We are happy to announce a new vacancy for a PDRA position in Computer Vision to work with Prof. Andrew Zisserman; more info here - eng.ox.ac.uk/jobs/job-detai… Application deadline is 17th March at 12 noon. Good luck!

We are looking for a student researcher to work on video understanding plus 3D, in Google DeepMind London. DM/Email me or pass it to someone if you feel it may be a good fit!

📢#DeepSeekR1 is amazing ! How does it work for medical tasks, and more importantly, do these reasoning / thinking process make sense ? We present you a comprehensive evaluation on popular LLMs on examination recommendation, disease diagnosis, treatment planning. paper:

📢 #ICLR2025 We’re presenting Track-On: a simple transformer-based model for online point tracking. It runs frame-by-frame, and uses no future information. Come by and chat! ⏰ Friday, April 25 | 3-5:30pm SGT 📍 Hall 3 + Hall 2B | Poster #101 🔗 kuis-ai.github.io/track_on

Check this out ! Something I have been really fascinated ! AIs for Soccer Games ! ⚽️⚽️⚽️⚽️

For the soccer fans, don't miss this !!!!⚽️⚽️⚽️⚽️⚽️

Fun work on AI4Soccer ! ⚽️⚽️⚽️⚽️

We’re hosting the 1st SLoMO Workshop at #ICCV2025 to discuss Story-Level Movie Understanding & Audio Descriptions !

Thanks for sharing !😊

🚀 Excited to announce that StreamFormer is accepted to #ICCV2025 as an Oral paper! Congrats to my excellent coauthors Yibin Yan Weidi Xie "Learning Streaming Video Representation via Multitask Training"! 📄Paper: arxiv.org/abs/2504.20041 🌐Web: go2heart.github.io/streamformer

Oral@#ICCV2025 ! Only a starting point for streaming video representation learning, stay tuned, a lot more interesting work to come !

After almost two years, RadFM is finally published Nature Communications Thanks for all the work from reviewers and editors. 🥳🥳 nature.com/articles/s4146… Please check the papers with changes on title: Towards generalist foundation model for radiology by leveraging web-scale 2D&3D

🚀 Excited to introduce our new work, Deep-DxSearch End-to-End Agentic RAG System Training for Traceable Diagnostic Reasoning (arxiv.org/pdf/2508.15746) 🧐Objective: the goal is to transform human-prior guided LLM-powered medical RAG into end-to-end data-driven optimized agentic