Sign in

Ben Beilharz

@engramm.ing
295 followers 528 following 252 posts

Cognitive Science PhD student working on (inverse) light transport for sensing, planning & acting @tsawallis.bsky.social's Perception Lab. AI for science. I like my models uncertain. he/him

PostsRepliesMedia
Ben Beilharz @engramm.ing · 17/09/2026
Two new interesting graphics papers. Gaussian Light Transport: patrick-attimont.com/projects/gau... Diffusion Restore: arxiv.org/abs/2605.08916
arxiv.org
Diffusion Restore: Real-Time Markov Chain Monte Carlo Light Transport
We present Diffusion Restore, a real-time framework for diffusion-based MCMC light transport. MCMC methods are highly suitable for sampling from complex high-dimensional distributions and for approxim...
0112
Reposted by Ben Beilharz
Valve @valvesoftware.com · 14/09/2026
Steam Frame is here! Our wireless VR headset + controllers, available now in the following models: Steam Frame 256GB Steam Frame 1TB Sign up now and learn more on @steampowered.com : store.steampowered.com/hardware/ste...
892490611
Ben Beilharz @engramm.ing · 07/09/2026
I paddle back from my statement, if this is actually what's happening:
000
Ben Beilharz @engramm.ing · 06/09/2026
DLSS5 technical report is available: research.nvidia.com/labs/adlr/fi...
research.nvidia.com
021
Reposted by Ben Beilharz
𝒩𝒾𝒸𝑜𝓁𝒶𝓈 𝒸 𝒢𝓇𝑒𝓎 @nicolascgrey.bsky.social · 24/08/2026
1235391308
Ben Beilharz @engramm.ing · 06/09/2026
This is now without context, because the user blocked you. Who was it?
100
Ben Beilharz @engramm.ing · 03/09/2026
Exactly what we NOT needed. Another open platform acquired by big tech. blogs.nvidia.com/blog/nvidia-...
blogs.nvidia.com
NVIDIA to Acquire Hugging Face
NVIDIA has agreed to acquire Hugging Face. Together, we will scale Hugging Face’s platform, strengthen its infrastructure and expand access to AI for developers and institutions worldwide.
000
Reposted by Ben Beilharz
Andrew Willmott @andrewwillmott.bsky.social · 31/08/2026
It's really hard to beat this for a practical, worked example of how modern CPU/memory architectures affect performance. It describes the issues with (and corresponding solutions for) a simple matrix transpose: gudok.xyz/transpose/.
gudok.xyz
What it takes to transpose a matrix
03412
Reposted by Ben Beilharz
Aaron Rupar @atrupar.com · 30/08/2026
Pathetic -- Google is now referring to Lake Ontario as "Lake America"
screencap of Google Maps showing "Lake America"
3438143342989
Ben Beilharz @engramm.ing · 31/08/2026
Proton Mail is waiting for you.
010
Ben Beilharz @engramm.ing · 30/08/2026
Post #ecvp 😬
080
Ben Beilharz @engramm.ing · 28/08/2026
I finally get why people call this conference a family. @ecvp.bsky.social #ecvp
050
Ben Beilharz @engramm.ing · 27/08/2026
If you are interested in MRD, I am editing a blog post (will probably be updated a couple of times): engramm.ing/journal/rese... This will more casually outline the challenges and ideas of the metamer work. I am happy for people to chime in and discuss these things. 2/2
engramm.ing
MRD: Using Physically Based Differentiable Rendering to Probe Vision Models for 3D Scene Understanding・エングラム
While deep learning methods have achieved impressive success in many vision benchmarks, it remains difficult to understand and explain the representations and decisions of these models. Though vision ...
010
Ben Beilharz @engramm.ing · 27/08/2026
Hey @ecvp.bsky.social thanks for joining the material perception session. I have uploaded my slides to my website: engramm.ing/journal/rese... Also, if you want to reuse any assets from the talk, please write me a DM so I can give you the edit password for the files. #ecvp2026 1/2
engramm.ing
[ECVP26] Talk: Applications of Physically Based Differentiable Rendering in Vision Science・エングラム
Physically Based Rendering (PBR) is the physical simulation of light transport within a scene, i.e., how light participates in events such as reflections, refractions and scattering when interacting w...
