RayFronts code has been released ! github.com/RayFronts/RayF… 🤖 Guide your robot with semantics within & beyond depth. 🖼️ Stop using slow SAM crops + CLIP pipelines. RayFronts gets dense language aligned features in one forward pass. 🚀 Test your mapping ideas in our pipeline !

Introducing UFM, a Unified Flow & Matching model, which for the first time shows that the unification of optical flow and image matching tasks is mutually beneficial and achieves SOTA. Check out UFM’s matching in action below! 👇 🌐 Website: uniflowmatch.github.io 🧵👇

Catch Deva Ramanan presenting UFM at the #CVPR2025 Workshop on 4D Vision in 20 minutes (3:30 pm CDT)! 4dvisionworkshop.github.io Location: 104 B, Music City Center x.com/YuchenZhan5425…

WTF? Comes with code. See in comment below 👇

Woah. First person camera view being warped to a third person camera view using a unified flow and matching model.

UFM: A Simple Path towards Unified Dense Correspondence with Flow Yuchen Zhang, Nikhil Keetha, Chenwei Lyu, Bhuvan Jhamb, Yutian Chen, Yuheng Qiu, Jay Karhade, Shreyas Jha, Yaoyu HU, Deva Ramanan, Sebastian Scherer, Wenshan Wang tl;dr: transformer-based architecture using covisibility

UFM: A Simple Path towards Unified Dense Correspondence with Flow Yuchen Zhang, Nikhil Keetha et 10 al tl;dr: wide-baseline and optical flow data help each other. DINO layer selection. No matching, just flow regression with RoMA- loss. A LOT of ablations 1/ arxiv.org/abs/2506.09278

Funny cycle - very small, but challenging, test-only WxBS dataset is becoming relevant for image matching again. ducha-aiki.github.io/wide-baseline-…

Catch our team Parv Kapoor Jay Patrikar AirLab presenting and demoing ViSafe at #RSS2025 tomorrow! We'll be showing our payload demo & high speed aerial collision avoidance results 🚀

come checkout our demo on how our setup does precise detections for far away agents🕵️ that are sometimes invisible to the naked eye

Want to learn how to empower 🤖 with real-time scene understanding and exploration capabilities? Catch Me, Cherie Ho & Yuheng Qiu presenting RayFronts at #RSS2025 SemRob Workshop (OHE 122) & Epstein Plaza at 10:00 am PST Today!

Awesome and insightful panel about the past, present and future of robotics!

ViSafe’s vision-only approach offers a lightweight, passive alternative that maintains strong safety performance, enabling broader deployment on agile, resource-constrained platforms. ✈️👀 #CMUrobotics Read the latest RI article covering the team's work! ri.cmu.edu/visafe-smarter…

Lots of awesome technical nuggets similar to this in the panel! Checkout the key takeaways here: linkedin.com/posts/nik-v9_r…

Only after labelling a dataset by your own you know how dirty it is

Excited to share recent work with Kaihua Chen and Deva Ramanan where we learn to do novel view synthesis for dynamic scenes in a self-supervised manner, only from 2D videos! webpage: cog-nvs.github.io arxiv: arxiv.org/abs/2507.12646 code (soon): github.com/Kaihua-Chen/co…

I'm very surprised at how quickly π³ can create a 3D model from a video that was just randomly generated by Kling AI 2.1 Master.