Ilya Chugunov

171 posts

Ilya Chugunov banner
Ilya Chugunov

Ilya Chugunov

@_ilya_c

Adobe Research Scientist doing computational photography stuff, previously Princeton PhD. I post more on bsky. All opinions are my own.

Katılım Nisan 2022
116 Takip Edilen259 Takipçiler
Sabitlenmiş Tweet
Ilya Chugunov
Ilya Chugunov@_ilya_c·
(1/2) Tired of your panoramas being turned into sad flat rectangles? Our latest #SIGGRAPHAsia2024 work "Neural Light Spheres" lets you turn panoramic captures into dynamic wide FOV renders (with real-time rendering!). Code, data, and info: light.princeton.edu/publication/ne…
English
1
5
62
5.3K
Ilya Chugunov
Ilya Chugunov@_ilya_c·
no animal sold in a pet store should cost less than trader joe’s chicken by weight
English
0
0
1
56
Ilya Chugunov retweetledi
#ICCP2026
#ICCP2026@ICCP_conference·
ICCP 2026 is coming to @Princeton, July 13-15! Paper submissions are open, deadline April 10. Accepted papers published in ICCP Proceedings or IEEE PAMI Special Issue. Take a sneak peek at already confirmed speakers on our website: iccp2026.iccp-conference.org! #ICCP2026
#ICCP2026 tweet media
English
0
5
13
924
Ilya Chugunov
Ilya Chugunov@_ilya_c·
@jon_barron Japan drives me insane with this, depending on what city you're in the walking-standing sides will flip
English
0
0
2
100
Jon Barron
Jon Barron@jon_barron·
I'm visiting the UK and I couldn't shake the feeling that walking around feels a little uncanny compared to the rest of the anglosphere. Gemini figured it out --- the UK is the only anglo country where driving/walking/escalator orientations aren't consistent.
Jon Barron tweet media
English
4
3
30
4.5K
Ilya Chugunov
Ilya Chugunov@_ilya_c·
I want to start my own conference where the papers don't have to have any evaluations. No baselines, no tables, no PSNRs, no FID scores or IoU metrics. The only review score is "are the methods presented in this paper actually interesting (1-5)"
English
4
1
7
860
David Lindell
David Lindell@DaveLindell·
So cool to see our work featured by @veritasium! Our results on rendering novel views of propagating light are featured at the beginning, and check out @anagh_malik's interview at 17:30. youtube.com/watch?v=P-4pbF… anaghmalik.com/FlyingWithPhot…
YouTube video
YouTube
Anagh Malik@anagh_malik

📢📢📢 I’ve been watching @veritasium since high school, so it’s surreal to see our work Flying with Photons featured in their video! FWP allows "flying" in the scene while seeing light propagate.

