Zhongdao Wang

I am a third-year Ph.D candidate at the Department of Electronic Engineering, Tsinghua Univerisity, advised by Prof. Shengjin Wang. Before that, I received bechalor's degree in Mathematics and Physics from Tsinghua University. My current research interests are task-oriented self-supervised learning, espicially applications on multiple object tracking (MOT) and re-identification (re-ID). Previously, I also did research on metric learning and its applications on face recognition, person/vehicle re-identification. I work closely with Prof. Liang Zheng.

Email  /  CV  /  Google Scholar  /  Github

profile photo

  • [NEW] Two papers are accepted to ECCV 2020!
  • One paper is accepted to CVPR 2020 as oral presentation!
  • One paper is accepted to AAAI 2020 as oral presentation!
  • One paper isaccepted to CVPR 2019.

Conference papers

CycAs: Self-supervised Cycle Association for Learning Re-identifiable Descriptions
Zhongdao Wang, Jingwei Zhang, Liang Zheng, Yixuan Liu, Yifan Sun, Yali Li, Shengjin Wang
ECCV, 2020

Learning re-identifiable features by self-supervised learning. The supervision signal is cycle consistency emerging from instance association in a forward-then-backward cycle.

Towards Real-time Multi-Object Tracking
Zhongdao Wang, Liang Zheng, Yixuan Liu, Yali Li, Shengjin Wang
ECCV, 2020
[code] star

By incorporating the appearance embedding model into the detector, we introduce JDE, the first open-source real-time multiple object trackor with a running speed of 22 ~ 38 FPS. This speed takes all the steps into account, including detection, appearance embedding extraction and association. Code is released! If you are looking for an easy-to-use and fast pedestrian detector/tracker, JDE is a good option!

Circle Loss: A Unified Perspective of Pair Similarity Optimization
Yifan Sun, Changmao Cheng, Yuhan Zhang, Chi Zhang, Liang Zheng, Zhongdao Wang, Yichen Wei,
CVPR, 2020   (Oral Presentation)

A unified perspective for proxy-based and pair-based metric learning.

Softmax Dissection: Towards Understanding Intra-and Inter-class Objective for Embedding Learning
Lanqing He*, Zhongdao Wang*, Yali Li, Shengjin Wang (* indicates equal contribution)
AAAI, 2020   (Oral Presentation)

Investigation on intra- and inter-class objectives of the softmax cross-entropy loss function, and a new loss that dissects the two parts for accelerating massive classification.

Linkage based face clustering via graph convolution network
Zhongdao Wang, Liang Zheng, Yali Li, Shengjin Wang
CVPR, 2019
[code] star

A supervised solution to the face clustering task using graph convolutional networks.

Orientation invariant feature embedding and spatial temporal regularization for vehicle re-identification
Zhongdao Wang*, Luming Tang*, Xihui Liu, Zhuliang Yao, Shuai Yi, Jing shao, Junjie Yan, Shengjin Wang, Hongsheng Li, Xiaogang Wang (* indicates equal contribution)
ICCV, 2017
[Key point annotation for Veri-776 dataset] star

An orientation-invariant solution to the vehicle re-identification problem.

Journal papers

Node-Adaptive Multi-Graph Fusion Using Extreme Value Theory
Jingwei Zhang, Zhongdao Wang, Yali Li, Shengjin Wang,
IEEE Singnal Processing Letters (SPL), 2020

Multi-graph fusion using extreme value theory for multi-view clustering.

Design and source code modified based on Jon Barron's website.