Image-space control variates for rendering

In ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia), 2016

Image-space control variates allow leveraging coherence in renderings. We show here an example of our re-rendering application, leveraging temporal coherence. We used 1024/64 samples per pixel for rendering the control/difference images, and our final reconstruction (Ours, far right) offers a significant improvement over standard Path tracing, despite the magnitude of the changes.

Abstract

We explore the theory of integration with control variates in the context of rendering. Our goal is to optimally combine multiple estimators using their covariances. We focus on two applications, re-rendering and gradient-domain rendering, where we exploit coherence between temporally and spatially adjacent pixels. We propose an image-space (iterative) reconstruction scheme that employs control variates to reduce variance. We show that recent works on scene editing and gradient-domain rendering can be directly formulated as control-variate estimators, despite using seemingly different approaches. In particular, we demonstrate the conceptual equivalence of screened Poisson image reconstruction and our iterative reconstruction scheme. Our composite estimators offer practical and simple solutions that improve upon the current state of the art for the two investigated applications.

Downloads

• Publication

  • Paper, PDF (8MB)
  • Publisher's official version (free via ACM Author-Izer)
  • Publisher's official version (may require a subscription)

• Slides

  • Presentation slides, PDF (9MB)
  • Presentation slides, PowerPoint (22MB)

• Data

  • Test suite (large website, may load slowly) - an interactive comparison of the results

Cite

@article{rousselle16image,
    author   = {Rousselle, Fabrice and Jarosz, Wojciech and Nov\'ak, Jan},
    title    = {Image-space Control Variates for Rendering},
    journal  = {ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia)},
    volume   = {35},
    number   = {6},
    pages    = {169:1--169:12},
    year     = {2016},
    month    = dec,
    doi      = {10/f9cphw},
    keywords = {physically based rendering, gradient-domain, Poisson reconstruction}
}

Copyright Disclaimer