I am currently working as Boya Postdoctoral Researcher at Peking University with Prof. Song-Chun Zhu. Before that, I received Ph.D in Computer Science from EECS, Peking University, supervised by Prof. Yizhou Wang, and received B.Sc in Communication Engineering from Beijing Jiaotong University.

My current research interests are robot learning, multi-agent learning, and computer vision, particularly in building agents with physical and social common sense.

🔥 News

  • 2023.02:  🎉🎉 One paper was accepted by CVPR’23.
  • 2022.01:  🎉🎉 One paper was accepted by ICLR’23.

📝 Publications

* : co-first author, ✉ : corresponding author

CVPR 2023

GFPose: Learning 3D Human Pose Prior with Gradient Fields

Hai Ci, Mingdong Wu, Wentao Zhu, Xiaoxuan Ma, Hao Dong, Fangwei Zhong✉, Yizhou Wang

CVPR 2023, Project , Paper, Code code PWC

  • A versatile framework to model plausible 3D human poses in gradient fields for various applications.
ICLR 2023

Proactive Multi-Camera Collaboration for 3D Human Pose Estimation

Hai Ci*, Mickel Liu*, Xuehai Pan*, Fangwei Zhong✉, Yizhou Wang

ICLR 2023, Project , Paper

  • A novel MARL framework to solve proactive multi-camrea collaborations for 3D HPE in human crowds.
AAAI 2023

RSPT: Reconstruct Surroundings and Predict Trajectories for Generalizable Active Object Tracking

Fangwei Zhong*, Xiao Bi*, Yudi Zhang, Wei Zhang, Yizhou Wang

AAAI 2023 (Oral), Project, Paper

  • A framework to form a structure-aware motion representation by Reconstructing Surroundings and Predicting the target Trajectory.
NeurIPS 2022

TarGF: Learning Target Gradient Field to Rearrange Objects without Explicit Goal Specification

Mingdong Wu*, Fangwei Zhong*, Yulong Xia, Hao Dong

NeurIPS 2022, Project , Paper, Code

  • A framework based on a target gradient field trained by score-matching to tackle object rearrangement without explicit goal specification.
NeurIPS D&B 2022

MATE: Benchmarking Multi-Agent Reinforcement Learning in Distributed Target Coverage Control

Xuehai Pan*, Mickel Liu*, Fangwei Zhong✉, Yaodong Yang✉, Song-Chun Zhu, Yizhou Wang

NeurIPS 2022 Datasets and Benchmarks Track,

Project , Paper, Code code

  • A gamification of the multi-camera multi-target target coverage problem, and an all-in-one multi-agent reinforcement learning benchmark
ICML 2022

Disentangling Disease-related Representation from Obscure for Disease Prediction

Chu-Ran Wang, Fei Gao, Fangwei Zhong✉, Fangwei Zhong, Yizhou Yu, Yizhou Wang

ICML 2022, Paper

  • A disentanglement learning strategy under the guidance of alpha blending generation in an encoder-decoder framework (DAB-Net).
ICLR 2022

ToM2C: Target-oriented Multi-agent Communication and Cooperation with Theory of Mind

Yuanfei Wang*, Fangwei Zhong*, Jing Xu, Yizhou Wang

ICLR 2022, Paper, Code code

  • A Target-oriented Multi-agent Communication and Cooperation mechanism using Theory of Mind.
ICML 2021

Towards Distraction-Robust Active Visual Tracking

Fangwei Zhong, Peng Sun, Wenhan Luo, Tingyun Yan, Yizhou Wang

ICML 2021, Project, Paper, Code code, Environment code

  • A mixed cooperative-competitive multi-agent game: a target and multiple distractors form a collaborative team to play against a tracker.
  • A bunch of practical methods: a reward function for distractors, a cross-modal teacher-student learning strategy, and a recurrent attention module for the tracker.
NeurIPS 2020

Learning Multi-Agent Coordination for Enhancing Target Coverage in Directional Sensor Networks

Jing Xu*, Fangwei Zhong*, Yizhou Wang

NeurIPS 2021 Project, Paper, Code code,

  • a Hierarchical Target-oriented Multi-Agent Coordination (HiT-MAC), which decomposes the target coverage problem into two-level tasks: targets assignment by a coordinator and tracking assigned targets by executors.
AAAI 2020

Pose-Assisted Multi-Camera Collaboration for Active Object Tracking

Jing Li*, Jing Xu*, Fangwei Zhong*, Xiangyu Kong, Yu Qiao, Yizhou Wang

AAAI 2020, Project, Paper, Code code, Environment code, Demo

  • An efficient yet effective multi-camera collaboration system for collaborative multiCamera active object tracking.

