The source code for VKSR is now available on GitHub: github.com/mweiherer/vksr
Give it a try and let me know what you think!
Introducing Vecchia Kernel Surface Reconstruction (VKSR), an implicit surface reconstruction method that efficiently scales kernel surface reconstruction to dense and large-scale point clouds with millions of points.
I'm presenting VKSR at #ECCV2026 in Malmö now at poster #190!
I just retuned from #ECCV2026, and it was a blast! But seriously, @KLM, loosing our luggage twice in one week kinda leaves an aftertaste, covering the taste of your wonderful sandwiches! Make a step and send us free sandwiches!
With @VisionBernie
Introducing Vecchia Kernel Surface Reconstruction (VKSR), an implicit surface reconstruction method that efficiently scales kernel surface reconstruction to dense and large-scale point clouds with millions of points.
I'm presenting VKSR at #ECCV2026 in Malmö now at poster #190!
(4/5) At the same time, VKSR maintains strong performance in a sparse setting, unlike widely used Screened Poisson Surface Reconstruction.
(5/5) This is joint work with Chukwudi Williams Umah and @VisionBernie!
Project page: mweiherer.github.io/vksr/
Maximilian Weiherer retweeted
My team at @Google has just open sourced GNM, our parametric 3D statistical model of the human head, which is learned from a large dataset of 3D scans. It provides fine-grained control over facial identity (170 head and 80 teeth components), expressions (383 components, with a range of -3 to +3) and head pose.
You can get all the data here! github.com/google/GNM/tree/m…
Congratulations to everyone for the release, and stay tuned for the next ones ;-)
Maximilian Weiherer retweeted
Proud to present SchmidhubAI
What do you think @SchmidhuberAI ?
#SIGBOVIK
Excited to announce our latest (submitted to) SIGBOVIK 2026 @sigbovik paper: "SchmidhubAI: Accurate Historical Paper Attribution". We built an AI system that, given any modern AI paper, automatically determines which of its ideas were already published by Jürgen Schmidhuber.
Maximilian Weiherer retweeted
We congratulate Gerd Faltings as the 2026 Abel Prize laureate! 🎉 He recives the Abel Prize "for introducing powerful tools in arithmetic geometry and resolving long-standing diophantine conjectures of Mordell and Lang".
Maximilian Weiherer retweeted
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-kernels…
Was super cool to have you here @UniFAU, @CSProfKGD!
Maximilian Weiherer retweeted
We will have @CSProfKGD presenting on Friday 13th at noon in lecture hall 13 at @UniFAU
"In Search of Universal Concepts"
Very much looking forward to this visit!
(twitter was the platform that initiated our contact, this is why I break my silence on this platform)
Maximilian Weiherer retweeted
Today we’re excited to unveil a new generation of Segment Anything Models:
1️⃣ SAM 3 enables detecting, segmenting and tracking of objects across images and videos, now with short text phrases and exemplar prompts.
🔗 Learn more about SAM 3: go.meta.me/591040
2️⃣ SAM 3D brings the model collection into the 3rd dimension to enable precise reconstruction of 3D objects and people from a single 2D image.
🔗 Learn more about SAM 3D: go.meta.me/305985
These models offer innovative capabilities and unique tools for developers and researchers to create, experiment and uplevel media workflows.
🚀Excited to share our paper "Towards Integrating Multi-Spectral Imaging with Gaussian Splatting" was accepted at VMV 2025! 🎉
We fuse RGB & multi-spectral imagery (red, green, red-edge, near-infrared) into the 3D Gaussian Splatting framework.
🔗Project: meyerls.github.io/towards_mu…
(3/4) To explore this, we systematically compare three strategies for incorporating multi-spectral data into 3DGS and show how spectral cross-talk (the exchange of information between RGB and multi-spectral data) enhances both RGB and spectral reconstructions.
(4/4) Great work led by Josef Grün & Lukas Meyer and with @maxweiherer, @VisionBernie, Marc Stamminger, and @_linus_franke!
📄Paper: arxiv.org/pdf/2509.00989
📷Code: github.com/j-gruen/MS-Splatt…
#VMV25 #GaussianSplatting #NeRF #Multispectral #Rendering
Maximilian Weiherer retweeted
Towards Integrating Multi-Spectral Imaging with Gaussian Splatting
Contributions:
• Formulation and comparison of multi-spectral integration strategies: We introduce and systematically evaluate three optimization paradigms—SEPARATE, SPLIT, and JOINT—for incorporating additional spectral bands into 3DGS.
• We propose a set of optimizations for the JOINT strategy for multi-spectral reconstruction, along with practical insights from this analysis.
• We examine how spectral cross-talk between spectral bands can improve reconstruction quality.
Maximilian Weiherer retweeted
NeurIPS is pleased to officially endorse EurIPS, an independently-organized meeting taking place in Copenhagen this year, which will offer researchers an opportunity to additionally present their accepted NeurIPS work in Europe, concurrently with NeurIPS.
Read more in our blog post and on the EurIPS website:
blog.neurips.cc/2025/07/16/n…
eurips.cc/