Supplementary Material for Self-Supervised Post-Correction for Monte Carlo Denoising

Jonghee Back1, Binh-Son Hua2, Toshiya Hachisuka3, and Bochang Moon1

1 GIST, South Korea, 2 VinAI Research, Vietnam, 3 University of Waterloo, Canada


This supplementary material illustrates equal-time comparisons between recent Monte Carlo (MC) image denoising methods and their corrected results using our method.

We test three denoising methods as our correlated images: KPCN [Bako et al. 2017], AMCD [Xu et al. 2019] and AFGSA [Yu et al. 2021].

We also compare deep combiner (DC) [Back et al. 2020] using retrained models with the results corrected by our approach given a same running time.

In the tables, we provide the total running time, samples per pixel (spp) and numerical accuracy (relative L2 [Rousselle et al. 2011] with small value 1e-2) for each result.

Equal-Time Comparisons between MC Image Denoising Methods and Our Post-Correction








Equal-Time Comparisons between DC and Our Post-Correction






[Bako et al. 2017] Steve Bako, Thijs Vogels, Brian Mcwilliams, Mark Meyer, Jan Novák, Alex Harvill, Pradeep Sen, Tony Derose, and Fabrice Rousselle. 2017. Kernel-Predicting Convolutional Networks for Denoising Monte Carlo Renderings. ACM Trans. Graph. 36, 4, Article 97 (2017), 14 pages

[Xu et al. 2019] Bing Xu, Junfei Zhang, Rui Wang, Kun Xu, Yong-Liang Yang, Chuan Li, and Rui Tang. 2019. Adversarial Monte Carlo Denoising with Conditioned Auxiliary Feature Modulation. ACM Trans. Graph. 38, 6, Article 224 (2019), 12 pages.

[Yu et al. 2021] Jiaqi Yu, Yongwei Nie, Chengjiang Long, Wenju Xu, Qing Zhang, and Guiqing Li. 2021. Monte Carlo Denoising via Auxiliary Feature Guided Self-Attention. ACM Trans. Graph. 40, 6, Article 273 (2021), 13 pages.

[Back et al. 2020] Jonghee Back, Binh-Son Hua, Toshiya Hachisuka, and Bochang Moon. 2020. Deep Combiner for Independent and Correlated Pixel Estimates. ACM Trans. Graph. 39, 6, Article 242 (2020), 12 pages.

[Rousselle et al. 2011] Fabrice Rousselle, Claude Knaus, and Matthias Zwicker. 2011. Adaptive Sampling and Reconstruction Using Greedy Error Minimization. ACM Trans. Graph. 30, 6, Article 159 (2011), 12 pages


We'd like to thank nacimus (Bathroom), Cem Yuksel (Hair), Christian Schüller (Dragon), Guillermo M. Leal Llaguno (Sanmiguel), julioras3d (Chopper-titan), bluewhalestudios (Hotel), Jan-Walter Schliep, Burak Kahraman and Timm Dapper (Landscape) for providing the following test scenes and Joey Litalien, Jan Novák and Benedikt Bitterli for the interactive viewer.