Intelligent Vision for Robotics in Complex Environment

News

2022-06: Four papers accepted by IROS 2022!

Our papers “Siamese Object Tracking for Vision-Based UAM Approaching with Pairwise Scale-Channel Attention”, “Local Perception-Aware Transformer for Aerial Tracking”, “HighlightNet: Highlighting Low-Light Potential Features for Real-Time UAV Tracking”, and “End-to-End Feature Decontaminated Network for UAV Tracking”, are accepted by IROS 2022 in Kyoto, Japan!

2022-03: Two papers accepted by CVPR 2022!

Our papers “Unsupervised Domain Adaptation for Nighttime Aerial Tracking” and “TCTrack: Temporal Contexts for Aerial Tracking” are accepted by CVPR 2022 in New Orleans, Louisiana, USA!

2022-02: Two papers accepted by ICRA 2022!

Our papers “Tracker Meets Night: A Transformer Enhancer for UAV Tracking” (also for IEEE Robotics and Automation Letters, RA-L) and “Ad2Attack: Adaptive Adversarial Attack on Real-Time UAV Tracking” are accepted by ICRA 2022 in Philadelphia (PA), USA!

2021-11: Academic Stars of Tongji!

- Bowen Li is awarded the Academic Star Model of Tongji University (top 5 among all undergraduate students of Tongji).

- Ziang Cao is awarded the Academic Star of Tongji University (top 15 among all undergraduate students of Tongji).

Congratulations!

2021-10: National Scholarship of China!

- Fuling Lin is awarded the National Scholarship of China for Graduate Students.

- Bowen Li is awarded the National Scholarship of China for Undergraduate Students.

Congratulations!

2021-07: One paper accepted by ICCV 2021!

Our paper “HiFT: Hierarchical Feature Transformer for Aerial Tracking” is accepted by ICCV 2021!

2021-07: Two papers accepted by IROS 2021!

Our papers “SiamAPN++: Siamese Attentional Aggregation Network for Real-Time UAV Tracking”, and “DarkLighter: Light up the Darkness for UAV Tracking” are accepted by IROS 2021 in Prague, Czech Republic!

2021-03: Latency-aware tracking benchmark

We formulate a latency-aware benchmark with new metrics for a more realistic evaluation. A new predictive visual tracking baseline is developed to compensate for the latency stemming from the onboard computation. More details can be found here!

2021-02: Four papers accepted by ICRA 2021!

Our papers “Siamese Anchor Proposal Network for High-Speed Aerial Tracking”, “ADTrack: Target-Aware Dual Filter Learning for Real-Time Anti-Dark UAV Tracking”, “Online Recommendation-based Convolutional Features for Scale-Aware Visual Tracking”, and “Mutation Sensitive Correlation Filter for Real-Time UAV Tracking with Adaptive Hybrid Label”, are accepted by ICRA 2021 in Xi’an, China!

2021-02: The best review of object tracking in 2020

Our paper “Correlation Filters for Unmanned Aerial Vehicle-Based Aerial Tracking: A Review and Experimental Evaluation” is awarded as the best review of object tracking in 2020! (See more)

2021-02: UAVDark135

A pioneering UAV dark tracking benchmark consists of 135 videos with a variety of objects. (See more)

2020-07: Five papers accepted by IROS 2020!

Our papers “DR^2Track: Towards Real-Time Visual Tracking for UAV via Distractor Repressed Dynamic Regression”, “Augmented Memory for Correlation Filters in Real-Time UAV Tracking”, “Automatic Failure Recovery and Re-Initialization for Online UAV Tracking with Joint Scale and Aspect Ratio Optimization”, “Towards Robust Visual Tracking for Unmanned Aerial Vehicle with Tri-Attentional Correlation Filters”, and “Learning Consistency Pursued Correlation Filters for Real-Time UAV Tracking” are accepted by IROS 2020 in Las Vegas, USA!

2020-07: Outstanding Graduate Awards

- Yiming Li, Weijiang Xiong, Junjie Ye and Yujie He are awarded with Outstanding Graduate of Shanghai.

- Juntao Xu is awarded with Outstanding Graduate of Tongji University.

Congratulations!

2020-02: One paper accepted by CVPR 2020!

