I recently completed my PhD at Saarland University (Max Planck Institute for Informatics) under the supervision of Gurprit Singh. I am currently continuing as a postdoctoral researcher at the Max Planck Institute for Informatics with Karol Myszkowski.

My research focuses on Monte Carlo integration for rendering and machine learning, with a particular interest in variance reduction and sampling theory. More recently, I have been exploring how sampling correlations—both per-pixel and across the image domain—can be leveraged to improve real-time physically based rendering, and how perceptual modeling can further guide sampling to maximize visual quality.

My main research interests include:

• Blue-noise error distributions
• Multiple importance sampling
• Multi-class and structured sampling
• General variance-reduction techniques for Monte Carlo methods

Latest publications

Histogram Stratification for Spatio-Temporal Reservoir Sampling Corentin Salaün, Martin Bálint, Laurent Belcour, Eric Heitz, Gurprit Singh and Karol Myszkowski ACM Siggraph 2025 (Conference track) Published in SIGGRAPH 2025 Conference Papers, 2025 Recommended citation: Salaun, Corentin. (2025). "Histogram Stratification for Spatio-Temporal Reservoir Sampling" SIGGRAPH 2025 Conference Papers.

Online Importance Sampling for Stochastic Gradient Optimization Corentin Salaün, Xingchang Huang, Iliyan Georgiev, Niloy J. Mitra, Gurprit Singh ICPRAM 2025 (Full Paper) Best Student Paper Award Published in 14th International Conference on Pattern Recognition Applications and Methods, 2025 Recommended citation: Salaun, Corentin. (2025). "Online Importance Sampling for Stochastic Gradient Optimization" ICPRAM 2025.

Multiple Importance Sampling for Stochastic Gradient Estimation Corentin Salaün, Xingchang Huang, Iliyan Georgiev, Niloy J. Mitra, Gurprit Singh ICPRAM 2025 (Short Paper) Published in 14th International Conference on Pattern Recognition Applications and Methods, 2025 Recommended citation: Salaun, Corentin. (2025). "Multiple Importance Sampling for Stochastic Gradient Estimation" ICPRAM 2025.