Skip to main content
GameDev.net gamedev.net
Research Paper

This is an academic paper or technical research. Key findings may require technical background to fully understand.

Explore Research Radar

PRO Tired of ads? Read GameDev.net ad-free and help keep the community independent with GameDev Pro — $3/month.

arXiv cs.GR
arXiv cs.GR Research
· 8 months, 2 weeks ago • Andrew Tinits, Stephen Mann

Nonlinear Noise2Noise for Efficient Monte Carlo Denoiser Training

Briefing

This article presents a novel theoretical framework for applying nonlinear functions in Noise2Noise training, addressing a critical limitation in denoising HDR images. Developers should take note of how this method can reduce the impact of outliers during training, which has been a significant hurdle in achieving high-quality results with noisy data.

The findings suggest that specific combinations of loss and tone mapping functions can be utilized without introducing substantial bias. This breakthrough enables graphics programmers and artists to refine their workflows, making it easier to produce high-quality visuals in games while relying solely on noisy training data, thus streamlining the denoising process.

“Certain nonlinear functions can be applied without adding significant bias.”

— Research Team · Discussing the findings of their study.
Original source
Read on arXiv cs.GR
At a glance
what
Introduction of nonlinear Noise2Noise for Monte Carlo denoiser training
who
Researchers from arXiv cs.GR
impact
Improves denoising efficiency for HDR images
context
Addresses limitations of traditional Noise2Noise training
Signal Positive

The advancement offers practical solutions for common graphics challenges.

Discuss

Follow noise2noise updates

See relevant stories in your personalized news feed.

Sign in to follow

Continue on GameDev.net

Useful next steps related to this story.

Game development news without the noise

One useful weekly briefing. No daily flood.

Sending your confirmation email…

Discussion

Loading comments...

Recommended resources

Graphics Programming Resources

See full guide
Real-Time Rendering, Fourth Edition cover
Editor pick Community pick

Real-Time Rendering, Fourth Edition

Amazon · Book

Real-Time Rendering combines fundamental principles with guidance on the latest techniques to provide a complete reference on three-dimensional interactive computer graphics. It will help you increase speed and improve image quality and learn the features and limitations of acceleration algorithms and graphics APIs. This latest fourth edition has been updated to include a chapter on virtual reality and augmented reality and covers new topics such as visual appearance, global illumination, and curves and curved surfaces. It is for anyone serious about computer graphics who wants to learn about algorithms that create synthetic images fast enough that the viewer can interact with a virtual environment.

GameDev.net may earn a commission if you purchase through these links. This helps fund the site at no extra cost to you.

Programming with wgpu in Rust cover
Editor pick

Programming with wgpu in Rust

Amazon · Book

Unlock the full power of modern graphics programming with wgpu and Rust. This comprehensive guide takes you from foundational GPU concepts to advanced real-time rendering and compute techniques—equipping you to build fast, safe, and cross-platform graphics applications. Written for intermediate to advanced Rust developers, this book provides clear explanations, hands-on examples, and detailed insights into how GPUs process and render data. You’ll explore everything from the fundamentals of buffers, shaders, and pipelines to advanced topics like deferred rendering, shadow mapping, and GPU compute workloads.

GameDev.net may earn a commission if you purchase through these links. This helps fund the site at no extra cost to you.

GameDev.net may earn a commission if you purchase through these links. This helps fund the site at no extra cost to you.