About me
I'm a hands-on research leader grounded in 10+ years of 3D graphics research across academia and industry. I currently lead multimodal generative AI research at Tencent Video AI Center in New York and manage a 15-person team.
My work spans video generation, spatial intelligence, and 3D graphics, with an emphasis on research leadership and research-to-product execution. I also co-founded an AI startup, taking it from the original idea through product design and technical development to seed funding.
I received my PhD from IST Austria (now ISTA). See my recent publications and projects.
News
Make-It-Poseable will appear at SIGGRAPH Asia 2026.
Let There Be Light won the Best of Show Award at SIGGRAPH Asia 2025 Real-Time Live!
Experience
Lead Research Scientist & Manager
New York, USA
Manager since late 2024; lead a 15-person research team and own research strategy, roadmaps, hiring, performance reviews, and technical mentoring.
Lead controllable video generation and spatial VLM research, focusing on multimodal post-training and in-context learning within a hybrid VLM-DiT framework while prototyping native spatial representations.
Led generative video super-resolution and SDR-to-HDR from research through production. Architected a video-agent research prototype for workflow decomposition, simulated trajectory synthesis, retrieval, and closed-loop agent learning.
Senior Graphics Researcher
New York, USA
Postdoctoral Researcher
The Human Computer Interaction Lab
Berlin, Germany
Co-founder, Product & Technical Lead
Shenzhen, China
Conceived and co-founded an AI-assisted personalized brick-toy startup, defining the product vision and leading product design and technical development from idea through seed funding.
The company later grew to 1M+ cumulative users and exceeded US$15M in annual revenue.
Postdoctoral Researcher
Bickel's Group
Vienna, Austria
Research Assistant (PhD student)
Bickel's Group
Vienna, Austria
Research Intern
Internet Graphics Group
Beijing, China
Visiting Student
Smart Geometry Processing Group
London, UK
Visiting Student
The Digital Reality Group
London, UK
Research Assistant (Master student)
Dept. Electronic Engineering and Information Science
Hefei, China
Research Intern
Internet Graphics Group
Beijing, China
Projects
Academic blog: GraphiCon
State of the art technologies of Computer graphics. Interesting graphics papers.
GraphiCon
- Since:
- 12 Mar 2017
- Site link:
- https://zhuanlan.zhihu.com/graphicon
GraphiCon is an UGC media focus on computer graphics, it is founded by several graphics researchers. Currently it is hosted on Zhihu and Wechat official account platform.
GAMES Webinar
Weekly computer graphics webinar
GAMES Webinar
- Since:
- 22 Jun 2017
- Site link:
- www.games-cn.org
GAMES stands for Graphics And Mixed Environment Seminar, aims at spreading computer graphics techniques and assisting young graphics researchers.
Computational design of functionalities
Algorithm assisted design for 3d printable objects with functionalities.
Computational design of functionalities
- Date:
- Jul 2017
- Site link:
- http://visualcomputing.ist.ac.at/publications/2017/MechRet/
In our SIGGRAPH 2017 paper: Functionality-aware Retargeting of Mechanisms to 3D Shapes, We present an interactive design system to create functional mechanical objects. Our computational approach allows novice users to retarget an existing mechanical template to a user-specified input shape.
Deployable structures
Structures that can be compressed and deployed into certain shapes.
Deployable structures
- Date:
- Jan 2016
- Site link:
- http://staff.ustc.edu.cn/~xjchen99/scissor.html
Scissor structures are used to generate deployable objects for space-saving in a variety of applications, from architecture to aerospace science. While deployment from a small, regular shape to a larger one is easy to design, we focus on a more challenging task: designing a planar scissor structure that deploys from a given source shape into a specific target shape. We propose a two-step constructive method to generate a scissor structure from a high-dimensional parameter space. Topology construction of the scissor structure is first performed to approximate the two given shapes, as well as to guarantee the deployment. Then the geometry of the scissor structure is optimized in order to minimize the connection deflections and maximize the shape approximation. With the optimized parameters, the deployment can be simulated by controlling an anchor scissor unit. Physical deployable objects are fabricated according to the designed scissor structures by using 3D printing or manual assembly.
BrickMe Studio
BrickMe Studio is the world's first AI-assisted system for generating brick assemblies from a single portrait.
BrickMe Studio
- Since:
- June 2020
- Site link:
- https://brickme.deecamp.com
BrickMe Studio is the world's first AI-assisted system for generating brick assemblies from a single portrait.
