Tho Le

I'll soon present our CVPR work JRDB-PanoTrack, an open world panoptic segmentation and tracking benchmark. See you soon at poster session 5, 10:30AM Arch 4A-E Poster #267 #cvpr2024 @HamidRezatofigh @JRDB_Benchmark @poppingG5

Excited to present "DifFUSER"—our new diffusion model for robust multi-sensor fusion in 3D object detection & BEV segmentation at #ECCV2024! 🎉 📍 Find us: Poster 119, Session 2, on Oct 1st, 16:30–18:30 Let's discuss #DiffusionModels & #ComputerVision! ldtho.github.io/DifFUSER

Are you planning to attend @CVPR? Join us at the Robot Visual Perception in Human Crowded Environments workshop! Seattle Convention Center - Seattle , USA Date: Tuesday, June 18 Time: 1:30 PM–5:30 PM Room: Arch 210 More info: jrdb.erc.monash.edu/workshops/cvpr…

JRDB dataset has been trusted by top researchers globally, from leading academic institutions to industries. Curious about how JRDB dataset helps accelerate robotic research? Join us at our CVPR workshop on June 18th 1:30 PM to 5:30 PM. jrdb.erc.monash.edu/workshops/cvpr… @JRDB_Benchmark

Our JRDB- PanoTrack Dataset (@CVPR paper) along with all the evaluation & visualisation toolkits are now publicly available. The Closed/Open-world panoptic segmentation and tracking leaderboards for #CVPR2024 workshop challenges will be up soon.

Computer Vision conference's acceptance criteria these days: #CVPR2024 #eccc2024 #AI #ComputerVision

One thing we know is that if future AI systems are built on the same blueprint as current Auto-Regressive LLMs, they may become highly knowledgeable but they will still be dumb. They will still hallucinate, they will still be difficult to control, and they will still merely…