Long Zhao @ Google

Brief Bio. I am a Research Scientist at Google Research. Before joining Google, I obtained my Ph.D. in Computer Science from Rutgers University in 2022, advised by Distinguished Professor Dimitris N. Metaxas.

My current research interests lie primarily in (1) unsupervised/self-supervised representation learning, (2) human-centric multimodal learning with vision-language models, and (3) image/video generation by generative adversarial networks, with their applications to modeling human actions/activities.

340 Main Street
Los Angeles, CA 90291
Google LAX

Email: longzh [at] google [dot] com
[Curriculum Vitae]

Background

Education

Sept. 2016 - Jan. 2022

Rutgers, The State University of New Jersey - New Brunswick. Piscataway, NJ

Ph.D. in Computer Science

Supervised by Prof. Dimitris N. Metaxas
Ph.D. Thesis: "Towards Efficient and Reliable Skeleton-Based Human Pose Modeling"
GPA: 4.0/4.0

Sept. 2012 - Jun. 2015

Tongji University. Shanghai, China

M.S. in Software Engineering

Supervised by Prof. Jinyuan Jia and Prof. Shuang Liang
Master Thesis: "Sketch-Based 3D Model Retrieval"

Sept. 2008 - Jul. 2012

Tongji University. Shanghai, China

B.Eng. in Software Engineering

GPA: 4.56/5.0, which is equivalent to 90.6 on 100 basis

Experience

Nov. 2021 - Present

Perception Team, Google Research. Los Angeles, CA, USA

Research Scientist.

View-Invariant Pose Embedding
Human-Centric Multimodal Learning with Vision-Language Models

Dec. 2020 - May 2021

Brain Team, Google Research. Mountain View, CA, USA

Student Researcher. Host: Dr. Han Zhang

Boosting Transformers for High-Resolution Image Generation [NeurIPS'21]
Improving Efficiency and Interpretability for Vision Transformers [AAAI'22]

May 2020 - Dec. 2020

Mobile Vision Team, Google Research. Los Angeles, CA, USA

Research Intern & Student Researcher. Host: Dr. Ting Liu

View-Disentangled Human Pose Representation Learning [CVPR'21 (Oral)]
View-Invariant, Occlusion-Robust Probabilistic Pose Embedding [IJCV'21]

Sept. 2016 - Present

Computer Science Department, Rutgers University. Piscataway, NJ, USA

Research Assistant. Supervised by Prof. Dimitris N. Metaxas

3D Human/Hand Pose Estimation from RGB Images [CVPR'19] [CVPR'20]
Face/Pose/Video Generation with Generative Adversarial Networks [IJCAI'18] [ECCV'18] [IJCV'20]
Domain Generalization via Adversarial Training & Meta-Learning [CVPR'20] [NeurIPS'20]
Representation Learning on Graphs with Graph Convolutional Networks [NeurIPS'19]

Sept. 2016 - Jun. 2018

Computer Science Department, Rutgers University. Piscataway, NJ, USA

Teaching Assistant.

Principles of Information and Data Management (198:336) - Fall 2016
Principles of Programming Languages (198:314) - Spring 2017
Distributed Systems (198:417) - Fall 2017
Internet Technology (198:352) - Spring 2018

Dec. 2013 - Nov. 2014

Visual Computing Group, Microsoft Research Asia (MSRA). Beijing, China

Research Intern. Mentor: Dr. Yichen Wei

Generic Object Proposal Generation [CVPR'15]
Salient Object Detection [ACCV'14]
Won the award of excellence in "MSRA Stars of Tomorrow Internship Program"

Sept. 2012 - Jun. 2015

School of Software Engineering, Tongji University. Shanghai, China

Research Assistant. Supervised by Prof. Jinyuan Jia and Prof. Shuang Liang

Sketch-based Image & 3D Model Retrieval [PCM'14]
Best View Selection of 3D Models from Sketches [CGI'15]

Jul. 2011 - Aug. 2012

Performance Team, SAP SuccessFactors. Shanghai, China

Java Developer Intern.

Worked on developing internal tools using Hadoop & HBase

Selected Publications

(^* indicates equal contributions)

Exploiting Unlabeled Data with Vision and Language Models for Object Detection

Shiyu Zhao, Zhixing Zhang, Samuel Schulter, Long Zhao, Vijay Kumar B.G, Anastasis Stathopoulos, Manmohan Chandraker, Dimitris Metaxas.

European Conference on Computer Vision (ECCV), 2022.

Hierarchically Self-Supervised Transformer for Human Skeleton Representation Learning

Yuxiao Chen, Long Zhao, Jianbo Yuan, Yu Tian, Zhaoyang Xia, Shijie Geng, Ligong Han, Dimitris N. Metaxas.

European Conference on Computer Vision (ECCV), 2022.

COMPOSER: Compositional Reasoning of Group Activity in Videos with Keypoint-Only Modality

Honglu Zhou, Asim Kadav, Aviv Shamsian, Shijie Geng, Farley Lai, Long Zhao, Ting Liu, Mubbasir Kapadia, Hans Peter Graf.

European Conference on Computer Vision (ECCV), 2022.

