Aaron Lefohn

Vice President of Graphics Research at NVIDIA. Opinions are my own.

[Wenzel Jakob {deprecation notice} @wenzeljakob]

Methods like NeRF and Gaussian Splats model the world as radioactive fog, rendered using alpha blending. This produces great results.. but are volumes the only way to get there?🤔 Our new SIGGRAPH'25 paper directly reconstructs surfaces without heuristics or regularizers.

[Wenzel Jakob {deprecation notice} @wenzeljakob]

If you try to optimize geometry using a differentiable renderer, there is an elephant in the room: geometry causes discontinuous visibility changes, which mess up the derivatives. To use indirect cues like shadows in geometric reconstructions, this issue must be fixed. (1/7)

[Aaron Lefohn @aaronlefohn]

@wenzeljakob Congratulations!

[Wenzel Jakob {deprecation notice} @wenzeljakob]

The free web version of "Physically Based Rendering: From Theory To Practice" is now based on the 4th edition of the book. Enjoy! (Link: https://t.co/N7CroKjKLC)

[𝖒𝖒𝖆𝖑𝖊𝖝 @mmalex]

so excited by the explosion of gaussian splatting; cant wait for the community to start playing with painterly styles; on dreams, we used splats with tiny alpha-textured meshes called 'flecks' (not flat!) that selectively can replace the little 'blob' with painterly texture 1/n

[Daqi Lin @DaqiLin]

Our new conditional resampled importance sampling (CRIS) opens up possibilities for designing resampling algorithms where marginal probabilities are unknown. As an example, we show that ReSTIR PT can be de-correlated like applying final gather to photon mapping.

[Aaron Lefohn @aaronlefohn]

This great foundational advancement in ReSTIR now enables reuse of sub-paths.

Our theory research directly targets the needs of real-time rendering.

[Markus Kettunen @nmkettunen]

Thrilled to announce our work "Conditional Resampled Importance Sampling and ReSTIR" (SIGGRAPH Asia 2023). We extend RIS and ReSTIR into conditional probability spaces, enabling novel forms of sample reuse. 1/7

[NVIDIA AI Developer @NVIDIAAIDev]

Differentiable Slang easily integrates with existing codebases—from #Python, PyTorch, and #CUDA to HLSL.

Here we introduce code examples using differentiable Slang to demonstrate the use across different rendering apps and ease of integration.

🧵 2/2

https://t.co/e7RMjdXWbX

[NVIDIA AI Developer @NVIDIAAIDev]

New #NVIDIAResearch paper: SLANG.D: Fast, Modular and Differentiable Shader Programming: shows how a single language serves as a unified platform for real-time, inverse, and differentiable rendering.

Collaboration with @MIT, @UCSanDiego, & @UW.
🧵 1/2

https://t.co/5DHzQldMMV

[Theresa Foley @TangentVector]

The Slang project is seeking experienced GPU/graphics/AI compiler programmers who want to be part of the development of an ecosystem for AI-powered real-time graphics. My DMs are open.

[Marco Salvi @marcosalvi]

I have been working with machine learning in graphics for a few years now and SLANG.D is the tool I wished I had from the start. Being able to easily sprinkle some gradient descent on your HLSL code & learn from data is invaluable. This is a major milestone for rendering.

🧵1/2

[Bart Wronski 🇺🇦🇵🇸 @BartWronsk]

Some awesome work of my teammates on differentiable shading language, compatible with HLSL, and interoperable with CUDA/Python/PyTorch/C++.
https://t.co/CSVTnoSmPY

This makes ML + graphics *significantly* easier. I'm excited to see what researchers and engineers do with it.:)

[Yong He @csyonghe]

Bringing autodiff to shaders is a challenging task. It takes years of effort to design the language that integrates differentiation as a first-class citizen, allowing autodiff to work seamlessly with custom types, arbitrary control flow, generics and dynamic dispatch.

[Yong He @csyonghe]

Slang's Visual Studio Code extension fully supports the autodiff features.

[Aaron Lefohn @aaronlefohn]

Slang is now fully differentiable.

Generate PyTorch plugins from shader code.

Create rendering algorithms using appearance-based optimization.

Build differentiable renderers using your current shader codebase.

DiffSlang connects real-time rendering and learning.

[Yong He @csyonghe]

Slang is an open-source, cross-platform shading language that targets D3D, Vulkan, GLSL, CUDA and C++. Today, it is fully differentiable, which means you can autodiff your existing shader code!

[NVIDIA AI Developer @NVIDIAAIDev]

🏃 Rapid recap of #NVIDIAResearch released recently at #SIGGRAPH2023 👇

https://t.co/HAEelHxXSI

[Aaron Lefohn @aaronlefohn]

Fantastic collaboration with @FidlerSanja's team. Working together to improve the quality of geometry extracted with inverse rendering.

[Tianchang Shen @TianchangS]

Excited to share our #SIGGRAPH2023 project for better mesh extraction in learning & reconstruction pipelines.
Project page: https://t.co/lADmnKmETj
Joint work w/ @yoslber Jon Hasselgren @kangxue_yin @zianwang97 @ChenWenzheng @ZGojcic @FidlerSanja @nmwsharp @JunGao33210520
(1/n)

[Kayvon Fatahalian @kayvonf]

🎾 Haotian's AI tennis players are back... in 3D. 🎾

AI characters learn to play full 3D simulated tennis points by watching professional play depicted in large numbers of broadcast tennis videos.

https://t.co/rtrkfDp5Zh

1/

[Jan Kautz @jankautz]

Honored to see that our work on Precomputed Radiance Transfer (2002) made it onto the SIGGRAPH Seminal Graphics Papers list: https://t.co/GVDbeQav57. Joint work with Peter-Pike Sloan and John Snyder.

[Sebastian Aaltonen @SebAaltonen]

The FlexiCubes technique by Nvidia (Flexible Isosurface Extraction for Gradient-based Mesh Optimization) seems very promising.

https://t.co/cpySU2bD9W