Carview!

CARVIEW

MOTORHOMES

Select Language

HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Wed, 29 Oct 2025 03:36:42 GMT access-control-allow-origin: * etag: W/"69018bca-89e6" expires: Sun, 28 Dec 2025 21:50:13 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: EA44:3157C7:7F654E:8F15BE:6951A3BD accept-ranges: bytes age: 0 date: Sun, 28 Dec 2025 21:40:13 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210086-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1766958013.213849,VS0,VE215 vary: Accept-Encoding x-fastly-request-id: c570f946981ef633e792163dcaccd138006dbbcc content-length: 10195 Zhe Wang (Allen) @ Computational Vision Group -- UC Irvine

Zhe Wang

Email: buptwangzhe2012[at]gmail[dot]com

About Me

I am an AI Research Scientist at Meta, where I build and ship next-generation video generation systems. Previously, I was a member of Adobe Firefly, serving as a main contributor to Video Extension, Text-to-Video, and RGBA Video generation. Before that, I led cross-city and cross-platform model development and deployment at Cruise, driving large-scale perception modeling into production. Earlier in my career, I interned at AWS Rekognition and Adobe Research, working across computer vision, deep learning, and multimodal generation. My research has garnered 8,000+ citations on Google Scholar. I obtained my Ph.D. from UC Irvine where I was advised by Prof. Charless Fowlkes.

News

News! [Oct, 2025] We are hiring AI Research Scientist intern working on video generation in Meta for 2026, contact me directly if you are interested ( LINK ).
News! [Oct, 2025] Generate videos with transparent backgrounds is on Firefly Web App ( BLOG ).
[Oct, 2024] We are hiring research intern working on T2V/T2I/VLM in Firefly for 2025 summer, contact me directly if you are interested ( LINK ).
[Oct, 2024] Generate Video (beta) is on Firefly Web App ( BLOG ).
[Sep, 2024] We are bringing generative AI (T2V and GenExtend Model) to video with Adobe Firefly Video Model ( BLOG ).
[Mar, 2024] I moved to generative computer vision after working about 10 years in CV related perception problems.
[Jul, 2022] One paper on multi-view 3d human pose estimation gets accepted by ECCV 2022.
[May, 2022] I uploaded my phd thesis and presentation, have a look if you are interested!
[Feb, 2022] Thanks for the continuous interest from Academia for our GPA dataset, I recently provide single video setting meta file, and improve the gaussian blur quality (oval shape blur) to mitigate defect of gaussian blur.
[Nov, 2021] I succesfully defended my Ph.D dissertation!

Industrial Highlights

Firefly Video Foundation Model
Firefly Video model supports text-to-video and image-to-video generation. This is in public beta now through Adobe Firefly.com.
2024 & 2025
[ Link ]

GenExtend
GenExtend extends an input video clip temporally using generative models. This is in public beta now in Premiere Pro.
2024
[ Link ]

Cruise Origin
Deploy the Cruise Origin self-driving vehicle — a steering wheel-free car platform — in San Francisco and Austin.
2023
[ Link ]

Cruise Delta
Launch the Cruise Delta self-driving vehicle in San Francisco and Austin.
2022 & 2023
[ Link ]

Selected Publications [ Full List ]

Robust Estimation of 3D Human Body Pose with Geometric Priors

PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimation
H. Ma, Z. Wang, Y. Chen, D. Kong, L. Chen, X. Liu, X. Yan, H. Tang, and X. Xie
European Conference on Computer Vision (ECCV) , 2022.
[ Paper ] [ Code ]

Robust Estimation of 3D Human Body Pose with Geometric Priors
Z. Wang
Ph.D thesis, 2021
[ Slides ] [ Paper ] [ BibTex ]

The Best of Both Worlds: Combining Model-based and Nonparametric Approaches for 3D Human Body Estimation
Z. Wang, J. Yang, and C. Fowlkes
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), ABAW workshop, 2022
[ Paper ] [ BibTex ]

SSCAP: Self-supervised Co-occurrence Action Parsing for Unsupervised Temporal Action Segmentation
Z. Wang, H. Chen, X. Li, C. Liu, Y. Xiong, J. Tighe, and C. Fowlkes
IEEE Winter Conf. on Applications of Computer Vision (WACV), 2022
[ Paper ] [ BibTex ] [ Poster ] [ Slides ]

