Hyunsoo Lee

Undergrad @ SNU ECE & Math | Exchange Student @ UC Irvine

[Soulart @SoulartOfficial]

We are announcing the Early Access Program (EAP) of Soulart Station.

Register now to be the first to receive and experience it.
https://t.co/TzSqhPSu2v

[Hyunsoo Lee @HysooLee]

Huge thanks to collaborators: @changwoon_choi , @clementjambon , @YVinker , and Prof. Young Min Kim for constructive feedback throughout this project.

[8/n]

[Hyunsoo Lee @HysooLee]

What if "style" in 3D was not just painted on the surface, but carved into the shape itself?
Introducing Geometric Stylization: Redefining style as geometry.

Image-Guided Geometric Stylization of 3D Meshes, Accepted to CVPR 2026.
Will be presented at Saturday, June 6, 11:45 AM – 1:45 PM MDT, ExHall F.

🌐 Project page: https://t.co/IrxdDeWw4p
📑 Paper: https://t.co/QbuNMBPKJh
💻 Code: https://t.co/AnDZ6EYFKH

[1/n]

[Hyunsoo Lee @HysooLee]

We further show that our framework can be flexibly extended with additional controls: text and localized deformation signals.

[7/n]

[InSpatio｜影溯 @InSpatio_AI]

We don’t generate videos. 🎬
We generate worlds from videos. 🌍

Introducing InSpatio-World — the world's first open-source real-time 4D world model‼️

Your input: a video clip
Our output: a dynamic, navigable, persistent world

🕹️ explore freely across viewpoints
⏪ control time forward and backward
🔓 open-source and ready to build on :)

Live demo:
🔗 https://t.co/nqU4ZZSqOH

Code & weights:
🔗 https://t.co/rNMilkvKUV

Project page:
🔗 https://t.co/WfMjYGbfzq

[Sookwan Han @SookwanH]

We can synchronize multiple diffusion models for one collaborative generation. But WHY does it work? HOW should we synchronize?

**SyncSDE** explains
- WHY it works: probabilistic framework for synchronizing diffusion models
- WHERE to focus: where synchronization heuristics should be strategically focused, to reduce trial-and-error
- Strong results on diverse tasks

Accepted in CVPR 2025!!
Arxiv: https://t.co/tQi5bLT37G
Website: https://t.co/GgE6sovWzA

[1/n]