AD-VAT+: An Asymmetric Dueling Mechanism for Learning and Understanding Visual Active Tracking

Fangwei Zhong, Peng Sun, Wenhan Luo, Tingyun Yan, Yizhou Wang

IEEE TPAMI, Paper, Code code, Environment code

  • Employ more advanced environment augmentation technique and two-stage training strategies to improve the performance of the tracker in the case of challenging scenarios such as obstacles.
  • Analyze the target’s behaviors as the training proceeds and visualize the latent space of the tracker for a better understanding.
ICLR 2019

AD-VAT: An Asymmetric Dueling mechanism for learning Visual Active Tracking

Fangwei Zhong, Peng Sun, Wenhan Luo, Tingyun Yan, Yizhou Wang

ICLR 2019, Paper, Code code, Environment code

  • A novel adversarial RL method which adopts an Asymmetric Dueling mechanism (tracker vs. target) for robust active visual tracking
CVPR 2019

CRAVES: Controlling Robotic Arm with a Vision-based, Economic System

Yiming Zuo*, Weichao Qiu*, Lingxi Xie, Fangwei Zhong, Yizhou Wang, Alan L Yuille

CVPR 2019, Project, Paper, Code code, Controller code, Environment code

  • A vision system for low-cost arm control: trains a vision model in virtual environment, and applies it to real-world images after domain adaptation (a semi-supervised approach).
  • One virtual environment for collection data and reinforcement learning.
  • Two real-world datasets for evaluation.

End-to-end Active Object Tracking and Its Real-world Deployment via Reinforcement Learning

Wenhan Luo*, Peng Sun*, Fangwei Zhong*, Wei Liu, Tong Zhang, Yizhou Wang

IEEE TPAMI, Paper, Code code, Environment code

  • Deploy End-to-end active object tracker trained in virtual environment in real-world robot.
ICML 2018

End-to-end Active Object Tracking via Reinforcement Learning

Wenhan Luo*, Peng Sun*, Fangwei Zhong, Wei Liu, Tong Zhang, Yizhou Wang

ICML 2018, Project, Paper, Training code, Gym-tvizdoom, Gym-unrealcv

  • An end-to-end active objet tracking solution via deep reinforcement learning, where a ConvNet-LSTM function approximator is adopted for the direct frame-to-action prediction.
WACV 2018

Detect-SLAM: Making Object Detection and SLAM Mutually Beneficial

Fangwei Zhong, Sheng Wang, Ziqi Zhang, Chen Zhou, Yizhou Wang

WACV 2018, Paper, Video

  • A robotic vision system which integrates SLAM with a deep neural network-based object detector to make the two functions mutually beneficial.
ACM MM 2017

Unrealcv: Virtual worlds for computer vision

Weichao Qiu, Fangwei Zhong, Yi Zhang, Siyuan Qiao, Zihao Xiao, Tae Soo Kim, Yizhou Wang, Alan Yuille

ACM Multimedia 2017 Open Source Software Competition,

Project, Paper, Code code,

  • An open-sourced project to help computer vision researchers build virtual worlds using Unreal Engine 4 (UE4).

🎖 Selected Honors and Awards

  • 2021.09 ACM China SIGAI Doctoral Dissertation Award.
  • 2021.06 China National Postdoctoral Program for Innovative Talents.
  • 2021.06 Outstanding Graduate in Peking University and Beijing.
  • 2021.05 Top 10 Outstanding Research Award in the School of EECS, Peking University.
  • 2016.08 The Second Place in DJI RoboMasters Summer Camp Competition (2/10).

📖 Educations

  • 2016.09 - 2021.07, Peking University, Ph.D in Computer Science.
  • 2012.09 - 2016.07, Beijing Jiaotong University, B.S. in Communication Engineering.

💻 Professional Service

  • Journal Reviewer: Nature Machine Intelligence, IJRR, IEEE TIP, IEEE TVT, IEEE RA-L, ACM TOMM, Sensors, IEEE Access
  • Conference Reviewer: ICML 2020/21/22/23, NeurIPS 2020/21/22, ICLR 2020/21/22/23, CVPR 2020/21/22/23, ICCV 2019/21/23, ECCV 2020/22, ACCV 2020, WACV 2021/22/23
  • PC Member: AAAI 2020/21/22/23