150
Ben Beilharz @engramm.ing · 27/08/2026
Good morning @ecvp.bsky.social. Sadly, the conference is already nearing its end, but @lschmittwilken.bsky.social still has something nice for you! So now you have a date: 10:30 in the Active Vision and Visual Salience session (Bayview Suite). Have a great day everyone! :) #ecvp2026
030
Ben Beilharz @engramm.ing · 26/08/2026
G'day @ecvp.bsky.social, it's time for another treat: Lina will present her poster on "Causality perception under visual uncertainty" at W1AM2. In the afternoon, I will be giving my first in-person talk in the Material Perception session about differentiable rendering in vision science. #ECVP2026
040
Ben Beilharz @engramm.ing · 25/08/2026
Hey @ecvp.bsky.social we got twice the fun for you today (order of visit recommended): T5AM6 Swantje will talk about Dead Leaves stimuli with which you can model natural image statistics. T5AM1 @lreining.bsky.social will talk about how to segment these dead leaves, testing NNs for optimality.
072
Ben Beilharz @engramm.ing · 24/08/2026
goooooood mornin‘ @ecvp.bsky.social! If you are an early bird and you like to see something and talk about multi focal display, or just want to know how to easily implement your experiments on one, go ahead and pay Tarek a visit at M3AM1 :) #ecvp26
041
Reposted by Ben Beilharz
Tom Wallis @tsawallis.bsky.social · 23/08/2026
TU Darmstadt's Perception Lab is at #ECVP2026 in force. We hope you can make it to see our work! @engramm.ing @lreining.bsky.social @lschmittwilken.bsky.social
A graphic depicting the times that TU Darmstadt's Perception Lab members have posters at the European Conference on Visual Perception, 2026.
0184
Reposted by Ben Beilharz
ECVP 2026 @ecvp.bsky.social · 23/08/2026
🎉 TODAY’S THE DAY! 🎉 #ECVP2026 begins today in Bournemouth! 👁️🧠 We’re ready for a fantastic week of vision science, workshops, talks, symposia, posters and plenty of opportunities to connect with the ECVP community. Welcome to ECVP 2026 — let’s get started! 🚀 #VisionScience #Perception
0127
Ben Beilharz @engramm.ing · 09/08/2026
@lreining.bsky.social link's dead. :/
000
Reposted by Ben Beilharz
Lars C. Reining @lreining.bsky.social · 09/08/2026
Excited this is finally out! I am happy to have built this together with Rabea Turon, @tsawallis.bsky.social and all the others!🎉
163
Ben Beilharz @engramm.ing · 06/08/2026
Freu mich drauf. :]
000
Ben Beilharz @engramm.ing · 05/08/2026
I am moving over to codeberg.org in the coming days. Maybe you should do the same. I'll mirror to GitHub for the time being.
codeberg.org
Codeberg.org
Codeberg is a non-profit community-led organization that aims to help free and open source projects prosper by giving them a safe and friendly home.
000
Ben Beilharz @engramm.ing · 03/08/2026
Happy to announce that MRD got accepted by the Journal of Vision (marks the first submission of my phd). Thanks to all editors and reviewers making the paper better over time. If you don't want to wait for the official JoV version, I updated the preprint & repo with the latest code and results.
284
Ben Beilharz @engramm.ing · 28/07/2026
We got rogue AI before GTA6. Three 0-days in a weekend. Where's my blackwall? www.youtube.com/watch?v=q2KC...
youtube.com
Did an AI Really Hack Hugging Face?
YouTube video by LiveOverflow
000
Reposted by Ben Beilharz
Stephen Hill @selfshadow.com · 27/07/2026
SIGGRAPH 2026 links: blog.selfshadow.com/2026/07/26/s.... I'll continue to add to the list as more things show up online.
blog.selfshadow.com
SIGGRAPH 2026 Links
05225
Ben Beilharz @engramm.ing · 25/07/2026
It is fucking horrendous what happened today at the Berlin Pride Parade. Kinda sucks that hate progresses quicker than tolerance and acceptance. Currently only German news: www.tagesschau.de/inland/gesel... TL;DR some sicko drove into a group of people after the demonstration. 1 dead person.
tagesschau.de
Berliner CSD abgebrochen - mutmaßlich Fahrzeug in Menschenmenge gefahren
Der Berliner Christopher Street Day ist abgebrochen worden. Laut Polizei ist mutmaßlich ein Fahrzeug in eine Menschenmenge gefahren, mehrere Menschen seien verletzt.
