Sign in

Valentin Deschaintre

@vdeschaintre.bsky.social
1.1K followers 227 following 120 posts

Doing research at Adobe in Computer Graphics/Vision/ML on appearance & content authoring and generation. I also like photography, and baking, but I try to keep it under control! valentin.deschaintre.fr

PostsRepliesMedia
Valentin Deschaintre @vdeschaintre.bsky.social · 24/03/2026
Got a paper around appearance at CVPR 2026? Make the most of it at the AppX workshop! Submit by April 7: appx-workshop.github.io @elegarces.bsky.social @iliyangeo.bsky.social Giseppe Vecchio Michael Fischer #CVPR2026 #ComputerVision #NeuralRendering #AppearanceUnderstanding #AppX2026
appx-workshop.github.io
APPX Workshop @ CVPR 2026
011
Valentin Deschaintre @vdeschaintre.bsky.social · 23/01/2026
Another year, another Thesis Fast Forward, fantastic set of work! Thanks Ruben for keeping it going!
031
Reposted by Valentin Deschaintre
Ruben Wiersma @rubenwiersma.nl · 23/01/2026
The #SIGGRAPH Thesis Fast Forward 2026 is here! Learn about the future of CG in 9 PhD theses: from Monte Carlo PDE solvers and photorealistic 3D avatars to fantasy-based rehab games and modular umbrella meshes. 🎓✨ youtu.be/FlrTZEJeEXs
youtu.be
SIGGRAPH Thesis Fastforward 2026
YouTube video by ACMSIGGRAPH
056
Reposted by Valentin Deschaintre
Annie Gottfried Cohen @annsplaining.bsky.social · 03/12/2025
I asked my husband if there is a French equivalent of "sticks and stones will break my bones" and he gave me "the toad's drool doesn't reach the white dove" and "the train of your insults is rolling on the rails of my indifference" and now I would like both of these on a t-shirt
44813
Reposted by Valentin Deschaintre
Alban Gauthier @albangauthier.bsky.social · 17/11/2025
The implementation of our paper called "An evaluation of SVBRDF Prediction from Generative Image Models for Appearance Modeling of 3D Scenes" is now available at: github.com/graphdeco-in... Project Page: repo-sam.inria.fr/nerphys/svbr...
github.com
GitHub - graphdeco-inria/svbrdf-evaluation: Official implementation of the article "An evaluation of SVBRDF Prediction from Generative Image Models for Appearance Modeling of 3D Scenes"
Official implementation of the article "An evaluation of SVBRDF Prediction from Generative Image Models for Appearance Modeling of 3D Scenes" - graphdeco-inria/svbrdf-evaluation
161
Reposted by Valentin Deschaintre
Biblioteca de la Universidad de Zaragoza @bibliounizar.bsky.social · 03/11/2025
Julia Guerrero Viu @juliagviu.bsky.social, investigadora aragonesa recibe el premio SCIE-Zonta-Sngular 2025, que impulsa la visibilidad de las mujeres en la informática 🤩https://shre.ink/oeUz ¡Enhorabuena! Orgullosos de nuestra #genteunizar 🫶
072
Valentin Deschaintre @vdeschaintre.bsky.social · 30/10/2025
I was at #AdobeMAX this week to present #projectSurfaceSwap! Our surface selection and replacement techs! Check it out here: youtu.be/Xg4n60hYfhA?... What a blast the Sneaks were!
090
Reposted by Valentin Deschaintre
Ruben Wiersma @rubenwiersma.nl · 07/10/2025
The #SIGGRAPH Thesis Fast Forward 2026 submissions are now open! Deadline: Nov 7th 2025. Let the world know what you've been working on in your PhD! More info at research.siggraph.org/thesisff/ and check out last year's iteration youtube.com/watch?v=8esF...
044
Valentin Deschaintre @vdeschaintre.bsky.social · 18/09/2025
Who's Adam?
120
Valentin Deschaintre @vdeschaintre.bsky.social · 19/08/2025
Sneak peak into the live event. That's the most important part of the FF, the other 2.5 hours pale in comparison.
030
Valentin Deschaintre @vdeschaintre.bsky.social · 19/08/2025
Can we meet to have beer/coffee and talk about field trend, open scientific questions, the future of our species (this may get mildly depressing) and the meaning of life instead?
020
Reposted by Valentin Deschaintre
Ruben Wiersma @rubenwiersma.nl · 18/08/2025
If you're confused, I uploaded the presentation as well www.youtube.com/watch?v=jv7o...
youtube.com
Uncertainty for SVBRDF Acquisition using Frequency Analysis - SIGGRAPH 2025
YouTube video by Ruben Wiersma
163
Reposted by Valentin Deschaintre
Ruben Wiersma @rubenwiersma.nl · 18/08/2025
Our fast forward for SIGGRAPH 2025. Keep SIGGRAPH weird! www.youtube.com/watch?v=sYel...
youtube.com
Fast Forward SIGGRAPH 2025 - Uncertainty for SVBRDF Acquisition using Frequency Analysis
YouTube video by Ruben Wiersma
2166
Valentin Deschaintre @vdeschaintre.bsky.social · 13/08/2025
I will be presenting in the 10.30 session in room 208-209. I will share thoughts about implicit/generative and explicit appearance representations. See you there if that sounds interesting!
