Anthony Hu

Launching @kyutai_labs x @Gen_intuition - research collab at the frontier World models and agents capable of spatiotemporal reasoning are the next frontier. As builders of this technology, we have a responsibility to educate. This is impossible without both unique data and world-leading talent. Through our research collaboration, both labs will share data and talent seamlessly. - The necessary data lives inside only a few companies, and isn’t available on the internet. @gen_intuition owns the largest, most diverse, ground truth action labelled dataset. @Kyutai_labs utilizes the data for open research, resuming their culture of publishing on the frontier - Talent moves between the two labs, switching freely between open-ended research and application Join us in NYC and Europe - open roles below jobs.ashbyhq.com/medal & kyutai.homerun.co/technical-staf…

It's a unique playground to build agents with a general intuition. Working with the @gen_intuition team has been so much fun so far, and I can't wait to explore what's ahead of us!

I'm starting a new journey building general agents and world models for video games at General Intuition. The past 6 years at Wayve working on self-driving cars have been an absolute blast, where I've learned so much alongside wonderful colleagues.

We’re starting a company to build the next frontier of embodied intelligence - General Intuition We believe the time has come to move from words to worlds, and we’re developing foundation models capable of acting in any environment. Starting with video games!

Introducing GAIA-2 🌎Generative world modeling just stepped up a gear. GAIA-2 is the latest development of Wayve’s video-generative world model tailored for driving. GAIA-2 offers richer, more realistic, and highly controllable synthetic driving scenarios, accelerating Wayve’s path to safe driver assistance and automated driving at scale. Learn more about GAIA-2 in our Blog: wayve.ai/thinking/gaia-… #GAIA2 #GAIA #EmbodiedAI

It’s #GAIA1 world, we just drive in it. Generate realistic driving videos using only prompts. See how it works below 🧵

Wayve presents Model-based Imitation LEarning (MILE): a method to jointly learn a world model and a driving policy from an offline corpus of driving data. This model is another step in achieving a more generalisable driver - paper to appear at #NeurIPS2022 wayve.ai/blog/learning-…

Our paper "Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation" will be presented in Session 10 at #ICCV2021. youtube.com/watch?v=mTy85t… Timeslots: Oct 13 (Wed) 5-6PM (EDT) Oct 15 (Fri) 10-11AM (EDT) w/ @IgnasBud @robertocipolla

Very excited to present our paper "Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wild" at #ICCV2021! w/ @IgnasBud and @robertocipolla Paper + code: github.com/akashsengupta1…

@wayve_ai future prediction for autonomous driving is being presented as an oral presentation (top 3.4% acceptance rate) at ICCV this week. Congrats to the team who made this possible! @hu_anth @ZMurez @NikhilM37190363 Sofía Dudas @jeffrey_hawke @vijaycivs @robertocipolla

Today we are sharing a new approach to prediction in dynamic urban driving scenes, one of driving's hardest problems. Our algorithm, FIERY, is able to predict multimodal futures in bird's eye view from monocular cameras. Blog, research paper and code here: wayve.ai/blog/fiery-fut…

Our computer vision team will be at the ECCV conference virtual poster session today to present our work on “Probabilistic Future Prediction for Video Scene Understanding” at 2pm (UTC+1) and then at midnight (UTC+1). Looking forward to seeing you there. arxiv.org/abs/2003.06409

Thrilled to present our research on probabilistic future prediction of complex real-world urban scenes at #ECCV2020! Huge thanks to the @wayve_ai team for making this work possible. anthonyhu.github.io/research/proba…

High dimensional, multi-modal probabilistic scene prediction has fascinated me for a long time and I'm excited to share our first attempt to model this problem. A critical component for autonomous driving and model-based reinforcement learning.

Ever wondered about how to structure a deep learning codebase, run reproducible experiments, and efficiently work remotely? Check out my guide: anthonyhu.github.io/research-workf… #research #machinelearning #pytorch #deeplearning

Happy to share our final paper of 2019 proposing a spatial-temporal video-pixel embedding loss which sets a new state of the art for video instance segmentation. Here's the paper arxiv.org/abs/1912.08969 and demo video youtu.be/dc-3meFF6z0