010
Ben Beilharz @engramm.ing · 20/07/2026
We actually get to witness Nvidia open sourcing their Neural Appearance Models in this lifetime. Thanks for finally dropping the guesswork reimplementing all of that. github.com/NVlabs/neura... Their new SIGGRAPH is also out: research.nvidia.com/labs/rtr/pub...
github.com
GitHub - NVlabs/neuralappearance
Contribute to NVlabs/neuralappearance development by creating an account on GitHub.
0114
Ben Beilharz @engramm.ing · 14/07/2026
Disclaimer: I wrote the code, Claude the documentation and tests. I want to do another pass on it, which is why I want to postpone the release a bit. 5/5
000
Ben Beilharz @engramm.ing · 14/07/2026
I've implemented a couple of models I found interesting. The idea would be to have people filing PRs with their experiments to gather a library of neural material approaches in which we can easily benchmark and evaluate different approaches. 4/5
100
Ben Beilharz @engramm.ing · 14/07/2026
You can author the materials in a couple of different ways: Either you take preexisting measured BSDFs from Merl and RGL, analytical models from Mitsuba, or BYO @shader-slang.bsky.social shaders being loaded via SlangPy. There is very rudimentary support for @materialx.bsky.social at the moment. 3/5
100
Ben Beilharz @engramm.ing · 14/07/2026
It is not ready yet, but what I have so far: - Common helpers to tabulate and reparameterise BSDF data - Loaders for MERL and RGL's data - Pipelines, losses, and evaluations - A few models - Adapters to use the fitted BSDFs (Mitsuba; coopvec). - RTC-like neural texture compression. 2/5
100
Ben Beilharz @engramm.ing · 14/07/2026
After the neural material/shading talks at EGSR, I was wondering if there's a need for a package to facilitate this type of research. It would be nice to have a package with reusable building blocks and a set of reimplementations to benchmark against. This is currently what I am working on. 1/5
130
Ben Beilharz @engramm.ing · 08/07/2026
That sounds right up your alley. Happy for you, enjoy the break! :)
020
Ben Beilharz @engramm.ing · 07/07/2026
Just wrapped up implementing ボンド's path space API. Previously, I had a loop that constructed the path incrementally without storing the vertices. Now, the whole path is created at once and then evaluated. While this is still a unidirectional case, it would enable me to implement BDPT next.
010
Ben Beilharz @engramm.ing · 07/07/2026
Just seeing that @guerrilla.bsky.social has a new tab in their nav bar pointing to the Decima engine: www.guerrilla-games.com/decima I don't want to hypothesise, but if that means that the engine is being released at one point, we could see serious competition on the market for UE, Unity and Godot.
guerrilla-games.com
We create awe-inspiring worlds - Guerrilla
Guerrilla is one of Europe's leading game companies and a wholly-owned subsidiary of Sony Interactive Entertainment Europe. We started in 2003, and have pushed the boundaries of technical and artistic...
010
Ben Beilharz @engramm.ing · 06/07/2026
Alan's (@demofox.bsky.social) blog post is what could solve a lot of anxiety for new graphics people: blog.demofox.org/2026/07/01/w...
blog.demofox.org
What To Learn To Be A Real Time Graphics Programmer
I get asked fairly often what people need to know to be hireable as a graphics programmer. I figured it was time to make a page to link instead of re-typing it each time. We are in a strange time w…
0294
Ben Beilharz @engramm.ing · 06/07/2026
Pretty cool new sim work by TUM: www.youtube.com/watch?v=-1Tx... Paper: ge.in.tum.de/download/ST-...
youtube.com
SIGGRAPH'26: Spatiotemporal FLIP for Fast Free-Surface and Two-Phase Sims With Very Large Time Steps
YouTube video by Nils Thuerey
084
Ben Beilharz @engramm.ing · 06/07/2026
All of them were working when they first added the links. I think something went wrong afterwards.