021
Valentin Deschaintre @vdeschaintre.bsky.social · 09/08/2025
Come check out what the newest generation has been up to!
030
Valentin Deschaintre @vdeschaintre.bsky.social · 08/08/2025
It's in Vancouver ;)
100
Valentin Deschaintre @vdeschaintre.bsky.social · 08/08/2025
🧑‍🎓I have the pleasure to present at the Best of Eurographics session at #SIGGRAPH2025 next week! I’ll be looking back at our work on appearance authoring, editing & understanding — come by and say hi! 🔗 s2025.conference-schedule.org/presentation...
s2025.conference-schedule.org
Presentation - SIGGRAPH 2025 Conference Schedule
171
Reposted by Valentin Deschaintre
Ruben Wiersma @rubenwiersma.nl · 30/07/2025
Mark your calendars for the #SIGGRAPH 2025 Thesis Fast Forward in-person event. Recent PhD graduates will present their thesis work and answer questions on Monday 11 August, 10:30-11:30am: s2025.conference-schedule.org/presentation...
083
Reposted by Valentin Deschaintre
Keenan Crane @keenancrane.bsky.social · 20/06/2025
Folks in the #SIGGRAPH community: You may or may not be aware of the controversy around the next #SIGGRAPHAsia location, summarized here www.cs.toronto.edu/~jacobson/we... If you're concerned consider signing this letter docs.google.com/document/d/1... via this form docs.google.com/forms/d/e/1F...
1267
Reposted by Valentin Deschaintre
Torsten Sattler @sattlertorsten.bsky.social · 12/06/2025
Attending @cvprconference.bsky.social and looking for a PhD or postdoc position in the area of 3d reconstruction (Gaussian splatting, nerfs, scene understanding, etc.)? Find me or drop me an email ;)
0169
Reposted by Valentin Deschaintre
Iliyan Georgiev @iliyangeo.bsky.social · 12/06/2025
Editing appearance, geometry, lighting with precision is easy with a 3D scene representation. But it's so much more difficult with just an image or photograph. Enter IntrinsicEdit: Precise generative image manipulation in intrinsic space (SIGGRAPH 2025)! intrinsic-edit.github.io
intrinsic-edit.github.io
IntrinsicEdit: Precise generative image manipulation in intrinsic space
We present a powerful generative editing workflow built on the RGB-X diffusion framework, enabling pixel-precise, semantic image manipulation across tasks like relighting, color adjustment, and object...
2136
Valentin Deschaintre @vdeschaintre.bsky.social · 12/06/2025
⭐ Image editing without paired data, delivering on the promise of RGB<->X! Edit in Intrinsic space, get back your image with the desired modification! The method is quite fun too, conditional token optimization and noise inversion.
162
Valentin Deschaintre @vdeschaintre.bsky.social · 09/06/2025
I feel like "j'veux dire..." would be a pretty solid equivalent
010
Valentin Deschaintre @vdeschaintre.bsky.social · 06/06/2025
Lots of ad for this in Old Street tube stop. Who is this ad for. Is the goal negative buzz?
100
Valentin Deschaintre @vdeschaintre.bsky.social · 17/05/2025
Thanks! And congrats on the paper award!
110
Valentin Deschaintre @vdeschaintre.bsky.social · 17/05/2025
Thanks for diving in the crazy worlds of materials with me!
010
Valentin Deschaintre @vdeschaintre.bsky.social · 17/05/2025
Thanks Ana, hope to see you in Vancouver?
110
Valentin Deschaintre @vdeschaintre.bsky.social · 17/05/2025
Thanks Vova!
000
Reposted by Valentin Deschaintre
Angela Dai @adai.bsky.social · 16/05/2025
Brilliant insights from @michael-j-black.bsky.social on the importance of data and 3D+ for 4D foundation models that understand humans, and the future of embodied intelligence in the last keynote talk of #Eurographics2025! See you next year in Aachen :)
0191
Reposted by Valentin Deschaintre
Angela Dai @adai.bsky.social · 15/05/2025
Awesome look into the future of humanoid robots and what we can learn from character animation from Karen Liu’s keynote at #Eurographics2025!
071
Reposted by Valentin Deschaintre
Angela Dai @adai.bsky.social · 14/05/2025
Amazing keynote by Alyosha Efros on the role of data in visual computing at #Eurographics2025! Thought-provoking insights from generative models to 3D perception :)
0143
Valentin Deschaintre @vdeschaintre.bsky.social · 13/05/2025
Thanks for the kind word! Maybe we will have the opportunity to collaborate sometimes in the future! 🤞
010
Valentin Deschaintre @vdeschaintre.bsky.social · 13/05/2025
Finally, I cannot thank enough @ygryadit.bsky.social for her everyday support, and all she taught me about academia! To everyone who made, or continues, to make the journey more fun and interesting, thank you!
