering, gradient-domain rendering is comparable to Monte Carlo denoisers, particularly with photon density estimation.

Monte Carlo methods for physically-based light transport simulation are broadly adopted in film productions and animation/visual effects industries. Monte Carlo methods, however, often result in noisy images and have slow convergence. As such, improving the convergence of Monte Carlo rendering remains an important open problem. Gradient-domain light transport is a recent family of techniques that can accelerate Monte Carlo rendering by up to an order of magnitude, leveraging a gradient-based estimation and a reformulation of the rendering problem as an image reconstruction. This state of the art report comprehensively elaborates the fundamentals of gradient-domain rendering, as well as the specifics behind gradient-domain uni- and bidirectional path tracing and gradient-domain photon density estimation. We also discuss various image reconstruction schemes which are important components of gradient-domain rendering. Finally, we benchmark various gradient-domain techniques against state-of-the-art denoising methods before discussing open problems.

More Information

Publisher Paper preprint

Citation

B. S. Hua, A Gruson, Victor Petitjean, M Zwicker, D Nowrouzezahrai, Elmar Eisemann, and Toshiya Hachisuka, A Survey on Gradient-Domain Rendering, Computer Graphics Forum, 38, pp. 455–472, 2019.

BibTex

@article{bib:hua:2019,
    author       = { Hua, B. S. and Gruson, A and Petitjean, Victor and Zwicker, M and Nowrouzezahrai, D and Eisemann, Elmar and Hachisuka, Toshiya  },    
    title        = { A Survey on Gradient-Domain Rendering },
    journal      = { Computer Graphics Forum },
    volume       = { 38 },
    year         = { 2019 },
    pages        = { 455--472 },
    doi          = { 10.1111/cgf.13652 },
    dblp         = { journals/cgf/HuaGPZNEH19 },
    url          = { https://publications.graphics.tudelft.nl/papers/103 },
}