Adobe Research Scientist doing computational photography stuff, previously Princeton PhD. All opinions are my own.
Joined April 2022
- Tweets206
- Following118
- Followers268
- Likes260
Pinned Tweet
#CVPR2026 Sunday 11:45AM-1:45PM ExHall F 102! We'll be presenting Hist2Style -- headed by @dekelgalor
-- an interactive stylization pipeline which distills hallucinatory (but cool) generative image edits into non-hallucinatory (and fast) bilateral grid tone and color edits.
Ilya Chugunov retweeted
Infinigen2.0.0a3 now has procedural generators for 6 room types, several new procedural objects and materials, and a prototype system for multi-room house tours. See thread for videos, a visual changelog, and release notes.
Ilya Chugunov retweeted
Real-world scenes are highly repetitive 🧱🌿: bricks, blades of grass, leaves, and many other structures appear again and again. At #NeurIPS2026, we present SCION, which learns a compact vocabulary of reusable 3D primitives and instantiates them throughout a scene, capturing its underlying compositional structure. This brings learned photorealistic 3D Scenes down to the ~1 MB range 🤏 while introducing natural structures for editing and animation 🎬.
Great work by @williamkoch314, @amogh7joshi, @cvachha, and Cheng Zheng.
You'll see the authors in Sydney this December!
Project Page: light.princeton.edu/SCION
Paper: arxiv.org/abs/2610.02322
I think maybe the solution to paper reviewing is to invert the conflict of interest section.
If I'm the most relevant person to look at your submitted work then we better have at least a small conflict of interest somewhere.
why does the genetic thing that makes you lactose intolerant also make you addicted to cheesecake, someone should study this
Ilya Chugunov retweeted
📸 It has become an annual tradition for ICCP to host a photo challenge. This year we introduced a track for creative use of AI in image editing. This years winners are, drumroll please… 🥁
- Mel White
- Sanjeev Koppal (AI category)
See our website for all submissions!
wow what a cool city that everyone should visit and go rock climbing with ilya at
It has been a blast sharing the excitement, science, art, and community of CVPR with all of you.
The Publicity Chairs are officially signing off.
Seattle, you’re up next! 🌲 See you at #CVPR2027! 👋
@deblinaforAI @anfurnari @CSProfKGD @YVinker @_vztu
My advice to PhD students in 2026:
eat more dietary fiber, most students aren’t getting enough of it in their meals and it greatly helps with your gut health
Project Link: dekelgalor.com/hist2style
Co-Authors: Adam Pikielny, Zhoutong Zhang, Ke Wang, Laura Waller, Jiawen Chen
If I know you on a first name basis and you send me an LLM generated email that’s an instant -9999 ilya social credit points
every composer ever: "I have completed my best piece so far, time to name it `waltz. no 6` or something to make sure it's completely ungooglable 180 years from now"
Ilya Chugunov retweeted
🎓ICCP 2026 Summer School - July 11-12 at Princeton!
Extend your ICCP trip with hands-on courses on cryo-EM, adaptive optics, SPAD imaging, and interferometry. Applications reviewed on a rolling basis - apply early!
iccp2026.iccp-conference.org…
#ICCP2026
Ilya Chugunov retweeted
We have an important result to share: if you reduce multiple dense vision tasks into a single RGB-image-prediction task, fine-tuning a strong image generator (in our case Nano Banana Pro) matches or beats all specialized models for monodepth, normals, and semantic segmentation.
Yay, finally! Introducing Vision Banana🍌 from @GoogleDeepMind, our unified model that outperforms SoTA specialist models on various vision tasks!
By treating 2D/3D vision tasks as image generation, we unlock a new foundation for CV.
Project page: vision-banana.github.io
(1/5)
