Torsten Sattler
@SattlerTorsten
Computer Vision researcher, academic
Desired for accurate camera poses for long sequences in dynamic scenes, and also dense 3D reconstruction? Check out our #CVPR25 paper WildGS-SLAM – bridging robust SLAM and high-fidelity 3D Gaussians in the wild!
🥳Excited to share our latest work, WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments, accepted to #CVPR2025 🌐 We present a robust monocular RGB SLAM system that uses uncertainty-aware tracking and mapping to handle dynamic scenes.
🚨🚨 New #SLAM #RealityCapture Dataset! 🚨🚨 The Oxford Spires Dataset. New mapping benchmark in Oxford's famous colleges and Blenheim Palace focusing on high quality reconstruction (#nerf #gaussiansplatting) and colorization. dynamic.robots.ox.ac.uk/datasets/oxfor… #spatialai #oxford
Understanding and reconstructing the 3D world are at the heart of computer vision and graphics. At #CVPR2025, we’ve seen many exciting works in 3D vision. If you're pushing the boundaries, please consider submitting your work to #3DV2026 in Vancouver! (Deadline: Aug. 18, 2025)
We have extended the deadline for paper submissions to July 1st.
Working on camera calibration or camera pose estimation? Want to finally know how calibrations and poses are obtained for the data you ae using? Come join us at the CALIPOSE workshop at @ICCVConference . Details are on the workshop website: cmt3.research.microsoft.com/CALIPOSE2025
Interested in 3D Vision, 3D Scene Understanding and 3D Generation? We have open PhD/PostDoc positions in Lugano, Switzerland - I'm at #CVPR2025, please reach out 🚀 or send me an email @CVPR
Attending @CVPR and looking for a PhD or postdoc position in the area of 3d reconstruction (Gaussian splatting, nerfs, scene understanding, etc.)? Find me or drop me an email ;)
Come to our poster and learn about our work on privacy-preserving visual localization.
One of the papers is #GaussianSplatting Feature Fields for #PrivacyPreserving #VisualLocalization by Maxime Pietrantoni, @kgcs96 and @SattlerTorsten. If interested in the subject, Maxime will present his work tomorrow at #CVPR2025, Poster Session 1, Poster 85, ExHall D.
🚀 Just wrapped up my internship at Google with a paper submission! LODGE is a new large-scale level-of-detail 3DGS method, enabling efficient rendering even on mobile devices. Thanks to the amazing team at Google Zurich for making this possible! 📄lodge-gs.github.io
🚀🚀🚀Announcing our $13M funding round to build the next generation of AI: 𝐒𝐩𝐚𝐭𝐢𝐚𝐥 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬 that can generate entire 3D environments anchored in space & time. 🚀🚀🚀 Interested? Join our world-class team: 🌍 spaitial.ai #GenAI #3DAI
Large-scale visual SLAM for in-the-wild videos Shuo Sun, @SattlerTorsten, Malcolm Mielle, Achim J. Lilienthal, Martin Magnusson tl;dr: PVO->base; RAFT+GLOMAP->initialization; Mask2Former->dynamic objects; Metric3D->BA; SIM(3) PGO->loop closure arxiv.org/abs/2504.20496
A Guide to Structureless Visual Localization Vojtech Panek, Qunjie Zhou, Yaqing Ding, Sérgio Agostinho, @ZKukelova, @SattlerTorsten, @lealtaixe tl;dr: structureless localization review arxiv.org/abs/2504.17636
Applications are now 📣OPEN📣 for #PAISS2025 4th edition of THE AI summer school #Grenoble 1-5 Sept! Speakers so far @ylecun @JeromeRevaud @dimadamen @mcaron31 @ArthurGretton @gabrielpeyre @le_science4all Alex Cristia Justin Carpentier Mariia Vladimirova paiss.inria.fr
Preprint of today: Huang et al., "3R-GS: Best Practice in Optimizing Camera Poses Along with 3DGS" -- zsh523.github.io/3R-GS/ Accurate camera poses matter a lot for high-quality 3D reconstruction. Use 3DGS-MCMC & MLP-based global pose refinement & adherence to SfM correspondence
A thread of thoughts on radiance fields, from my keynote at 3DV: Radiance fields have had 3 distinct generations. First was NeRF: just posenc and a tiny MLP. This was slow to train but worked really well, and it was unusually compressed --- The NeRF was smaller than the images.
Spatial AI is increasingly important, and the newest papers from #NVIDIAResearch, 3DGRT and 3DGUT, represent significant advancements in enabling researchers and developers to explore and innovate with 3D Gaussian Splatting techniques. 💎 3DGRT (Gaussian Ray Tracing) ➡️…
📢 This is mostly addressed to the 3D reconstruction fanatics: We uploaded our Eurographics 2025 paper. We just went full nerd-mode and investigated in extreme detail if and how the approximations of the 3DGS renderer compared to accurate volumetric rendering matters for the…
I'm excited to present "Fillerbuster: Multi-View Scene Completion for Casual Captures"! This is work with my amazing collaborators @Normanisation, @yash2kant, Vasu Agrawal, @MZollhoefer, @akanazawa, @c_richardt during my internship at Meta Reality Labs. ethanweber.me/fillerbuster/