Working on Graphite/Geogram, improved the GUI, reorganized in a tree (leaves much more space for the 3D view). Will be released soon. For the curious, the repo is up to date here: github.com/BrunoLevy/Gr...
Image: OpenSCAD test suite (displayed in a retro-futurist style for fun...)
#genuary2026 day 1: "One color, one shape."
Save the date : on organise un workshop sur les maths de l'IA à Bordeaux les 4-6 novembre. N'hésitez pas ! plus on est de fous plus on rit ! (dans les limites des capacités d'accueil haha)
Dates & infos :
wmathsia2026.sciencesconf.org
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.
A teaser figure showing the process of metamers rendered differentially (MRD). Target scene parameters are used to render a target scene. A new scene is initialized from some starting point, and renders are created from this scene. The loss between the initial and target scenes is measured. MRD allows the gradients wrt the loss to be propagated to the scene parameters (e.g. lighting, geometry or material) for gradient-based optimization.
Legit super excited about this work coming out. My amazing doctoral student @ben.graphics has been working on an idea to use physically based differentiable rendering (PBDR) to probe visual understanding. Here, we generate physically-grounded metamers for vision models. 1/4
arxiv.org/abs/2512.12307
My general recommendation: don't ever miss a talk by Chris Wyman, regardless of the topic.
I don’t like to publicly criticize science articles, but this work perfectly illustrate what I think is wrong in graphics nowadays : yunfan.zone/gswt_webpage/
This work is clearly doing shell map like texture synthesis, and do not cite any previous work.
The word « texture » is not even mentioned.
I don't really know, as you said, taa is quite popular right now. As for normal mal filtering, i remember that The Order 1886 and Ghost of Tsushima implemented techniques similar to LEAN/LEADR mapping but with other math framework (www.advances.realtimerendering.com)
For specular anti-aliasing there is Tokuyoshi ans Kaplanian's joint work www.jcgt.org/published/00... (and their previous one), they use screen pixel derivatives to modify roughness of the material.
Art of Porco Rosso → buff.ly/pbrxatW
#animation #conceptart #modelsheet #characterdesign #visdev
I usually teach the basics of git to PhD students and one of the tools that works well is learngitbranching.js.org when I presenting ways to organize their repos vi branches
Here's an update on my small-voxel renderer, inspired by the look of games like Doom and Quake. The work I shared last year focused on handling simpler environment geometry. Since then, I've greatly expanded the scope of what I can depict in this art style. More info in thread.
Anisotropic perlin noise on Suzanne
New at #SIGGRAPH2025:
Can we make Perlin Noise stretch along some underlying vector field? Well it turns out it's possible with two simple additions to the original method! No need for advection or convolutions.
Find the paper and implementations here:
github.com/jakericedesi...
Wanna see how to speedup x12 flyover procedural FBM worlds (possibly animated) ?
→ Check Mathéo Moinet & I new Eurographics paper "Fast sphere tracing of procedural volumetric noise for very large and detailed scenes" (tuesday) inria.hal.science/hal-05046040 (paper, video,+slides soon)
Heroes sketchbook commission of Porco Rosso’s plane. Incidentally my favorite film.
Nice writeup ! I really love how it's presented with the step-by-step highlights. It makes me want to do something similar to create tutorials/exercises for students
easy isometric pixel art is still coming... with a ton of new features!
create anything with easymetric soon on @aseprite.org
Do you like voxels and implicits' representations and wish you could animate them in real-time?
Now you can!
Check out our article: "Real-time rendering of animated meshless representations" to know how: hal.science/hal-05095359
Shout out to @pacomeluton.bsky.social for this fantastic work!
The infinite poetry into the waves of Frederick Judd Waugh (1861-1940)
"Neon"
for(float i,d,z;i++<1e2;){vec3 p=z*normalize(r.xyy-FC.rgb*2.)-1.;d=max(p.y,0.);p.y-=d+d;p.z+=t;z+=d=.3*(.1*d+.01+length(vec2(cos(p.z*6.)/6.,1.-dot(cos(p),sin(p).yzx)))/++d/d);o+=(cos(p.z+vec4(0,1,3,0))+1.5)/d/z;}o=tanh(o/8e2);
Left: an input CSG tree and a much smaller pruned tree computed using our method. Right: a rendered scene showing the number of active nodes per cell. Our method reduces the active nodes to less than 20 from the initial 6023 nodes of the input tree.
I am proud to announce our Eurographics 2025 paper "Lipschitz Pruning: Hierarchical Simplification of Primitive-Based SDFs"! With Mathieu Sanchez (joint first author), @axelparis.bluesky.social, @elie-michel.bsky.social, Thibaud Lambert, @tamyboubekeur.bsky.social, Mathias Paulin and Théo Thonat.
3D Gaussian splatting relies on depth-sorting of splats, which is costly and prone to artifacts (e.g., "popping"). In our latest work, "StochasticSplats", we replace sorted alpha blending by stochastic transparency, an unbiased Monte Carlo estimator from the real-time rendering literature.
Dear every fucking piece of technology,
I don't want to download an app.
I don't want to make an account.
I'm not turning on notifications unless I want to.
You dont need my location or permissions for everything.
Fuck off. Leave me alone. I will use you how I want.
I want dumb tech back!
I fell in löve (pun intended) with Lua for parameterizing my C/C++ programs and for prototyping.
#screenshotsaturday
I remember testing Nuklear and liking it. It's ANSI C though, not C++
"Qui peut l'ångström peut le méga-parsec !"
Hi Everyone, I'm a French researcher working on computer graphics.
My research focuses on Procedural generation and Geometry.
@pltc.bsky.social Congrats "Docteur" !
SamplingLab : a small tool to live code and visualize samples
aduprat.com/tools/sampli...
Since my last post with a protein render seems to have found an audience, here is a video displaying another large system