English
4
0
32
3.2K
Ilya Chugunov
Ilya Chugunov@_ilya_c·
Name a better combo: speaker saying “without further ado” + there being further ado
English
0
0
0
109
Ilya Chugunov retweetledi
Jingwei Ma
Jingwei Ma@JingweiMa2·
Excited to present UltraZoom at SIGGRAPH Asia next Tuesday (Dec.16)! UltraZoom converts sparse phone captures of an object into a single gigapixel-resolution image that you can seamlessly explore. Threads below. Website: ultra-zoom.github.io Paper: arxiv.org/abs/2506.13756
English
2
3
12
994
Ilya Chugunov
Ilya Chugunov@_ilya_c·
@ana_dodik isn’t twitter ragebaiting is statistically the number one way to get tenure? /s
English
1
0
1
281
Ana Dodik
Ana Dodik@ana_dodik·
Tired of senior folks dragging people's posters on socials. Has contentification cooked our sociality so much? Consider: 1. In-person feedback. 2. Broad feedback. 3. Make resources. 4. Post positive examples. 5. Censor the name. 6. Laugh about it in a groupchat if you must.
English
4
1
41
4.3K
Ilya Chugunov retweetledi
Xindi Wu
Xindi Wu@cindy_x_wu·
I’m at #NeurIPS2025 from 12.2–12.7! I work on data-centric Video Generation and VLMs/VLAs recently (MOTIVE, COMPACT, ICONS, etc.), and I’m generally interested in building more scalable and capable multimodal systems. DMs open for a coffee chat! 😃 Excited to meet old and new friends!
English
2
4
70
6.3K
Jon Barron
Jon Barron@jon_barron·
A traditional AI generated stop motion knolling. Workflow: I photographed my meal, asked Gemini for a top down orthographic view, then asked it for a knolling, then used Veo/Flow for a stop motion animation using the two generations as first and last frames.
English
9
5
108
10.2K
Ilya Chugunov
Ilya Chugunov@_ilya_c·
Sometimes I wish I could go back in time and edit my papers to have the correct buzzwords. "light fields for image stitching and panoramic reconstruction" ? Nah baby that's World Models (tm)
English
0
0
3
109
Ilya Chugunov retweetledi
Tyler Zhu
Tyler Zhu@tyleryzhu·
Today seems to be a fitting day for @GoogleDeepMind news, so I'm excited to announce our new preprint! Prior work suggests that text & img repr's are converging, albeit weakly. We found these same models actually have strong alignment; the inputs were too impoverished to see it!
Tyler Zhu tweet media
English
11
25
138
30.5K
Ilya Chugunov retweetledi
Victor Chu
Victor Chu@bkv2chu·
Imagine AR as immersive as Vision Pro and as light as Meta Ray Bans. That's the promise of holographic displays. The flaw? Eyelashes can cast image-destroying shadows. In our #SIGGRAPHAsia2025 paper, we train an AI to generate "eyelash-proof" holograms at 80 FPS to fix this.
English
5
16
63
7.6K
Ilya Chugunov
Ilya Chugunov@_ilya_c·
An absolute must-see at #NeurIPS this year for any neural spline field fans who were like "yeah neural spline fields are cool but I wish they were even more neurally and spliney"
Felix Heide@_FelixHeide_

Excited to present editable Neural Atlas Graphs at #NeurIPS 2025 (Spotlight)! We introduce a learned atlas representation in which each dynamic object is a 2D planar layer (the atlas). All time-dependent appearance and fine motion are captured directly within this 2D layer using a learned planar flow field and a view-dependent field. Neural Atlas Graphs allow for texture-editable neural representations at high resolution! Project page: …ceton-computational-imaging.github.io/nag/ Amazing project led by @jaypschneider, with Pratik Bisht, @_ilya_c, Andreas Kolb, Michael Moeller, with the Universität Siegen, @Princeton, the Lamarr Institute, and @torc_robotics.

English
0
0
2
344
Ilya Chugunov retweetledi
Felix Heide
Felix Heide@_FelixHeide_·
Excited to present editable Neural Atlas Graphs at #NeurIPS 2025 (Spotlight)! We introduce a learned atlas representation in which each dynamic object is a 2D planar layer (the atlas). All time-dependent appearance and fine motion are captured directly within this 2D layer using a learned planar flow field and a view-dependent field. Neural Atlas Graphs allow for texture-editable neural representations at high resolution! Project page: …ceton-computational-imaging.github.io/nag/ Amazing project led by @jaypschneider, with Pratik Bisht, @_ilya_c, Andreas Kolb, Michael Moeller, with the Universität Siegen, @Princeton, the Lamarr Institute, and @torc_robotics.
English
1
14
110
8.5K
Ilya Chugunov
Ilya Chugunov@_ilya_c·
“reviewing is broken” “publishing is broken” “conferences are broken” I say we scrap the whole system and post research findings directly to tumblr dot com
English
0
0
3
116

Keşfet

@andrewyguo @Princeton @jon_barron @DaveLindell @veritasium @anagh_malik @ana_dodik @elonmusk