🎶Can a robot learn to play music? YES! — by teaching itself, one beat at a time 🎼 🥁Introducing Robot Drummer: Learning Rhythmic Skills for Humanoid Drumming 🤖 🔍 For details, check out: robotdrummer.github.io
Ni PP ni VOX aconseguiran esborrar la valencianitat lingüística i cultural de la ciutat d’Alacant. Diuen que ací no es parla valencià però és perquè no han posat l’orella:
🎥 AnyCam: Learning to Recover Camera Poses and Intrinsics from Casual Videos TL;DR: fast transformer-based model for camera parameters; feed-forward; from diverse, unlabeled YouTube videos; uncertainty-based loss and pre-trained depth/flow networks; no direct motion supervision
🚀 Excited to finally share our work: 'HyperNVD: Accelerating Neural Video Decomposition via Hypernetworks' accepted at #CVPR2025 ! Let’s break down how it works in this thread 🧵👇
I think Europe and the UK need their own DeepMind-like AI lab which is not connected to the US in any form
As I wait for the SigLIP2 OpenCLIP and timm checkpoints to upload, a few comments. Like v1, these are very strong image encoders, but even better. They should become a go to ViT encoder for many tasks. Yet, even when we had ViT encoders from SigLIP, DFN CLIP, DataComp CLIP,…
Throughout 2024 my work on end to end RL and legged robots has led to an ICRA finalist, an RSS finalist, a CoRL oral, and a submission to [a journal, to be revealed later, really proud]. I am really lucky to keep working on RL with access to multiple frontier legged systems.
Serious question, when does RL ever leave? And I thought reward/cost engineering is always a thing (way before prompt engineering is a thing).
HOT 🔥 fastest, most precise, and most capable hand control setup ever... Less than $450 and fully open-source 🤯 by @huggingface, @therobotstudio, @NepYope This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀 A thread 🧵
Per no patir, com més tonta millor! 🫶'DEL FANDOM AL TROLEIG. Una sàtira del bla bla bla' de Berta Prieto 🎟️Últimes entrades disponibles. salabeckett.cat/espectacle/del…
A new optimizer that's better than Adam. I think I've heard that before. But, I tried this one, I was doing regression testing on some recent optimizer cleanup w/ real scenarios, I threw this in the mix. It did beat Adam, every time (so far). This one appears worth a closer look.
Our NeurIPS paper is published on arXiv. In this paper, we propose a new optimizer ADOPT, which converges better than Adam in both theory and practice. You can use ADOPT by just replacing one line in your code. arxiv.org/abs/2411.02853
🚀 Open-sourcing CoMix: framework for Comics Understanding! 🎯 Features: - Unified dataset processing 🔒 - Detection benchmarks - Captioning benchmarks (VLM 🤗 models) - Multi-task evals 🔍 Based on: ICDAR'24, NeurIPS'24 & ECCV'24 papers. ⭐️ GitHub: github.com/emanuelevivoli…
Day 3 at #SCEWC2024! At the @govern booth, we are showcasing how computer vision and #AI can enhance citizens' quality of life and improve our territories! It is the last day of the congress, so here’s a look at the impactful initiatives we have been sharing in our exhibit…
#ECCV2024 Catalan @eccvconf dinner at the Bar Magenta in Milan: a diverse and consolidated community in computer vision research.
We are launching Salamandra 2B & 7B multilingual LLMs trained at @BSC_CNS from scratch with nearly 8 trillion tokens in 35 EU languages+code. Spanish languages have been carefully curated, with Romance languages comprising >30% of the training dataset. huggingface.co/collections/BS… 👇
huggingface.co
Salamandra 🦎 - a BSC-LT Collection
Salamandra 🦎 - a BSC-LT Collection
Everyone talks about #AI, yet the vast majority of Bachelor programs in Germany still do not include a single Machine or Deep Learning course. Hence, students can practically only start studying #AI-related topics during their masters which puts them at a severe disadvantage.
Thrilled to announce that SAM 2 is out, bringing SAM level capability now to any video! 🤩🚀 Super excited to see what everyone builds!
Introducing Meta Segment Anything Model 2 (SAM 2) — the first unified model for real-time, promptable object segmentation in images & videos. SAM 2 is available today under Apache 2.0 so that anyone can use it to build their own experiences Details ➡️ go.fb.me/p749s5
Worth thinking about for us working on GNN theory.. Eg. Why not use a fully connected transformer instead?
it's like, "i'm crazy? no you're crazy! ilya fucking sutskever, the guy behind alexnet and openai, created a company called safe superintelligence! sam altman is raising $7 trillion to build The Final Invention. but yeah, i'm sure they're all definitely 100% wrong without a…
What's truly heartwarming and a testament to the atmosphere we've managed to build at EEML'24 is the fact that, even after six highly intensive days, we still have a full house attending the final day 😊 Thank you all so much, you've been a fantastic crowd! ❤️
United States Trends
- 1. Steelers 51.6K posts
- 2. Rodgers 21K posts
- 3. Chargers 36K posts
- 4. Tomlin 8,155 posts
- 5. Schumer 217K posts
- 6. #BoltUp 2,909 posts
- 7. Resign 102K posts
- 8. #TalusLabs N/A
- 9. Tim Kaine 17.8K posts
- 10. #HereWeGo 5,617 posts
- 11. Keenan Allen 4,742 posts
- 12. #RHOP 6,804 posts
- 13. Durbin 25.2K posts
- 14. Herbert 11.5K posts
- 15. Gavin Brindley N/A
- 16. #ITWelcomeToDerry 4,362 posts
- 17. Angus King 15.1K posts
- 18. Ladd 4,362 posts
- 19. 8 Democrats 8,594 posts
- 20. 8 Dems 6,654 posts
Something went wrong.
Something went wrong.