Nan Liu

I research multimodal and generative models. I have worked at World Labs, Google, and Adobe Research. I received my bachelor's from the University of Michigan and my master's from the University of Illinois.

I also invest in early-stage startups—feel free to reach out if you're building something interesting.

Email · CV · Scholar · GitHub · X · LinkedIn

Products

Multimodal World Model World Labs

Interactive 3D world generation from multimodal inputs
World Generation World Labs

Generating explorable 3D worlds from a single image
Virtual Try-On Google

Video diffusion model for virtual clothing try-on, launched on Google Shopping

Selected Publications

Fashion-VDM: Video Diffusion Model for Virtual Try-On

Johanna Karras, Yingwei Li, Nan Liu, Luyang Zhu, Innfarn Yoo, Andreas Lugmayr, Chris Lee, Ira Kemelmacher-Shlizerman

SIGGRAPH Asia 2024 · Website · Paper

Fashion-VDM teaser

Compositional Image Decomposition with Diffusion Models

Jocelin Su*, Nan Liu*, Yanbo Wang*, Joshua B. Tenenbaum, Yilun Du

ICML 2024 · Website · Paper · Code

M&M VTO: Multi-Garment Virtual Try-On and Editing

Luyang Zhu, Yingwei Li, Nan Liu, Hao Peng, Dawei Yang, Ira Kemelmacher-Shlizerman

CVPR 2024 · ★ Highlight · Website · Paper

M&M VTO teaser

Unsupervised Compositional Concepts Discovery with Text-to-Image Generative Models

Nan Liu*, Yilun Du*, Shuang Li*, Joshua B. Tenenbaum, Antonio Torralba

ICCV 2023 · Website · Paper · Code

Compositional Visual Generation with Composable Diffusion Models

Nan Liu*, Shuang Li*, Yilun Du*, Antonio Torralba, Joshua B. Tenenbaum

ECCV 2022 · Website · Paper · Code · Demo · MIT News

Learning to Compose Visual Relations

Nan Liu*, Shuang Li*, Yilun Du*, Joshua B. Tenenbaum, Antonio Torralba

NeurIPS 2021 · ★ Spotlight · ★ Outstanding Paper · Website · Paper · Code · MIT News