Heading to Hawaii for #ICCV25 — with a quick Bay Area stop this week! Excited to catch up with folks — ping me if you’d like to meet up or chat :)

📢📢📢We've released the ScanNet++ Novel View Synthesis Benchmark for iPhone data! 🥳 Test your models on RGBD video featuring real-world challenges like exposure changes & motion blur! Download the newest iPhone NVS test split and submit your results! ⬇️ https://t.co/hLnFwifvTL

Congratulations to Noah Goodwin @NoahGoodwin08 week 9 Offensive Player of the Week. Great Job Noah! 5 TDs @BerserkerScout @CoachLakey53 @Alabama_FN @CoachBlanksSLFB @KevinMoses38 @N_W_Sports

In the 'early days' of modern deep learning (2012-2015) when ConvNets such as AlexNet or VGG came out, it was considered almost impractical to train an ImageNet classifier from scratch. The required compute was typically a couple of GPUs on a single desktop machine, trained over

🚀We have PhD openings in my lab at TU Munich! Explore 3D/4D reconstruction & generation, semantic & functional understanding, and more - at the intersection of graphics, vision, and machine learning. 💼PhDs are 100% E13 positions 👉Apply: https://t.co/A0KhPKmSbD or via ELLIS!

Fantastic retreat this weekend by our research groups! Internal reviews, ideas brainstorming, paper reading, and much more! Of course also many social activities -- the highlight being our kayaking trip - lots of fun :)

2/ I find traders with insane stats, subscribe to their accounts, and let Alertsify auto execute their moves. Simple, effective, and profitable. 💰 Want in on the action?

All six of our submissions were accepted to #NeurIPS2025 🎉🥳 Awesome works about Gaussian Splatting Primitives, Lighting Estimation, Texturing, and much more GenAI :) Great work by @Peter4AI, @YujinChen_cv, @ZheningHuang, @jiapeng_tang, @nicolasvluetzow, @jnthnschmdt 🔥🔥🔥

We generate multiple videos along short, pre-defined trajectories that explore the scene in depth. Our scene memory conditions each video on the most relevant prior views while avoiding collisions. Great work by Manuel Schneider & @LukasHollein

We generate fully-navigable 3D scenes from text input in three stages. 1) A panoramic image scaffold defines the scene layout. 2) We expand it with video diffusion in an iterative scene generation pipeline. 3) Finally, we optimize a 3DGS scene from all generated frames.

Can we use video diffusion to generate 3D scenes? 𝐖𝐨𝐫𝐥𝐝𝐄𝐱𝐩𝐥𝐨𝐫𝐞𝐫 (#SIGGRAPHAsia25) creates fully-navigable scenes via autoregressive video generation. Text input -> 3DGS scene output & interactive rendering! 🌍 https://t.co/HBdrmU4Oqq 📽️ https://t.co/AQr0p4uWBZ

We just published v2 of the Avat3r paper with more analyses of the trained model: + More phone capture results + Comparisons with single-view methods + What happens if you: * vary number of input images? * add more train subjects? Check it out:

HairGS: Hair Strand Reconstruction based on 3D Gaussian Splatting (#BMVC2025) 📢 We reconstruct realistic 3D hair strands from multi-view images in ~1 hour. Results show robustness to a wide range of hair styles such as challenging curly hair. Project: https://t.co/qcOgGjSq7W

📢📢Want to build 𝟑𝐃 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬?📢📢 ➡️We're looking for Diffusion/3D/ML/Infra engineers and scientists in Munich & London. Get in touch and apply: https://t.co/atwcWtTyV5 #GenAI #foundationmodels #worldmodels #diffusion #transformers

📢Code release for 3DGS-LM (#ICCV2025)📢 Gaussian-Splatting made faster by changing the underlying optimizer. It's compatible with other changes that accelerate the optimization along different axes. https://t.co/ddGPY1T5KJ

We will present QuickSplat at #ICCV2025! 🎉 Data-driven 2DGS initialization and densification makes 3D surface reconstruction fast & accurate! 🚀 Projcet: https://t.co/SKJ7XRWTVt Arxiv:

ScaffoldAvatar: High-Fidelity Gaussian Avatars with Patch Expressions (#SIGGRAPH) We reconstruct ultra-high fidelity photorealistic 3D avatars capable of generating realistic and high-quality animations including freckles and other fine facial details. We operate on patch-based

We hit our first $1M+ day at @synthesiaIO a few weeks ago. 5 years ago, we literally danced on the table when our first paying customer – a math professor – put down his credit card to pay $30 for the v1 (SS below). Time flies 🥹

We will present Avat3r at #ICCV2025! 🥳 Avat3r brings animation to Large Reconstruction Models. One surprising finding was that we can get rid of any template-based deformation modeling and simply use cross-attention to an abstract facial expression code. https://t.co/EqyZcVbu4J

Stunning voice model by @synthesiaIO: *𝐄𝐗𝐏𝐑𝐄𝐒𝐒-𝐕𝐨𝐢𝐜𝐞* -> new SotA that perseveres identity, accent, expressiveness w/o fine-tuning -> two-stage transformer (AR + NAR), each with 800M params -> curriculum training with QK-layer normalization https://t.co/F4oxWniPNx

Matthias Niessner @ ICVSS 2025!

TL;DR RGB-D scan as input -> compact, CAD scene representation that also features materials in order to create a digital copy that features the looks of a real environment. Great work by @ZheningHuang in collaboration with @XiaoyangWu_, F. Zhong, @HengshuangZhao, J. Lasenby