🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Delighted to share an overview of my lab's eight #cvpr 2024 papers. Thanks to my amazing postdoc Sangmin Lee for spearheading the effort and to our invaluable collaborators AI at Meta Reality Labs at Meta Stability AI Toyota USA. See you in Seattle! @IllinoisCS The Grainger College of Engineering

🗣Exploring beyond interpreting actions that directly involve the camera wearer, how can egocentric audio-visual signals aid in understanding the natural social behaviors among all partners, ultimately enhancing our daily communication? #CVPR24 Project: vjwq.github.io/AV-CONV/

🎉Excited to share our #CVPR2024 paper on modeling multimodal social interactions! We explore how to understand social dynamics through verbal (e.g., language) and non-verbal (e.g., gaze, gestures) cues in multi-party environments. (1/4) Paper: arxiv.org/abs/2403.02090

My winter computer vision talk about language models in self-driving is ready for you on youtube: - a lot on LLM based Visual QA evaluation - a fair amount on language models in self-driving - a bit on GAIA and AV2.0 #llm #autonomous #selfdriving #GenAI youtube.com/watch?v=WNaJPy…

Diagnosing mpox with AI - Survey for clinicians now available in French, Spanish, Chinese, and English. If you are a clinician (nurse, doctor, PA, NP, student) please click below or else share with any clinicians: nyumc.qualtrics.com/jfe/form/SV_8e…

Our work accepted at #MICCAI2024 We leverage finetuned diffusion model to simulate disease progression. This approach generates realistic diseases progressions WITHOUT training on continuous observations. As we get more access to medical data and scale up I believe this will

Come to #CVPR 2024 Poster #220 this morning to chat with our team about how to build the Vision-Language Model of the HD map and traffic scene understanding task. The updated version of our benchmark will serve as a new challenge dataset in the ITSC 2024 workshop. James Matthew Rehg

Crypto是一个自由的世界，我们尊重一切最终能创造价值的项目，包括DeSci。我们相信只有自由，才能发挥人最大的主观能动性。但是大部分的科研工作者是不自由的，ta们受困于Funding，受困于环境，需要各种外界的帮助。 我们ABCDE

I keep saying that naming your research after OpenAI products is cringe. AutoGPT? Mini-Omni? LLaVA-o1? OpenSORA? You haven't seen the weights, you have little clue about the design, you're just latching onto the brand hype train. Have some dignity. At least go after Anthropic.

A delightful Sunday at #ICLR2025 in the Pediatric AI workshop pediamedai.com/ai4chl/ listening to an exciting talk by Jason Alan Fries describing his exciting work with Nigam Shah Stanford Health Care and others!

Thank you! Thrilled to announce the release of SocialGesture — the first dataset and VQA benchmark for multi-person gesture understanding in social interaction scenes! More exciting work based on this research is coming soon. Stay tuned! #CVPR2025