Multi-View 3D Point Tracking
Frano Rajič, @haofeixu.bsky.social, Marko Mihajlovic, Siyuan Li, Irem Demir, Emircan Gündoğdu, Lei Ke, Sergey Prokudin, @marcpollefeys.bsky.social, Siyu Tang
arxiv.org/abs/2508.21060
@marcpollefeys.bsky.social
Prof ETH Zürich, Director Microsoft Spatial AI Lab, CV/ML/Robotics
Multi-View 3D Point Tracking
Frano Rajič, @haofeixu.bsky.social, Marko Mihajlovic, Siyuan Li, Irem Demir, Emircan Gündoğdu, Lei Ke, Sergey Prokudin, @marcpollefeys.bsky.social, Siyu Tang
arxiv.org/abs/2508.21060
📰Paper: arxiv.org/abs/2501.04597
🔥Code: github.com/cvg/Frontier...
🌍Page: boysun045.github.io/FrontierNet-...
w/ Boyang Sun, Hanzhi Chen, Stefan Leutenegger, Cesar Cadena, @marcpollefeys.bsky.social
Minimal Solvers for Full DoF Motion Estimation from Asynchronous Tracks
@petrhruby.bsky.social, @marcpollefeys.bsky.social
arxiv.org/abs/2508.17537
1/4 🚀 We’re excited to release MVTracker (ICCV 2025 Oral), the first data-driven multi-view 3D point tracker. MVTracker tracks arbitrary 3D points across multiple cameras, handling occlusions and varied camera setups without per-sequence optimization.
29.08.2025 12:38 — 👍 8 🔁 2 💬 1 📌 1MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion
Zador Pataki, @pesarlin.bsky.social Johannes L. Schonberger, @marcpollefeys.bsky.social
tl;dr: using monodepth to reconstruct w/o co-visible triplets. Many ablations and details. M3Dv2 FTW
demuc.de/papers/patak...
FlowR: Flowing from Sparse to Dense 3D Reconstructions
@tobiasfshr.bsky.social, Samuel Rota Bulò, @yung-hsuyang.bsky.social, @nikv9.bsky.social, Lorenzo Porzi, Norman Müller, Katja Schwarz, Jonathon Luiten, @marcpollefeys.bsky.social, Peter Kontschieder
arxiv.org/abs/2504.01647
Collaboration of ETH Zürich, Tübingen AI Center, Microsoft and University of Bonn: @haofeixu.bsky.social @songyoupeng.bsky.social @hermannblum.bsky.social @danielbarath.bsky.social @marcpollefeys.bsky.social
24.04.2025 08:58 — 👍 5 🔁 1 💬 0 📌 0Faculty position in Computer Vision at ETH Zurich: ethz.ch/en/the-eth-z...
20.12.2024 19:52 — 👍 23 🔁 2 💬 0 📌 0This should actually be drawn in 512-dim space ;-)
27.11.2024 11:22 — 👍 11 🔁 0 💬 1 📌 0