Smoothing スムージング [Math short write-up] www.kynd.info/writing/smoo...

CCCL Runtime: A Modern C++ Runtime for CUDA developer.nvidia.com/blog/cccl-ru... #cpp #cplusplus

lens design is matching up very well with the diagram from the original design in Modern Optical Engineering - you can measure the focal length of a lens by passing collimated light through the front surface and seeing where it focuses

The program for DigiPro 2026 has been posted! There's a lot of great talks this year; I'm really looking forward to Alexey Stomakhin's talk on WetaFX's new fire solver/system. If you're familiar with Alexey's work, you already know it's going to be mind-blowing. digiproconf.org/program/

The alpha version of my new book "Optimal Transport for Machine Learners" is out, with in particular an online version with interactive figures www.gpeyre.com/ot4ml/

😎

When dealing with noisy and/or partial inputs to describe a shape, the standard approach in geometry processing is to reconstruct a clean implicit surface first, and to do geometry on it. But why this one in particular? What about all the other surfaces that would have been coherent with the input?

Here's a fun blog post about bytecode VMs in surprising places. Mildly hot take of mine: a bytecode VM in a GPU kernel is not as bad of an idea as one might think, and in some cases it can actually be the best solution. Fun examples in thread: (1/5) dubroy.com/blog/bytecod...

Working with Cyprien Plateau-Holleville, Charles Dapogny, Edouard Oudet and Roya Mohayaee on shape optimization / partial optimal transport. A lot of work remains to do (in particular, Virtual Element method in 3D) Preprint (2D version) here: arxiv.org/abs/2409.07873

Thrilled that our paper "AtomSlicer" was accepted to SIGGRAPH 2026! A 5-axis FFF slicer for constant-thickness field-aligned non-planar layers + continuous, collision-free toolpaths. Code is out: xavierchermain.github.io/publications... @univlorraine.bsky.social @cnrs.fr @labo-loria.bsky.social

iveevi.github.io/rcgp/ Check out Venkat's work on a new graphics programming language on statically enforcing correctness of heterogeneous graphics programs (e.g., resources declared on host must match shaders at compile time). The compiler is open sourced and let us know if you have any comment!

A set of very disciplined Voronoi seeds.

The Tufte LaTeX class is pretty, but it hasn't been updated for years and doesn't work with well with Quarto/lualatex. But with #QuartoPub you can replicate it on your own in HTML/PDF with just a little bit of customization with CSS and some preamble-y things! github.com/andrewheiss/...

Plouf ! Our @eurographicsc.bsky.social course notes on Optimal Transport for Fluids, New and Old, with @pltc.bsky.social are out, here: diglib.eg.org/items/68a801...

Inspired by @sebaaltonen.bsky.social's "No Graphics API", I've written a GPU pointer-centric bindless wrapper on Vulkan+Metal. Still in rough shape, but I've written about it here: www.kevin-gibson.com/blog/writing...

Fun with semi-discrete Optimal Transport (getting ready for our course at Eurographics Conference next Monday and ACM SIGGRAPH in July with Cyprien Plateau-Holleville)

New blog post! In "Quantizing tangent frames", we look at various established methods to represent tangent frames in the vertex data, squeeze a few variants into 32 bits per vertex and look at the resulting precision. zeux.io/2026/04/30/q... Share, like and subscribe!

3DGS is fast to rasterize but bad for ray tracing, Radiant Foam is great for ray tracing but not good to rasterize; this paper promises a differentiable representation good for both. I haven't read the paper yet, but it's going straight to the top of the to-read pile! powerfoam.github.io

Preparing for our course "Optimal Transport Based fluid simulation: New and Old" with Cyprien Plateau-Holleville (EUROGRAPHICS and SIGGRAPH) 2D simulation of "droplet hurdles race" (De Maleprade et.al). Features high surface tension, viscosity and fluid boundary - domain interactions.

This is a great talk: www.youtube.com/watch?v=etGM... Two fun facts that I learned: - The Cycles render kernel can compile as any of C++17, CUDA, or HIP (from the same code, with lots of preprocessor macro magic) - Shaders are compiled to bytecode which is executed by a VM running on the GPU

New Paper🙂 Nanite has shown that small triangles can be rendered fast in compute, we're exploring how fast for large meshes with up to 18.9 billion triangles, without the need to precompute LOD structures. Paper: github.com/m-schuetz/Cu... Source: github.com/m-schuetz/Cu...

Looking forward to see you in Aachen and/or Los Angeles!

I finally wrapped up the second post on the series about micro-shadowing. On this one I go over a basic approach based on a microsurface, and I show results for different materials and lighting conditions. irradiance.ca/posts/micros...

Ultra-fast Screen-Space Refractions and Caustics via Newton’s Method jcgt.org/published/00...