000
Ben Beilharz @engramm.ing · 04/07/2026
* Mirco's Optimized and Aligned Anisotropic Monte Carlo Sampling Patterns: diglib.eg.org/items/3ba69a... * My personal favourite (and best paper awarded): Elie's Volumetric Inverse Rendering via Neural Radiative Transfer diglib.eg.org/items/dcda7e... 5/5
diglib.eg.org
Volumetric Inverse Rendering via Neural Radiative Transfer
Volumetric inverse rendering seeks to recover the optical properties of participating media from images. Existing approaches either rely on differentiable stochastic light transport simulation, which ...
010
Ben Beilharz @engramm.ing · 04/07/2026
* Laurent's Neural Texture Compression using Hypernetworks: diglib.eg.org/items/0229b2... * Sergio's Neural Render Proxies for Interactive and Differentiable Lighting: studios.disneyresearch.com/2026/07/01/n... 4/5
diglib.eg.org
Neural Texture Compression using Hypernetworks
Recent work on neural texture compression has demonstrated that it is possible to learn small, per-material texture representations (composed of latent textures and a small Multi-Layer Perceptron deco...
100
Ben Beilharz @engramm.ing · 04/07/2026
* Song's Conditional Product NEE for Gaussian Process Implicit Surfaces: diglib.eg.org/items/732105... * Liliana's Keynote about Avian colouration: egsr2026.inria.fr/keynotes/#alba * Louis' A Hybrid Neural-Microfacet BRDF Model for Real-Time Rendering: diglib.eg.org/items/add38f... 3/5
diglib.eg.org
Conditional Product Next Event Estimation for Gaussian Process Implicit Surfaces
We propose a new next-event estimation (NEE) technique for Gaussian Process Implicit Surfaces (GPISes). We show that the distribution of surface normals available for sampling at a GPIS hit point coll...
100
Ben Beilharz @engramm.ing · 04/07/2026
* Fengshi's Illumination-Aware Spatial Subdivision for Path Guiding: diglib.eg.org/items/44e861... * Jorge's One-more-vertex Next-Event Estimation with Hierarchical Geometry Sampling: diglib.eg.org/items/941558... 2/5
diglib.eg.org
Illumination-Aware Spatial Subdivision for Path Guiding
Path tracing is ubiquitous in production rendering and path guiding has established itself as a powerful approach to mitigate its failure cases. Several widely used methods partition the scene using a...
120
Ben Beilharz @engramm.ing · 04/07/2026
#EGSR26 has come to an end, and I am sad it is already over. Tons of interesting chats, nice, approachable people, and overall just a great experience. Thanks to the organisers for making this possible; looking forward to next year in Lugano! My highlights: 1/5
120
Reposted by Ben Beilharz
Godot Engine @godotengine.org · 30/06/2026
For years we have been overwhelmed by the volume of code contributions we have to review, especially from new contributors. AI has made the problem much worse. That's why we are taking steps to reduce the burden on maintainers while still welcoming new contributors: godotengine.org/article/cont...
godotengine.org
Changes to our Contribution Policies – Godot Engine
Godot is growing fast, so here is how we are dealing with the huge increase in contributions.
472494585
Ben Beilharz @engramm.ing · 29/06/2026
See you at MANER/EGSR tomorrow! :)
100
Reposted by Ben Beilharz
Delio Vicini @deliovicini.bsky.social · 26/06/2026
I'm excited to share the open-source release of hdMitsuba, a USD Hydra delegate for the Mitsuba 3 renderer. By leveraging the Hydra API, this project brings Mitsuba directly into OpenUSD workflows. The GIF below shows interactive rendering within usdview. 🧵
1143
Ben Beilharz @engramm.ing · 26/06/2026
If Mr. BVH says that, I am relieved. The BVH itself was one of the more difficult implementations, just because I had never implemented one before that was trying to be more efficient, both from a memory and an intersection point of view.
010
Reposted by Ben Beilharz
Wenzel Jakob @wjakob.bsky.social · 26/06/2026
I am excited to announce Dr.Jit 1.4 and Mitsuba 3.9 which now feature GPU-accelerated compute/ray tracing on Apple Silicon. Significant optimizations in both projects accelerate tracing/code generation ~2x. They generate better code and have faster Python bindings.🧵
1318