Taesung Park

I am currently working at a startup as a co-founder. Hope to share more details soon!

Previously, I was a Research Scientist at Adobe Research, focusing on image editing using generative models. I received Ph.D. in Computer Science at UC Berkeley, advised by Prof. Alexei Efros. Previously I interned at Adobe in 2019, working with Richard Zhang, and at NVIDIA, working with Ming-Yu Liu in summer 2018. I received B.S. in Mathematics and M.S. in Computer Science, both at Stanford University. During my Master’s program, I was advised by Vladlen Koltun and Sergey Levine. I was funded by Samsung Scholarship for my Ph.D. study, and a recipient of Adobe Research Fellowship 2020.

Email  /  Google Scholar

profile photo
Software

Highlighted softwares developed from my research papers.

Firefly and Generative Fill is a text-to-image generative model that is trained on ethically sourced data, and is significantly faster than competitors in generating high-resolution images. I developed the upscaling pipeline of Firefly.

Photoshop Landscape Mixer transforms landscape images in various ways.
Contributes to Adobe’s nomination as World’s 4th Most Innovative AI company of 2022.
Based on Swapping Autoencoder.

GauGAN turns sketches into photos.
100 Greatest Innovations of 2019 by Popular Science.
Based on Semantic Image Synthesis with Spatially-Adaptive Normalization.


Research

I am mainly interested in image editing and image synthesis using machine learning.

One-step Diffusion with Distribution Matching Distillation
Tianwei Yin, Michaël Gharbi, Richard Zhang, Eli Shechtman,
Frédo Durand, Bill Freeman, Taesung Park
CVPR, 2024
arXiv / Project

Holistic Evaluation of Text-To-Image Models
Tony Lee*, Michihiro Yasunaga*, Chenlin Meng*, ... Taesung Park, ... Percy Liang,
NeurIPS, 2023
arXiv / Project

Expressive Text-to-Image Generation with Rich Text
Songwei Ge, Taesung Park, Jun-Yan Zhu, Jia-Bin Huang,
ICCV, 2023
arXiv / Project

Scaling up GANs for Text-to-Image Synthesis
Minguk Kang, Jun-Yan Zhu, Richard Zhang, Jaesik Park,
Eli Shechtman, Sylvain Paris, Taesung Park
CVPR, 2023 (Highlight)
arXiv / Project

Domain Expansion of Image Generators
Yotam Nitzan, Michaël Gharbi, Richard Zhang, Taesung Park,
Jun-Yan Zhu, Daniel Cohen-Or, Eli Shechtman
CVPR, 2023
arXiv / Project

BlobGAN: Spatially Disentangled Scene Representations
Dave Epstein, Taesung Park, Richard Zhang, Eli Shechtman, Alexei Efros
ECCV, 2022
arXiv / Project / Talk / Code / Demo

ASSET: Autoregressive Semantic Scene Editing with Transformers at High Resolutions
Difan Liu, Sandesh Shetty, Tobias Hinz, Matthew Fisher, Richard Zhang,
Taesung Park, Evangelos Kalogerakis
SIGGRAPH - Journal Track, 2022
PDF(low-res) / PDF(high-res) / Project

Contrastive Feature Loss for Image Prediction
Alex Andonian, Taesung Park, Bryan Russell, Phillip Isola, Jun-Yan Zhu, Richard Zhang
ICCVW, 2021
Paper

Swapping Autoencoder for Deep Image Manipulation
Taesung Park, Jun-Yan Zhu, Oliver Wang, Jingwan Lu, Eli Shechtman, Alexei Efros, Richard Zhang
NeurIPS, 2020
arXiv / Project

Contrastive Learning for Unpaired Image-to-Image Translation
Taesung Park, Alexei Efros, Richard Zhang Jun-Yan Zhu
ECCV, 2020
arXiv / Project / Code

Semantic Image Synthesis with Spatially-Adaptive Normalization
Taesung Park, Ming-Yu Liu, Ting-Chun Wang, Jun-Yan Zhu
CVPR, 2019. Best Paper Finalist. SIGGRAPH RTL Best of Show award
arXiv / Project / Code

CyCADA: Cycle-Consistent Adversarial Domain Adaptation
Judy Hoffman, Eric Tzeng, Taesung Park, Jun-Yan Zhu, Phillip Isola, Kate Saenko, Alexei Efros, Trevor Darrell
ICML, 2018
arXiv / Code

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
Jun-Yan Zhu*, Taesung Park*, Phillip Isola, Alexei Efros
ICCV, 2017 (Spotlight, * indicates equal contribution)
arXiv / Project / Code

Inverse Optimal Control for Humanoid Locomotion
Taesung Park, Sergey Levine
RSS Workshop, 2013
Paper


Dissertation
Machine Learning for Deep Image Synthesis
Taesung Park
EECS Department, UC Berkeley, 2021
Paper

webpage template from Jon Barron