Tampere University of Technology

TUTCRIS Research Portal

Blockwise Multi-Order Feature Regression for Real-Time Path Tracing Reconstruction

Research output: Contribution to journalArticleScientificpeer-review

Standard

Blockwise Multi-Order Feature Regression for Real-Time Path Tracing Reconstruction. / Koskela, Matias; Immonen, Kalle; Mäkitalo, Markku; Foi, Alessandro; Viitanen, Timo; Jääskeläinen, Pekka; Kultala, Heikki; Takala, Jarmo.

In: ACM Transactions on Graphics, Vol. 38, No. 5, 138, 06.2019.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Koskela, Matias ; Immonen, Kalle ; Mäkitalo, Markku ; Foi, Alessandro ; Viitanen, Timo ; Jääskeläinen, Pekka ; Kultala, Heikki ; Takala, Jarmo. / Blockwise Multi-Order Feature Regression for Real-Time Path Tracing Reconstruction. In: ACM Transactions on Graphics. 2019 ; Vol. 38, No. 5.

Bibtex - Download

@article{8757552c36544193a6c361e3419c04d4,
title = "Blockwise Multi-Order Feature Regression for Real-Time Path Tracing Reconstruction",
abstract = "Path tracing produces realistic results including global illumination using a unified simple rendering pipeline. Reducing the amount of noise to imperceptible levels without post-processing requires thousands of samples per pixel (spp), while currently it is only possible to render extremely noisy 1 spp frames in real time with desktop GPUs. However, post-processing can utilize feature buffers, which contain noise-free auxiliary data available in the rendering pipeline. Previously, regression-based noise filtering methods have only been used in offline rendering due to their high computational cost. In this paper we propose a novel regression-based reconstruction pipeline, called Blockwise Multi-Order Feature Regression (BMFR), tailored for path-traced 1 spp inputs that runs in real time. The high speed is achieved with a fast implementation of augmented QR factorization and by using stochastic regularization to address rank-deficient feature data. The proposed algorithm is 1.8× faster than the previous state-of-the-art real-time path tracing reconstruction method while producing better quality frame sequences.",
author = "Matias Koskela and Kalle Immonen and Markku M{\"a}kitalo and Alessandro Foi and Timo Viitanen and Pekka J{\"a}{\"a}skel{\"a}inen and Heikki Kultala and Jarmo Takala",
year = "2019",
month = "6",
doi = "10.1145/3269978",
language = "English",
volume = "38",
journal = "ACM Transactions on Graphics",
issn = "0730-0301",
publisher = "Association for Computing Machinery",
number = "5",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Blockwise Multi-Order Feature Regression for Real-Time Path Tracing Reconstruction

AU - Koskela, Matias

AU - Immonen, Kalle

AU - Mäkitalo, Markku

AU - Foi, Alessandro

AU - Viitanen, Timo

AU - Jääskeläinen, Pekka

AU - Kultala, Heikki

AU - Takala, Jarmo

PY - 2019/6

Y1 - 2019/6

N2 - Path tracing produces realistic results including global illumination using a unified simple rendering pipeline. Reducing the amount of noise to imperceptible levels without post-processing requires thousands of samples per pixel (spp), while currently it is only possible to render extremely noisy 1 spp frames in real time with desktop GPUs. However, post-processing can utilize feature buffers, which contain noise-free auxiliary data available in the rendering pipeline. Previously, regression-based noise filtering methods have only been used in offline rendering due to their high computational cost. In this paper we propose a novel regression-based reconstruction pipeline, called Blockwise Multi-Order Feature Regression (BMFR), tailored for path-traced 1 spp inputs that runs in real time. The high speed is achieved with a fast implementation of augmented QR factorization and by using stochastic regularization to address rank-deficient feature data. The proposed algorithm is 1.8× faster than the previous state-of-the-art real-time path tracing reconstruction method while producing better quality frame sequences.

AB - Path tracing produces realistic results including global illumination using a unified simple rendering pipeline. Reducing the amount of noise to imperceptible levels without post-processing requires thousands of samples per pixel (spp), while currently it is only possible to render extremely noisy 1 spp frames in real time with desktop GPUs. However, post-processing can utilize feature buffers, which contain noise-free auxiliary data available in the rendering pipeline. Previously, regression-based noise filtering methods have only been used in offline rendering due to their high computational cost. In this paper we propose a novel regression-based reconstruction pipeline, called Blockwise Multi-Order Feature Regression (BMFR), tailored for path-traced 1 spp inputs that runs in real time. The high speed is achieved with a fast implementation of augmented QR factorization and by using stochastic regularization to address rank-deficient feature data. The proposed algorithm is 1.8× faster than the previous state-of-the-art real-time path tracing reconstruction method while producing better quality frame sequences.

U2 - 10.1145/3269978

DO - 10.1145/3269978

M3 - Article

VL - 38

JO - ACM Transactions on Graphics

JF - ACM Transactions on Graphics

SN - 0730-0301

IS - 5

M1 - 138

ER -