Our paper “AutoTrack: Towards High-Performance Visual Tracking for UAV with Automatic Spatio-Temporal Regularization” is accepted by CVPR 2020 in Seattle, United States!

2020-01: Three papers accepted by ICRA 2020!

Our papers “BiCF: Learning Bidirectional Incongruity-Aware Correlation Filter for Efficient UAV Object Tracking”, “Training-set Distillation for Real-Time UAV Object Tracking, and “Keyfilter-Aware Real-Time UAV Object Tracking”, are accepted by ICRA 2020 in Paris, France!

2019-11: IROS 2019

The 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019) will be held on November 4 – 8, 2019 in Macau, China.

Welcome to the Workshop on Fast Neural Perception and Learning for Intelligent Vehicles and Robotics !

2019-07: One paper accepted by ICCV 2019!

Our paper “Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking” is accepted by ICCV 2019 in Seoul, Korea!

2019-07: One paper accepted by IROS 2019!

Our paper “Boundary Effect-Aware Visual Tracking for UAV with Online Enhanced Background Learning and Multi-Frame Consensus Verification” is accepted by IROS 2019 in Macau, China!

2019-06: Best Bachelor Thesis Awards

-Yiyong Sun is awarded with Best Bachelor Thesis of Tongji University.

- Fuling Lin is awarded with Best Bachelor Thesis Scholarship of Tongji-Hirschvogel.

Congratulations!

Conference Papers

In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, pp. 1-8, 2022.

In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, pp. 1-8, 2022.

In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, pp. 1-8, 2022.

In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, pp. 1-8, 2022.

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, Louisiana, USA, …

Benchmarks

NAT2021 is a pioneering benchmark designed for unsupervised domain adaptive nighttime tracking.

DarkTrack2021 is a nighttime tracking benchmark comprises 110 challenging sequences with 100K frames in total.

UAMT100 benchmark is built for UAM tracking method evaluation. It contains 100 image sequences recorded on a flying UAM platform.

Predictive visual tracking (PVT) is a latency-aware benchmark jointly assessing the tracking accuracy and efficiency of trackers.

A pioneering UAV dark tracking benchmark consists of 135 videos with a variety of objects.

People

Principal investigator

Avatar

Changhong Fu

Associate Professor

Artificial intelligence, Computer vision, Robotics

Engineers / Research Assistants

Avatar

Sihang Li

BEng in Mechanical Engineering

Robotics, Adversarial machine learning, Computer vision

Avatar

Liguo Zhang

M.Eng. in Mechanical Engineering

Visual object tracking, Unmanned aerial manipulator, Deep learning

Avatar

Teng Li

BEng in Mechanical Engineering

Visual object tracking, Machine learning, Robotics

Avatar

Xiaogang Yu

M.Eng. in Mechanical Engineering

Robotics, Visual object tracking, Unmanned aerial vehicle

PhD Students

Avatar

Haowen Zheng

PhD in Mechanical Engineering

Deep learning, Computer version, Edge intelligence

Master students

Avatar

Haolin Dong

MSc in Mechanical Engineering

Low-light image enhancement, Machine learning, Deep learning

Avatar

Jilin Zhao

MSc in Mechanical Engineering

Visual object tracking, Deep learning, Computer vision

Avatar

Junjie Ye

MSc in Mechanical Engineering

Visual object tracking, Deep learning, Robotics

Avatar

Weiyu Peng

MSc in Mechanical Engineering

Deep learning, Robotics, Visual Object tracking

Avatar

Kunhan Lu

MSc in Mechanical Engineering

Visual object tracking, Deep learning, Computer vision

Avatar

Yihu Wang

MSc in Mechanical Engineering

Visual object tracking, Deep learning, Robotics

Avatar

Yuchen Li

MSc in Mechanical Engineering

Visual object tracking, Machine learning, Deep learning

Avatar

Zijie Zhang

MSc in Mechanical Engineering

Adversarial attack, Computer vision, Robotics

Bachelor students

Avatar

Haobo Zuo

BEng in Mechanical Engineering

Low-light image enhancement, Visual object tracking, Unmanned aerial vehicle, Computer vision

Avatar

Haoyang Li

BEng in Mechanical Engineering

Object segmentation, Computer vision, Robotics

Avatar

Liangliang Yao

BEng in Mechanical Engineering

Low-light image enhancement, Visual object tracking, Machine learning, Computer vision