020
Valentin Deschaintre @vdeschaintre.bsky.social · 13/05/2025
I was fortunate to collaborate with and learn from incredible senior researchers but also students. Over the past 4 years at Adobe I had the chance to work with outstanding students who trusted me to come explore ideas as my interns or collaborators!
100
Valentin Deschaintre @vdeschaintre.bsky.social · 13/05/2025
I was very honoured to receive one of the two Eurographics Young Researcher Award 2025 yesterday! This is the combination of the work of many people, mentors, collaborators, students, friends who trusted me and taught me so much along the way!
9374
Valentin Deschaintre @vdeschaintre.bsky.social · 06/05/2025
Les libertés prisent par rapport à l'histoire originale. Je ne me souviens plus du détail ça fait un moment, mais de mémoire il y avait plusieurs modifications qui n'aidaient pas le rythme et avaient des implications importantes pour l'intégralité de l'histoire.
000
Valentin Deschaintre @vdeschaintre.bsky.social · 05/05/2025
J'ai tenu trois épisodes 🙈, pourtant j'adore les livres.
100
Reposted by Valentin Deschaintre
Ruben Wiersma @rubenwiersma.nl · 30/04/2025
Recent progress in multi-view 3D object capture is stunning! But is there a limit? In our upcoming #SIGGRAPH2025 paper "Uncertainty for SVBRDF Acquisition using Frequency Analysis," we show how to answer this question for material estimation. 1/6
The figure illustrates a pipeline for estimating uncertainty and spatially-varying material properties of a plush dinosaur toy. On the left, a rendered model shows the input: incoming lighting, the BRDF, and outgoing lighting from multi-view captures. Three plots in the center display spherical harmonics power spectra for incoming lighting, BRDF, and outgoing lighting. Right side: first, a colored uncertainty map (entropy) over the dinosaur surface is shown, computed within 0.5 milliseconds. Then, three images show estimated base color, roughness, and metallicity maps, computed in about 3 seconds.
1236
Valentin Deschaintre @vdeschaintre.bsky.social · 30/04/2025
This was a blast to work on, understanding the signal available to estimate the uncertainty in your material recovery! And it's fast enough to inform your future capture steps in real time!
142
Reposted by Valentin Deschaintre
Nicolas Holzschuch @nholzschuch.bsky.social · 29/04/2025
Here is the dataset exploration tool provided with the paper: citiesmoving.com/visualizatio...
citiesmoving.com
Visualizations - Cities Moving
132
Reposted by Valentin Deschaintre
Angela Dai @adai.bsky.social · 28/04/2025
📢𝐄𝐮𝐫𝐨𝐠𝐫𝐚𝐩𝐡𝐢𝐜𝐬 𝟐𝟎𝟐𝟓 is almost around the corner - immerse yourself in computer graphics innovation in London, May 12-16! Register now: eg25.cs.ucl.ac.uk/main/registr... See you in London! 👋
062
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
At SIGGRAPH @elie-michel.bsky.social et al. had a nice workshop discussing energy consumption, not sure if that contains what you are looking for, but Élie will know best!
160
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
Rewriting so it's understandable by others takes time, I don't know a good way round it :/. But I also like this process, it forces me to rethink whether what I suggest/ask is fair, in the process of explaining what I mean I often understand why something was done the way it was.
120
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
To be complete, I should probably add that adding many small things that are not so important eat a lot of time because they also need to be well justified and perhaps require thinking about a way to improve them when it would not necessarily improve the paper substantially.
110
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
I don't know, I am too worried LLM will either editorialise my review or interpret it wrong, giving wrong gradient to the authors (and we don't like bad gradient)
010
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
Honestly I think it keeps things focused. It's hard to state a degree of importance, and I have seen many rebuttals where authors avoid/miss the really important question by answering the less important ones. So if it's minor I simply let it go, it will likely not be important in the grand scheme
110
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
In recent years I also try to be more concise (without being short) by focusing on the things I believe are key limitations to address, reviewer have limited space for rebuttal, I want to make sure I don't drown them in thoughts.
120
Valentin Deschaintre @vdeschaintre.bsky.social · 23/04/2025
For what it's worth, borderline reviews are what eats most of my time too. I find that adding sticky notes to the pdf as I go along to mark my reflexions/comments/questions helps a lot with summarizing my thoughts in the end.
250
Reposted by Valentin Deschaintre
Le Chien Maigre @lechienmaigre.bsky.social · 04/04/2025
- T’as vu l’histoire des taxes la?! - Mdr ouais - Ca va nous couter un bras! - Ahahaha t’es con Michel
121257
Reposted by Valentin Deschaintre
Antoine Guédon @antoine-guedon.bsky.social · 03/04/2025
💻We've released the code for our #CVPR2025 paper MAtCha! 🍵MAtCha reconstructs sharp, accurate and scalable meshes of both foreground AND background from just a few unposed images (eg 3 to 10 images)... ...While also working with dense-view datasets (hundreds of images)!
43916