Brief Intro

He Zhang received his PhD from the department of Electrical and Computer Engineering at Rutgers, the State University of New Jersey under the supervision of Prof. Vishal M. Patel. His research focuses on Deep Learning, Sparse and Low-rank Representation, and Image Processing. Prior joining Rutgers, he received his Bachelor degree in Electronic Engineering and its Automation from Jiangnan University, China, where he was working with Prof. Le Yang.

Currently, He Zhang is a research scientist at Adobe. At Adobe, I am the main contributor for the Select Subject Portrait feature shipped in June 2020 Photoshop Release, Refine Hair feature shipped in 2020 Adobe MAX.

[CV] [Linkedin][Google Scholar][Github]

I am hiring self-motivated graduate student interns summer 2021. so please feel free to contact me if you are interested in working with me on a research project in computer vision, deep learning, and their real-world applications especially low-level vision and portrait editing related image editing technology.

News:

10/20: Refine Hair (combined with Object-aware Refine Edge) is shipped at 2020 Adobe MAX where an automatic trimap generation procedure is proposed. PiXimperfect **Check out others amazing features shipped by our group in this video PiXimperfect like amazing Sky replacement, amazing content-aware tracing.

10/20: One paper is accepted in WACV 2021.

06/20: My first feature in Photoshop. Select Subject Portrait is shipped in 2020 Photoshop June Release. Design an efficient framework for automatic high-quality portrait segmentation (with matted hair details ) running on mac & windows. [PiXimperfect] [PHLEARN]

11/19: One paper 'Generative Adversarial Networks with Fusion-discriminator for Single Image Dehazing' is accepted in AAAI-2020.

05/19: Two paper (1st author) accepted in IEEE TCSVT.

03/19: One paper is accepted in IEEE TMI.

01/19: One paper (1st author) is accepted in IJCV. We describe a synthesize method to solve the cross-domain face verification problem. Synthesis of High-Quality Visible Faces from Polarimetric Thermal Faces using Generative Adversarial Networks

09/18: I have successfully defended my PhD thesis. Thanks for all my committee members including: Dr. Peter Meer, Dr. Kristin Dana, Dr.Laleh Najafizadeh and Dr. Kevin S. Zhou

06/18: Two papers are accepted in BTAS'18. Unconstrained Face Detection Datasets (UFDD) has been released. Check the UFDD website.

04/18: Our team ranks the 1st place on ICME-2018 Heterogeneous Face Recognition: Polarimetric Thermal-to-Visible Matching.

04/18: Our team ranks 1st place overall (1st indoor and 3rd outdoor) in NTIRE-2018 Dehazing Challenge [code] and [paper].

03/18: I was invited to give a talk at State Key Lab of Rail Traffic Control

and Safety (Beijing Jiaotong University)

03/18: I am selected in the CVPR 2018 Doctor Consortium

02/18: Two papers are accepted in CVPR' 18 (both are with first author)

1. Density-aware Single Image De-raining using a Multi-stream Dense Network", CVPR'18 (PDF)(Code)

2. Densely Connected Pyramid Dehazing Network", CVPR'18, Salt Lake City, UT, 2018 (PDF)(Code)

12/17: One paper "Convolutional Sparse and Low-Rank Coding-Based Image Decomposition" is accepted to IEEE-TIP (pdf).

03/17: Our paper "Image De-raining Using a Conditional Generative Adversarial Network" has been covered by the The Outline [link] and Technology Review

Experiences:

  • Research Intern (Adobe Research, San Jose, CA) 05.2018-11.2018

  • Research Intern (Siemens Healthneer, Princeton, NJ) 01.2017-09.2017

Selected Academic Services:

  • Journal Reviewer: TPAMI, IJCV, TIP, TMM, PR, SPL, SP and Information Fusion.

  • Conference Reviewer: CVPR'19, ACCV'18, BMVC'18, WACV'19, ICCV'19, BMVC'19

Email: he.zhang92@rutgers.edu or hezhan@adobe.com
(not hezhang@adobe.com)