Imperfect Image-Space Control Variates for Monte Carlo Rendering

Chanu Yang and Bochang Moon
Gwangju Institute of Science and Technology (GIST)

Full-Resolution Results and Comparative Analysis

We include full-resolution images used in the comparisons with IDUW [Back et al. 2023] and CV (baseline), and control variates using polynomials, CV (polynomials) [Salaün et al. 2022]. The tested scenes come from a public scene repository [Bitterli 2016]. Please explore the results by clicking on an image below.

Comparisons with IDUW [Back et al. 2023] and CV (baseline)

Comparisons with CV (polynomials) [Salaün et al. 2022]

References

[Back et al. 2023] Jonghee Back, Binh-Son Hua, Toshiya Hachisuka, and Bochang Moon. 2023. Input-Dependent Uncorrelated Weighting for Monte Carlo Denoising. In SIGGRAPH Asia 2023 Conference Papers (Sydney, NSW, Australia) (SA ’23). Association for Computing Machinery, New York, NY, USA, Article 9, 10 pages.

[Bitterli 2016] Benedikt Bitterli. 2016. Rendering resources. https://benedikt-bitterli.me/resources/.

[Salaün et al. 2022] Corentin Salaün, Adrien Gruson, Binh-Son Hua, Toshiya Hachisuka, Gurprit Singh. 2022. Regression-based Monte Carlo Integration. ACM Trans. Graph. 41, 4, Article 79 (July 2022), 14 pages.

Acknowledgements

We would like to thank Joey Litalien, Jan Novák, and Benedikt Bitterli for the interactive viewer, as well as the following artists: nacimus (Bathroom), Wig42 (Breakfast Room; Grey & White Room; Staircase), NovaAshbell (Classroom), BhaWin (Glass ofWater), MrChimp2313 (House), and Jay-Artist (Kitchen).

License

All codes for the interactive viewer are licensed under the MIT License