Out-of-Domain Generalization from a Single Source: An Uncertainty Quantification Approach

Xi Peng, Fengchun Qiao, Long Zhao.

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2022.

[arXiv]

Global Matching with Overlapping Attention for Optical Flow Estimation

Shiyu Zhao, Long Zhao, Zhixing Zhang, Enyu Zhou, Dimitris Metaxas.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.

Are Multimodal Transformers Robust to Missing Modality?

Mengmeng Ma, Jian Ren, Long Zhao, Davide Testuggine, Xi Peng.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.

[arXiv]

Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding

Zizhao Zhang, Han Zhang, Long Zhao, Ting Chen, Sercan O. Arik, Tomas Pfister.

AAAI Conference on Artificial Intelligence (AAAI), 2022.

View-Invariant, Occlusion-Robust Probabilistic Embedding for Human Pose

Ting Liu^*, Jennifer J. Sun^*, Long Zhao, Jiaping Zhao, Liangzhe Yuan, Yuxiao Wang, Liang-Chieh Chen, Florian Schroff, Hartwig Adam.

International Journal of Computer Vision (IJCV), 2022.

Improved Transformer for High-Resolution GANs

Long Zhao, Zizhao Zhang, Ting Chen, Dimitris N. Metaxas, Han Zhang.

Annual Conference on Neural Information Processing Systems (NeurIPS), 2021.

Learning View-Disentangled Human Pose Representation by Contrastive Cross-View Mutual Information Maximization

Long Zhao, Yuxiao Wang, Jiaping Zhao, Liangzhe Yuan, Jennifer J. Sun, Florian Schroff, Hartwig Adam, Xi Peng, Dimitris Metaxas, Ting Liu.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. [Oral Presentation]

SMIL: Multimodal Learning with Severely Missing Modality

Mengmeng Ma, Jian Ren, Long Zhao, Sergey Tulyakov, Cathy Wu, Xi Peng.

AAAI Conference on Artificial Intelligence (AAAI), 2021.

Maximum-Entropy Adversarial Data Augmentation for Improved Generalization and Robustness

Long Zhao, Ting Liu, Xi Peng, Dimitris Metaxas.

Annual Conference on Neural Information Processing Systems (NeurIPS), 2020.

Towards Image-to-Video Translation: A Structure-Aware Approach via Multi-Stage Generative Adversarial Networks

Long Zhao, Xi Peng, Yu Tian, Mubbasir Kapadia, Dimitris N. Metaxas.

International Journal of Computer Vision (IJCV), Volume 128, Issue 10, 2020.

Knowledge as Priors: Cross-Modal Knowledge Generalization for Datasets without Superior Knowledge

Long Zhao, Xi Peng, Yuxiao Chen, Mubbasir Kapadia, Dimitris N. Metaxas.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

[arXiv]

Learning to Learn Single Domain Generalization

Fengchun Qiao, Long Zhao, Xi Peng.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

Rethinking Kernel Methods for Node Representation Learning on Graphs

Yu Tian^*, Long Zhao^*, Xi Peng, Dimitris N. Metaxas.

Annual Conference on Neural Information Processing Systems (NeurIPS), 2019.

Construct Dynamic Graphs for Hand Gesture Recognition via Spatial-Temporal Attention

Yuxiao Chen, Long Zhao, Xi Peng, Jianbo Yuan, Dimitris N. Metaxas.

British Machine Vision Conference (BMVC), 2019.

Semantic Graph Convolutional Networks for 3D Human Pose Regression

Long Zhao, Xi Peng, Yu Tian, Mubbasir Kapadia, Dimitris N. Metaxas.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.

Cartoonish sketch-based face editing in videos using identity deformation transfer

Long Zhao, Fangda Han, Xi Peng, Xun Zhang, Mubbasir Kapadia, Vladimir Pavlovic, Dimitris N. Metaxas.

Computers & Graphics, Volume 79, 2019.

Learning to Forecast and Refine Residual Motion for Image-to-Video Generation

Long Zhao, Xi Peng, Yu Tian, Mubbasir Kapadia, Dimitris N. Metaxas.

European Conference on Computer Vision (ECCV), 2018.

CR-GAN: Learning Complete Representations for Multi-view Generation

Yu Tian, Xi Peng, Long Zhao, Shaoting Zhang, Dimitris N. Metaxas.

International Joint Conference on Artificial Intelligence (IJCAI), 2018.

Object Proposal by Multi-branch Hierarchical Segmentation

Chaoyang Wang, Long Zhao, Shuang Liang, Liqing Zhang, Jinyuan Jia, Yichen Wei.

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.

Learning best views of 3D shapes from sketch contour

Long Zhao, Shuang Liang, Jinyuan Jia, Yichen Wei.

The Visual Computer, Volume 31, Issue 6-8, 2015.

[PDF]

Size and Location Matter: A New Baseline for Salient Object Detection

Long Zhao, Shuang Liang, Yichen Wei, Jinyuan Jia.

Asian Conference on Computer Vision (ACCV), 2014.

Sketch-Based Retrieval Using Content-Aware Hashing

Shuang Liang, Long Zhao, Yichen Wei, Jinyuan Jia.

Pacific-Rim Conference on Multimedia (PCM), 2014.