Avatar

Mutian Cai

BEng in Mechanical Engineering

Low-light image enhancement, Visual object tracking, Machine learning, Computer vision

Avatar

Xining Lu

BEng in Mechanical Engineering

Visual object tracking, Unmanned aerial vehicle, System automatization, Computer vision

Avatar

Yucheng Wang

BEng in Automation

Robotics, Computer vision, Visual object tracking, Unmanned aerial vehicle

Previous members

Avatar

Bowen Li

PhD Candidate in Robotics at CMU, USA

Visual object tracking, Unmanned aerial vehicle, Machine learning, Computer vision

Avatar

Guangze Zheng

PhD Candidate in Computer Science at HKU, China

Visual object tracking, Robotics, Machine learning

Avatar

Ziang Cao

PhD Candidate in Computer Science at NTU, Singapore

Visual object tracking, Deep learning, Unmanned aerial vehicle

Avatar

Fangqiang Ding

PhD Candidate in Robotics and Autonomous System at Edin., U.K.

Drone perception, Multi-modal fusion, Robotics

Avatar

Fuling Lin

PhD Candidate in Mechanical Engineering at HKU, Hong Kong, China

Visual object tracking, Unmanned aerial vehicle, Machine learning

Avatar

Ziyuan Huang

PhD Candidate in Mechanical Engineering at NUS, Singapore

Visual tracking for UAVs, Artificial Intelligence, Cognitive Systems

Avatar

Xinnan Yuan

MSc in Artificial Intelligence at NTU, Singapore

Visual object tracking, Adversarial attack, Computer vision

Avatar

Ran Duan

PhD Candidate at PolyU, Hong Kong, China

Image processing, Visual tracking, UAV visual navigation

Avatar

Jin Jin

MSc in Management, Technology, and Economics at ETH, Zurich, Switzerland

Robotics, Visual object tracking, Machine learning, Unmanned aerial vehicle

Avatar

Yiming Li

PhD Candidate in Robotics at NYU, USA

Visual object tracking, Pose estimation, Robotics

Avatar

Weijiang Xiong

MSc in Robotics, Aalto University, Finland

Visual object tracking, Machine learning

Avatar

Yujie He

MSc in Robotics at EPFL, Switzerland

Visual object tracking, Place recognition

Avatar

Juntao Xu

PhD candidate in Mechanical Engineering at HKU, Hong Kong, China

Visual object tracking, Artificial Intelligence, Robotics and Automation

Avatar

Changjing Liu

MSc in Instrument Science and Technology at SJTU, China

Robotics, Visual object tracking

Avatar

Xiaoxiao Yang

MSc in Automation at SJTU, China

Visual object tracking, Deep Learning, Robotics

Avatar

Yinqiang Zhang

MSc in Mechatronics and Information Technology at TUM, Germany

Visual object tracking, Computer vision

Avatar

Hao Chen

Computer Engineer at iFLYTEK, China

Visual SLAM, Unmanned aerial vehicle, Deep learning

Avatar

Jianqiao Lu

PhD candidate in Computer Science at HKU, China.

Biorobotics, Unmanned aerial vehicle, Machine learning

Avatar

Shaoqiu Xu

MSc in Mechanical Engineering, SJTU, China

Obejct tracking, Unmanned aerial vehicle, Pose estimation, Machine learning

Avatar

Xiangpeng Zeng

Computer Engineer at Eastmoney Securities, China

Object detection, Deep learning, Unmanned aerial vehicle

Avatar

Yiyong Sun

MSc in Mechanical Engineering at NUS, Singapore

Robotics, Visual object tracking, Unmanned aerial vehicle, Machine learning

Avatar

Yulin Li

MSc in Mechanical Engineering at UCSD, USA

Biorobotics, Unmanned aerial vehicle, Machine learning

Avatar

Zheng Shen

MSc in Mechanical Engineering at ETH, Switzerland

Fuzzy control, Unmanned aerial vehicle

Contact

  • Kaiwu Building, Tongji University, No.4800 Cao’an Highway, Jiading District, Shanghai #201804, China
  • changhongfu@tongji.edu.cn