Computer Vision Research Scientist at @simulon, music lover , fond of scientific/musical/geeky/useless stuff. I'm posting papers on whatever I found amazing :)

Paris, France
Joined June 2012
Oh by the way, I have embarked on a new adventure as a Research Scientist in Computer vision at @Simulon 🎉. I couldn't be more excited about the opportunities and challenges that lie ahead in making VFX available for everyone and empowering creators. Stay tuned!
3
6
52
0
Jiaming Zhang 1,‡ Cheng Liang 1,‡ Yichun Yang 1,‡ Chenkai Zeng 1,‡ Yutao Cui 1 Xinwen Zhang 1 Xin Zhou 1 Kai Ma 2 Gangshan Wu 1 Limin Wang 1,3,† 1 @NJU1902 2 Platform and Content Group (PCG), @TencentGlobal 3 OpenGVLab,Shanghai AI Laboratory
"SAM 2++: Tracking Anything at Any Granularity" TL;DR: unifies video tracking across masks, boxes & points; uses task-specific prompts, a unified decoder, and a task-adaptive memory to track at any granularity. Backed by a new large-scale dataset.
1
2
0
Alexandre Morgand retweeted
Check out Viu Bannes's Halloween gift: a 3D illusion of the Kingdom Hearts: Chain of Memories gameplay inside a Game Boy Advance cartridge, made with Blender. Enjoy: 80.lv/articles/magical-3d-il…
Instant Skinned Gaussian Avatars for Web, Mobile and VR Applications TL;DR: animates GS by leveraging parallel splat-wise processing to dynamically follow the underlying skinned mesh in real time while preserving high visual fidelity.
3
5
0
Alexandre Morgand retweeted
Imagine losing first authorship because you got hit by a blue shell on the last lap 💀
LLMs are injective and invertible. In our new paper, we show that different prompts always map to different embeddings, and this property can be used to recover input tokens from individual embeddings in latent space. (1/6)
Hanwen Jiang 1, Hao Tan 2, Peng Wang 2, Haian Jin 3, Yue Zhao 1, Sai Bi 2, Kai Zhang 2, Fujun Luan 2, Kalyan Sunkavalli 2, Huang Qixing 1, Georgios Pavlakos 1 1 @UTAustin, 2 @Adobe Research, 3 @Cornell University
RayZer: A Self-supervised Large View Synthesis Model TL;DR: Modern self-supervised methods share the same principle of learning by predicting "missing" data. GPT predicts missing next token leveraging language sequential prior. Can we do the same for self-supervised 3D model?
1
1
0
No paper in this post, but I'm back! Took two weeks off in the beautiful South Korea. Be ready, I've a full list of exciting research papers and also future news about our recent app release at @Simulon !
6
Alexandre Morgand retweeted
When I started working at Pixar in Sept 2000, I only animated with stepped keyframes, breakdowns and liner inbetweening because that is how I learned to animated at Disney. Pixar is the KING of spline animation and this bothered a lot people (I was told as much.) Eventually, me and a few liked minded animators convinced the tools team to write x-sheet, a tool that let us record our held keyframes in our shot recordings and then move them for timing and push them back into the spline editor. It was incredibly efficient and fast. THE INCREDIBLES was the first film we used this widely and I think you can see the results. Eventually, many of us landed on a workflow that was held keys/breakdowns for blocking and spline inbetweening for polish.
Rough blocking for my rooftop shots on #TheIncredibles.
28
676
6
10,858
Alexandre Morgand retweeted
Simulon has just opened up to the public! The app allows you to shoot, track, HDRI light, and render in your iPhone using its LiDAR 🤯 So you can take 3D models, composite them into your scene, and it lights everything really nicely using the environment you're shooting in. Find out more on Motion Mondays ➡️ som.bz/3WqtnOF
6
20
0
Alexandre Morgand retweeted
so good i had to animate it with grok!
Slot-loading
Alexandre Morgand retweeted
Tanat shared a VFX breakdown of the water shader from their indie game, built in Unity with toon shading, refraction ripples, and Voronoi noise for variation. 🌊 See more: teddit.net/r/Unity3D/comment… #unity3d #realtimeVFX #gamedev
1
101
1
1,393
0
Alexandre Morgand retweeted
There's a lot of both 2D and 3D effort that we're putting into Star Birds together with our friends at Toukana Interactive! Here's a quick behind the scenes look at some of the assets, 3D models, birb shenanigans and cutscenes that you might have already seen in the Early Access version! 🚀✨ Have you already joined us on this space adventure? Don't miss out! Star Birds in Early Access is available on Steam: kgs.link/4n1aYnq
Alexandre Morgand retweeted
I think we just built the most insane 3D VFX workflow. Precision reality editing with @Simulon + Blender. • Scan the real world in seconds • Edit in Blender • Sync + reopen in Simulon • View & shoot in real-time AR • Photoreal cloud render in minutes Will share a BTS of this clip shortly.
Lighting is key to creating believable VFX. @Simulon generates accurate 360 hdr lighting in seconds. But there’s something much stranger going on in this video. The chrome ball is being moved through the world as if it is actually there. How do you think this is possible? 👀 We’ve built something that will completely change how 3D visual effects are made and it’s coming to Simulon next week.
Alexandre Morgand retweeted
This week AI tackles retopology, we look at a mind-blowing VFX workflow on your iPhone, AND we launch a new #AfterEffects course you WILL NOT want to miss!💯 It's Motion Mondays! 🦛 Watch the latest episode now ➡️ som.bz/3WpRZXO
6
15
0
Alexandre Morgand retweeted
So cool seeing Simulon featured on CGWorld Japan 🙏 We’ve always been huge fans of Japanese creators. This is a fan-made render I created of @kowyokoyama’s Ma.K. mecha - shot when the JMSDF visited Cape Town. The crew lost their minds when I showed them. Shooting VFX on location and getting a photo-real render back in minutes really does feel like magic.
╭━━━━━━━╮ iPhone用VFXツール「Simulon」! ╰━━v━━━━╯ iPhoneで撮影した実写映像にフォトリアルな3DCGをシームレスに合成するインカメラVFXを実現 🔽詳細 cgworld.jp/flashnews/01-2025…