Dima Damen

3.9K posts

Dima Damen banner
Dima Damen

Dima Damen

@dimadamen

Professor of Computer Vision, @bristolcs. Senior Staff Research Scientist @GoogleDeepMind - passionate about the temporal stream in our lives.

Bristol, UK Katılım Nisan 2013
734 Takip Edilen11K Takipçiler
Sabitlenmiş Tweet
Dima Damen
Dima Damen@dimadamen·
🛑📢 HD-EPIC: A Highly-Detailed Egocentric Video Dataset hd-epic.github.io arxiv.org/abs/2502.04144 New collected videos 263 annotations/min: recipe, nutrition, actions, sounds, 3D object movement &fixture associations, masks. 26K VQA benchmark to challenge current VLMs 1/N
English
3
24
123
21K
Dima Damen retweetledi
Jack Parker-Holder
Jack Parker-Holder@jparkerholder·
Last year on 60 minutes we talked about the potential to ground our world models with Street View data. Thanks to hard work from @poolio and others this is now possible—Genie can simulate worlds based on real locations, and you can try it too!! Check out the blog for more info 🧞🚀
Ben Poole@poolio

Real-world models are here! Stoked to share how we're bringing real-world locations to life by integrating Street View into Genie. Try it now at labs.google/fx/projectgenie and read the blog for more info: blog.google/innovation-and…

English
5
5
65
9K
Dima Damen retweetledi
Jianyuan
Jianyuan@jianyuan_wang·
Introducing VGGT-Ω: scaling feed-forward reconstruction across static and dynamic scenes, and studying whether the learned geometric representations transfer beyond reconstruction.
English
13
142
1K
756.3K
Dima Damen
Dima Damen@dimadamen·
Congratulations @zhifan_zhu for being selected amongst the top 5% of reviewers @CVPR #CVPR2026. Zhifan has just submitted his thesis awaiting the oral exam / viva @bristolcs @BristolUni. #proudAdvisor #CVPR2026
#CVPR2026@CVPR

We are grateful to all of the 17,491 reviewers who helped make #CVPR2026 possible. We are especially pleased to recognize the following Outstanding Reviewers, whose high-quality reviews (as judged by their Area Chairs) placed them among the top 5% of reviewers.

English
0
2
21
2.7K
Dima Damen
Dima Damen@dimadamen·
So proud of my ex-students now outstanding ACs @CVPR #CVPR2026 @mwray0 @bristolcs @BristolUni @davmoltisanti @UniofBath @doughty_hazel @UniLeidenNews This is more important than papers... 3/3 of ex-student ACs are outstanding ACs ☺️
#CVPR2026@CVPR

Thank you to all the #CVPR2026 Area Chairs for steering the review process. We would like to give special recognition to the following Outstanding Area Chairs for their exceptional efforts and leadership.

English
2
1
20
2K
Dima Damen retweetledi
Elliott / Shangzhe Wu
Elliott / Shangzhe Wu@elliottszwu·
Check out Ariticraft 🦾 - a highly efficient agentic system that generates articulated 3D assets fully automatically at scale! 🚀 articraft3d.github.io
RuiningLi@RayLi234

🚀 Introducing Articraft, a coding agent for articulated 3D asset creation. Articraft writes code, executes it, receives validation feedback, and refines the result into simulation-ready 3D assets with parts, joints, and motion. We’re also releasing Articraft-10K: 10,000+ articulated objects across 250 categories, unlocking large-scale interactive scenes for robotics simulation and physical AI. 🔗 Project page: articraft3d.github.io 💻 Code: github.com/mattzh72/artic…

English
18
114
1.1K
215.6K
Dima Damen retweetledi
Vinayak Gupta
Vinayak Gupta@Vinayak22Gupta·
We compared against World Labs in a similar setting and it: 🏠 Hallucinates a second house on top of the existing roof. 🟢 Drops a green roof visible in almost every input. 💡 Locked to one specific lighting. 🚶 Bakes pedestrians into the scene. GenWildSplat handles all four✅
English
1
2
20
2.1K
Dima Damen
Dima Damen@dimadamen·
Nice reminder about the difference between policy learning and world models. Actions in world models can be random. ⁦@iclr_conf⁩ workshop on world models is still going strong. Thanks ⁦@Mengyue_Yang_⁩ great org!
Dima Damen tweet media
English
0
2
26
1.4K
Dima Damen
Dima Damen@dimadamen·
Now @iclr_conf 202 A/B - workshop on World Models @michaelrabbat reviewing the space of world models trained typically on narrow domains.
Dima Damen tweet media
English
0
1
13
1K
Dima Damen
Dima Damen@dimadamen·
From the V-Jepa2 presentation: Action anticipation in EPIC-KITCHENS is "a very challenging task and still there’s head room for improvement on this task". V-Jepa2 establishes a new SOTA but still <65% on both verb and noun acc. @mido_assran @iclr_conf workshop on World Models
Dima Damen tweet media
English
0
4
16
1.4K
Dima Damen
Dima Damen@dimadamen·
Packed room! An RL perspective on world models
Dima Damen tweet media
English
0
1
5
385
Dima Damen
Dima Damen@dimadamen·
A great line up of speakers
Dima Damen tweet media
English
0
0
1
355
Dima Damen
Dima Damen@dimadamen·
* AI (unlike humans) is trained with carefully curated text * Complementary learning systems theory gives a direction into how new knowledge is formed in the brain * Understanding the brain is one direction to make AI more efficient 3/3
English
0
0
4
428
Dima Damen
Dima Damen@dimadamen·
* People don’t just store a representation of the sentence of what was communicated to them. We also build our own representations. * The ability to organise behaviors in pursuit for goals including mastering and inventing new skills is the hallmark of what we consider Genius 2/3
English
1
0
4
520
Dima Damen
Dima Damen@dimadamen·
I so much enjoyed Jay McClelland's talk @iclr_conf #ICLR2026 workshop on New Frontiers in Associative Memory. Some quotes: * AI is staged until deployed and then they are frozen at inference... AI systems are not continual learners 1/3
English
2
1
21
2.6K
Dima Damen
Dima Damen@dimadamen·
For those attending @iclr_conf maybe we need a reminder that a panel moderator should be "like a great referee: if they did their job perfectly, you barely noticed they were there, but the game moved beautifully because of them."
English
5
0
17
1.9K

Keşfet

@poolio @zhifan_zhu @CVPR @bristolcs @BristolUni @mwray0 @davmoltisanti @UniofBath