Brewing up your new cool Robotics + NeRF papers? Submit your work to our Neural Fields in Robotics workshop for a chance to present at the poster session during #ICRA2024! 🦾 📢Extended Deadline: March 31st robonerf.github.io/2024/

Extremely proud of James Smith and Ram Ramrakhya for winning college-level awards for their research! James won the College of Computing Graduate Research Assistant Award and Ram won the College of Computing Rising Star Doctoral Student Research Award! Well deserved!

📢 9 spotlights, 12 posters accepted to the first workshop on Neural Fields in Robotic. A range of topics will be presented on #NeRF and #Robotics with orals and posters on May 17th at #ICRA2024! Come check these out. 🌐List of papers: robonerf.github.io/2024/

Excited to attend #icra2024 in Japan! Should be a fun conference to see the progress in robotics.

Excited to be giving a contributed oral talk tomorrow at the #GenAI4DM Workshop at #ICLR2024 about our latest work harnessing pre-trained diffusion models as vision-language representations learners that excel across a wide variety of control tasks! Details in 🧵below!

Visit our two Thursday papers and Friday RoboNerf workshop at #icra2024! Self-supervised mesh+pose estimation, collaborative perception, and NeRFs. You can't go wrong.

At my final technical talk ICRA with 🐝🐝.Zsolt Kira and team

Kicking off the RoboNeRF workshop at #icra2024 with Jeannette Bohg talking about NeRFs for manipulation! Come by Conference Center 419 to check it out! Great organization by Zubair Irshad and Nick Heppert.

I enjoyed moderating the great set of panelists. The question of where 3D/NeRFs fit into gpt4o type multimodal models got some interesting responses!

Extremely proud of Zubair's impactful body of work in this thesis! Happily we continue to collaborate so it is not as bittersweet :) #proudadvisor

Starting soon! Visit our ICE-G #cvpr workshop poster for condition editing of splats. The second one on scaling NeRF-based 3D learning is after: 📅 June 17th, 3:40-4:10 pm 🗺 Summit 332 🌐nerf-mae.github.io

Check out our three #cvpr main conference papers if you're in Seattle. Topics range from free unsupervised segmentation from diffusion models to multimodal embodied AI!

There are many vision-language-action (VLA) models rolling out, but their action tokenization schemes vary widely. In a new preprint we do a unified investigation, for discrete/continuous action spaces, across standardized settings for easy comparison! arxiv.org/abs/2406.07904

🚀 Introducing our #CVPR2024 paper GOAT-Bench 🐐! A Benchmark for Multi-Modal Lifelong Navigation. GOAT-Bench evaluates embodied navigation agents under: 🧠 Lifelong learning: Unlike episodic benchmarks, agents are tasked with navigating to multiple goals where it should

Come by poster 159 in the afternoon #cvpr poster session to see the unseen!

#RoboNerF workshop at #ICRA'24 was a great success! Missed attending in person? We have released the video recording of the entire workshop. Check it out below: 🎥Youtube link: youtu.be/jyEZtbXs3fg?si

Submit to the #ECCV2024 workshop on Foundation Models for downstream applications (including robotics)! Deadline Monday 07/22. Excited to be speaking at this workshop in Milan!

Congratulations to our Georgia Tech School of Interactive Computing colleagues for their achievements at ICML Conference!🎉 Irfan Essa: Best Paper🏆 for VideoPoet with Google DeepMind (including several GT alumni at Google Atlanta) and Judy Hoffman: Test of Time🏅 for DeCAF! More from GT this year: sites.gatech.edu/research/icml-…

3D representation learning can be difficult to scale. Why not use NeRFs + MAE as an unsupervised source for 3D learning? Check out our #ECCV2024 paper showing that this can be scaled and applied to many downstream 3D tasks Amazing work by Zubair Irshad with TRI. Georgia Tech School of Interactive Computing Machine Learning at Georgia Tech