Predicting Camera Viewpoint Improves Cross-dataset Generalization for 3D Human Pose Estimation
Z. Wang, D. Shin, and C. Fowlkes
European conference on computer vision (ECCV), 3DPW workshop, 2020
[ Paper ] [ Project Page ] [ BibTex ] [ Arxiv ] [pre-trained GPA model]

Geometric Pose Affordance: 3D Human Pose with Scene Constraints
Z. Wang, L. Chen, S. Rathore, D. Shin, and C. Fowlkes
(Arxiv), 2019
[ Paper ] [ Project Page ] [ Dataset Video ] [Dataset & Model] [ BibTex ]

Structured Triplet Learning with POS-tag Guided Attention for Visual Question Answering
Z. Wang, X. Liu, L. Chen, L. Wang, Y. Qiao, X. Xie, and C. Fowlkes
IEEE Winter Conf. on Applications of Computer Vision (WACV), 2018
[ Paper ] [ Poster ] [ Presentation ] [ Code ] [ Talk Video ] [ BibTex ]

Weakly Supervised PatchNets: Describing and Aggregating Local Patches for Scene Recognition
Z. Wang, L. Wang, Y. Wang, B. Zhang, and Y. Qiao
IEEE Transactions on Image Processing (TIP), 2017
[ Paper ] [ Extended Abstract ] [ Code& Model ] [ Feature ] [ Project Page ] [ BibTex ]

Transferring Object-Scene Convolutional Neural Networks for Event Recognition in Still Images
L. Wang, Z. Wang, Y. Qiao, and L. Van Gool
International Journal of Computer Vision (IJCV), 2018.
[ Paper ] [ Code& Model ] [ BibTex ]

Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
L. Wang, Y. Xiong, Z. Wang, Y. Qiao, D. Lin, X. Tang, and L. Van Gool
European Conference on Computer Vision (ECCV), 2016.
[ Paper ] [ Poster ] [ Code ] [ BibTex ]

Real-time Action Recognition with Enhanced Motion Vector CNNs
B. Zhang, L. Wang, Z. Wang, Y. Qiao, and H. Wang
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
[ Paper ] [ Project Page ] [ BibTex ]

Academic Service

Journal Reviewer

TVCG, TIP, TMM, PR, IJCV, TPAMI, etc

Conference Reviewer

ICML, AAAI, ICLR, NeurIPS, CVPR, ICCV, WACV, BMVC, ECCV, 3DV, Socal NLP, ICCV workshops, CVPR workshops, etc

TA/Readers:

ICS 6D, Discrete Math for Computer Science. ( Fall 2016 )

CS 175, Projects in AI in Minecraft. ( Winter 2017 )

CS 116, Computational Photography and Vision. ( Spring 2017 )

Invited Talks

Google Research (03/2024)
Apple (02/2024)
ECCV 2022 (10/2022)
CVPR 2022 (06/2022)
Electronic Arts (08/2021)
OPPO US Research Center (07/2021)
WACV 2018, Lake tahoe (03/2018)
Shenzhen Institute of Advanced Technology (01/2018)
ICASSP, Shanghai (03/2016)
ICCV 2015, Santiago de Chile, Chile (12/2015)

Contests

ActivityNet Large Scale Activity Recognition Challenge 2016: Untrimmed Video Classification (in conjunction with CVPR 2016), Rank: 1st place
ChaLearn Looking at People Challenge: Cultural Event Classification (in conjunction with ICCV 2015), Rank: 3rd place
THUMOS'15 Challenge: Action Classification (in conjunction with CVPR 2015), Rank: 5th place
ChaLearn Looking at People Challenge: Cultural Event Classification (in conjunction with CVPR 2015), Rank: 1st place
ChaLearn Looking at People Challenge: Action/Interaction Classification (in conjunction with CVPR 2015), Rank: 1st place
First Audio and Video Competition: Video Classification, Rank: 5th place

Outstanding interns I have worked with

Peize Sun (HKU), 2023, Cruise LLC

Bhavya Goyal (UW Madison), 2022, Cruise LLC

This site has been visisted

times in total.

Published with GitHub Pages

Original Source | Taken Source