PhD student at ETH Zurich & University of Tübingen. Training neural networks to solve 3D and 4D.
https://haofeixu.github.io/
Haofei Xu
Loading...
🎯 Large displacement motion estimation (optical flow, point tracking) has been a long-standing challenge. MegaFlow proposes a simple solution by leveraging pre-trained vision priors. Combined with a global matching formulation, we tackle both optical flow and point tracking in a unified structure.
🚀 Introducing MegaFlow: Zero-Shot Large Displacement Optical Flow!
🔥 Code, models, and demo — all available now!
It's incredibly exciting to see these old ideas shine again, empowered by modern architectures and tools. It’s equally rewarding to watch young students grow, and I'm so happy to be able to support their research journey!
A little backstory: we developed GMFlow (CVPR 2022, github.com/haofeixu/gmflow) and UniMatch (TPAMI 2023, github.com/autonomousvisi…) about 4 years ago.
🏆 SOTA zero-shot performance across multiple benchmarks and in-the-wild videos.
Check out the links below to see it in action! 👇
🌐 Project: kristen-z.github.io/projects/meg...
📄 Paper: arxiv.org/abs/2603.25739
💻 Code: github.com/cvg/megaflow
This project was done by our truly exceptional Master's student @dingxi-z.bsky.social (an incoming PhD student at ETH and MPI), together with amazing @fangjinhuawang.bsky.social and @marcpollefeys.bsky.social !
Today @kyutai-labs.bsky.social and @ellisinsttue.bsky.social launch @kesai.eu!
Robot learning is bottlenecked by the cost of physical interaction. Our mission is to advance the efficiency frontier of robust & safe physical AI through fully open and reproducible research.
kesai.eu/blog/2026-05...
Haofei Xu
Haofei Xu
Haofei Xu
Haofei Xu
kesai.eu
Kyutai and ELLIS Tübingen announce the official launch of KE:SAI – Kyutai ELLIS Scalable Autonomous Intelligence, a non-profit open science lab for physical AI.