I spent the last few days implementing the fantastic curvenet pixar article : dl.acm.org/doi/epdf/10.... it's very cool to be able to animate meshs without having to paint animation weight A pain to implement though ^^

At long last, the Teardown talk from the 2025 Graphics Programming Conference has been posted 🎉 youtube.com/watch?v=IM1D... (as well as another drop of some other great-looking stuff, check the channel)

If, like me, you wake up every morning wondering what the Voronoï diagram of arbitrary shapes looks like, here is the answer:

🔔 We’re #Hiring! The Forge Interactive is looking for Graphics Programmers and Interns to join our team! 🌐 Apply on LinkedIn: The Forge Interactive, Inc. #GraphicsProgramming #Intern #GameDevCareers #OpenRole

Preparing with @pltc.bsky.social our course on Optimal Transport for @eurographicsc.bsky.social He created a hands-on experience in Python, from Laguerre diagrams to semi-discrete OT, from simple bouncing balls to incompressible Euler fluid simulator. Looking forward to see you in Aachen !

Four weeks to the EGSR papers deadline! (I'm so excited to be on the Program Committee this year - cannot wait to see all the amazing papers submitted!) So - submit away! Important dates: Abstracts deadline: 4/8 Papers deadline: 4/15 Submit your work, and join in Bordeaux! egsr2026.inria.fr

- You are curious about Optimal Transport ? - You'd like to know what is "Semi-Discrete" ? - You'd like to know how it helps computational cosmology ? Short version of the story: github.com/BrunoLevy/Gr... P.S. post-doc position available very soon, stay tuned (or DM me if you can't wait).

Happy to share a major milestone: after years of development, we are officially launching Version 1.0 of the GeometricKernels library! To top it off, our accompanying paper has just been published in JMLR (MLOSS)! 🎉 github.com/geometric-ke...

My team just open-sourced our internal implementation of the OpenPBR BSDF from our path tracer: github.com/adobe/openpb...

Wrote up an article about how I ended up tackling this problem from January (spoiler: I gave up on barriers), might be useful to somebody out there learning Vulkan for compute amini-allight.org/post/dispatc...

Le Prix d'Honneur revient à Omar Cornut, créateur du logiciel open source ImGui qui est utilisé par toute l'industrie du jeu vidéo

Ray Tracing using HIP arxiv.org/abs/2603.00292

Continuing the adventures in neural networks, investigating how they can be hardware accelerated using Cooperative Vectors in HLSL, achieving impressive speedups. interplayoflight.wordpress.com/2026/02/21/a...

"Forget Superresolution, Sample Adaptively (when Path Tracing)" arxiv.org/abs/2602.08642

Cool. I posted a while ago about fast CDF reconstruction. I wonder if you are doing something like this or have a different technique. maxliani.wordpress.com/2024/03/09/a...

Ever seen one of these, and wondered *why* they work? 🧵 This is a so-called Bitonic sorting network, and the illustrations show the two common ways to implement them. From an implementation point of view the two methods are effectively the same, so which one you pick just comes down to preference.

I'm finally writing up how Nanite Tessellation works. The first few blogs posts are up. More will be coming. graphicrants.blogspot.com/2026/02/nani...

New post on how to do real-time diffuse global illumination using surfels and #WebGPU: juretriglav.si/surfel-based... #threejs

This was a fun quick experiment: Using an old photo I took in the forest, I used ml-sharp to generate gaussian splats from it, voxelized it in my toy renderer, relit the scene dramatically. #rendeting #metalrt

Geogram flies ! The other day, I was reading this article from NASA Ames Research Center: ntrs.nasa.gov/api/citation... What do they use for the geometric computations ? Geogram of course ! (thrilled !!) Geogram is a library with geometry processing algorithm: github.com/BrunoLevy/ge...

Gaboronoi (top) vs Anisotropic Voronoi (bottom), with 5000 Sites. I feel like the anisotropic voronoi's artifacts are more aesthetically pleasing. Gaboronoi is faster to train by a lot! I've compiled all of my recent voronoi experiments into a collab: colab.research.google.com/github/jaker...

My "How to Vulkan in 2026" @vulkan.org #Vulkan guide is now publicly available at www.howtovulkan.com I still consider it a preview, though I'm mostly happy with it and only plan on changing minor things and incorporating some feedback.

My "No Graphics API" blog post is live! Please repost :) www.sebastianaaltonen.com/blog/no-grap... I spend 1.5 years doing this. Full rewrite last summer and another partial rewrite last month. As Hemingway said: "First draft of everything is always shit".

Posting an older article on "Probability Theory for Physically Based Rendering", using tomorrow's #thisisigad Graphics Guild lecture as an excuse: jacco.ompf2.com/2019/12/11/p... There is also a part 2.

A modernized Utah Teapot model with interior! You can see/customize/download it, along with historical versions on the official teapot page: graphics.cs.utah.edu/teapot/