←
Ryousuke Yamada500 posts
Ryousuke Yamada avatar

Ryousuke Yamada

@FragileGoodwill

Research Scientist @ AIST | Visiting Postdoc @ UTN | HQ @ cvpaper.challenge

◷ Joined June 2017▧ 57 Photos and videos
Arpit Jadon avatar
⟳ Ryousuke Yamada retweeted
Arpit Jadon@arpitjadon_·Nov 13

Meet 𝗢𝗽𝗲𝗻 𝗩𝗶𝘀𝗶𝗼𝗻 𝗕𝗮𝗻𝗮𝗻𝗮 🍌 an open-weight alternative to @GoogleDeepMind's Vision Banana: one image generator fine-tuned to output segmentation, depth and normals as images while largely preserving native image editing abilities. 🌐 openvisionbanana.github.io/ 🚀 huggingface.co/spaces/huggin… 📄 arxiv.org/abs/2611.15844 💻github.com/OpenVisionBanana/… Work in collaboration with @v_pariza @jonaruthardt @FragileGoodwill @ASophiaKoepke @abursuc @y_m_asano (Part 1)

0:45
Yuki avatar
⟳ Ryousuke Yamada retweeted
Yuki@y_m_asano·Nov 13

🍌 What if an image generator could see? @GoogleDeepMind's Vision Banana showed it can, but kept data, recipe and weights closed. So we rebuilt it fully open: ✨ FLUX.2 Klein 4B/9B finetune w/ public data only ✨ 73.9 vs Vision Banana’s 69.9 mIoU zero-shot on Cityscapes ✨We release code + 148 checkpoints for the community/ Aaand we innovate: 📈 We develop a new test-time scaling recipe we call “legend-ensembling”: this gives strong gains by applying more compute without retraining anything. 🔗 openvisionbanana.github.io 📄 arxiv.org/abs/2611.15844 A @FunAILab work from A. Jadon, @v_pariza, @FragileGoodwill, @jonaruthardt, @ASophiaKoepke, @abursuc and me.

0:45
OguRyu🇩🇪 avatar
⟳ Ryousuke Yamada retweeted
OguRyu🇩🇪@Oguryu417·Oct 23

🎉 Excited to share that our paper “Where Did the Light Come From? Diagnosing Cast-Shadow Reasoning in VLMs and Borrowing the Missing Shadow Prior from Image Generation Models” has been accepted to #WACV2027 !! arxiv.org/abs/2610.30740 Huge thanks to all collaborators☀️👤

Post attachment
Yuki avatar
⟳ Ryousuke Yamada retweeted
Yuki@y_m_asano·Oct 9

Besides scalability, self-supervised learning is attractive because we have a good real-life example of humans/mammals learning strong visual systems. BUT: they do not learn from shuffled image collections. They learn from (essentially) a single long visual stream. This is what we tackle in our new @NeurIPSConf paper, led by Ivan, together with @LukasKnobel1 and me. 1/ arxiv.org/abs/2609.40333

Post attachment
kohsuke Ide avatar
⟳ Ryousuke Yamada retweeted
kohsuke Ide@IdeKohsuke·Sep 25

Excited to share that our paper was accepted as a Spotlight at #NeurIPS 2026 (E&D track)🎉 Huge thanks to everyone involved in the project! More details coming sometime in October. Now back to the ICLR grind 😅

Post attachment
Francis Engelmann avatar
⟳ Ryousuke Yamada retweeted
Francis Engelmann@FrancisEngelman·Sep 17

🍯 Got exciting 3D vision work that deserves another spotlight -- or a bold idea that isn’t quite a paper yet? Then the #3DV2027 Nectar Track is for you: showcase strong recent work, share early ideas and demos; all while enjoying bougatsa, meze, and rebetiko in beautiful Thessaloniki 🇬🇷 📅 Deadline: Feb 15, 2027 (more details coming soon) 👉 3dvconf.github.io/2027/call-…

Gabriele avatar
⟳ Ryousuke Yamada retweeted
Gabriele@gabrosi3·Sep 7

Going to @eccvconf? 👀 Workshop Radar is ready and online for #ECCV2026 🚀 🔎 Find workshops & tutorials ⭐ Save your favorites 📅 Explore the schedule in one place Open source. Mobile-friendly. No account required. Less tab chaos, more conference. 😄 workshopradar.vercel.app

Post attachment
Hirokatsu Kataoka | 片岡裕雄 avatar
⟳ Ryousuke Yamada retweeted
Hirokatsu Kataoka | 片岡裕雄@HirokatuKataoka·Aug 27

"Visual General Intelligence: A White Paper" is now on arXiv! arxiv.org/abs/2608.25924 We've discussed: - What is Visual Intelligence? - Can we develop it toward General Intelligence (VGI)? In the VGI white paper, we bring together diverse perspectives on Visual Intelligence and discuss how intelligence grounded in visual modalities and experiences could develop toward General Intelligence. Beyond proposing a single definition of VGI, we introduce, propose, and discuss different perspectives on what Visual Intelligence could be and how we could develop it toward General Intelligence. This paper is a starting point for a broader discussion on the future of Visual Intelligence. Many thanks to all the great co-authors!

Post attachment
Shumpei Takezaki avatar
⟳ Ryousuke Yamada retweeted
Shumpei Takezaki@shumpeiMaxwell·Jul 25

Self-Supervised Learning of Structured Dynamics from Videos TL;DR: Structured Dynamics Model decomposes video changes into primary and residual motion. Built on frozen DINOv2 features, it separates camera and object motion with weak supervision and outperforms prior representations on multiple motion tasks. arxiv.org/abs/2607.21576

Ryousuke Yamada avatar
Ryousuke Yamada@FragileGoodwill·Jul 2

ありがとうございました! 色んな方と交流できてとても楽しい時間でした😄

Ryosuke YamadaさんがDFKIで講演してくださってます。最近のCVPRで発表した内容も紹介してくださり、とても興味深いです!🇩🇪🇯🇵

Post attachment
Hirokatsu Kataoka | 片岡裕雄 avatar
⟳ Ryousuke Yamada retweeted
Hirokatsu Kataoka | 片岡裕雄@HirokatuKataoka·Jun 26

[#ECCV2026 LIMIT Workshop] This teaser describes a brief abstract at ECCV 2026 LIMIT Workshop in terms of topics, invited speakers, organizers, and website. Please join us in the workshop. Website: eccv2026-limit-workshop.limi… OpenReView (Paper deadline - 6 July): openreview.net/group?id=thec…

Post attachment

[LIMIT Workshop has been accepted at #ECCV2026 ] LIMIT: Representation Learning with Very Limited Resources explores how we can build powerful AI systems when data, labels, modalities, compute, or supervision are limited. Join us at LIMIT 2026! Website: eccv2026-limit-workshop.limi…⁠

Post attachment
Su Myat Noe avatar
⟳ Ryousuke Yamada retweeted
Su Myat Noe@imsu_myat·Jun 26

📢 Papers are now being accepted for the LIMIT Workshop at #ECCV2026! Topics include representation learning with limited data, labels, compute, and other resource constraints. 📅 Deadline: July 6, 2026 (23:59 AoE) 🔗 eccv2026-limit-workshop.limi…

Post attachment