Junjie Hu (胡君杰) Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen, Guangdong, China Email: hujunjie@cuhk.edu.cn

Biography

I am currently a researcher at the Shenzhen Institute of Artificial Intelligence and Robotics for Society in Shenzhen. I got my Ph.D. in the Graduate School of Information Sciences, Tohoku University, supervised by Prof. Takayuki Okatani in 2020. My research interest lies at the intersection of artificial intelligence and robotics, mainly including lifelong learning, 3D vision, robot perception, SLAM, and multi-robot perception.

News

• [01/2024] Two papers were accepted by ICRA'24.
• [12/2023] One paper on data-free depth distillation was accepted by KBS'23.
• [10/2023] One paper on lifelong depth estimation was accepted by TNNLS'23.
• [09/2023] One paper on LiDAR depth completion was accepted by RAL'23.
• [06/2023] Invited talk on scene depth estimation at the School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen. Slide is here.
• [06/2023] One paper on few-shot multi-agent perception was accepted by TPAMI'23.
• [04/2023] One paper on RGB-thermal based perception was accepted by RAL'23.
• [03/2023] Invited talk on robot perception at the School of Artificial Intelligence, South China Normal University. Slide is here.
• [03/2023] One paper on multi-robot SLAM was accepted by ICRA'23.
• [02/2023] One paper on scene recognition was accepted by TIM'23.
• [02/2023] One paper on heterogeneous multi-robot catching was accepted by TRO'23.
• [12/2022] The Survey on depth completion was accepted by TPAMI'22.

Journal Publications

• Dense Depth Distillation with Out-of-Distribution Simulated Images，
  Junjie Hu, Chenyou Fan, Mete Ozay, Hualie Jiang, Tin Lun Lam,
  Knowledge-Based Systems (KBS), 2023.
  paper
• Lifelong-MonoDepth: Lifelong Learning for Multi-Domain Monocular Metric Depth Estimation，
  Junjie Hu, Chenyou Fan, Liguang Zhou, Qing Gao, Honghai Liu, Tin Lun Lam,
  IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2023.
  paper, supplementary file , code
• Self-supervised Single-line LiDAR Depth Completion,
  Junjie Hu, Chenyou Fan, Xiyue Guo, Liguang Zhou, Tin Lun Lam,
  IEEE Robotics and Automation Letters (RAL), 2023.
  paper
• Few-Shot Multi-Agent Perception with Ranking-Based Feature Learning,
  Chenyou Fan, Junjie Hu and Jianwei Huang,
  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023.
  paper
• Explicit Attention-Enhanced Fusion for RGB-Thermal Perception Tasks,
  Mingjian Liang*, Junjie Hu*, Chenyu Bao, Hua Feng, Fuqin Deng, Tin Lun Lam,
  IEEE Robotics and Automation Letters (RAL), 2023.
  paper, code
• Asymmetric Self-play Enabled Intelligent Heterogeneous Multi-robot Catching System using Deep Multi-agent Reinforcement Learning,
  Yuan Gao, Junfeng Chen, Xi Chen, Chongyang Wang, Junjie Hu, Fuqin Deng, Tin Lun Lam,
  IEEE Transactions on Robotics (TRO), 2023.
  paper, demo
• Attentional Graph Convolutional Network for Structure-aware Audio-Visual Scene Classification,
  Liguang Zhou, Yuhongze Zhou, Xiaonan Qi, Junjie Hu, Tin Lun Lam, Yangsheng Xu,
  IEEE Transactions on Instrumentation and Measurement (TIM), 2023.
  paper
• Deep Depth Completion from Extremely Sparse Data: A Survey,
  Junjie Hu, Chenyu Bao, Mete Ozay, Chenyou Fan, Qing Gao, Honghai, Liu and Tin Lun Lam,
  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2022.
  paper, supplementary file
• BaSICNet: Lightweight 3D Hand Pose Estimation Network Based on Biomechanical Structure Information for Dexterous Manipulator Teleoperation,
  Wenrao Pang, Qing Gao, Yinan Zhao, Zhaojie Ju, Junjie Hu,
  IEEE Transactions on Cognitive and Developmental Systems , 2022.
  paper
• A Two-stage Unsupervised Approach for Low light Image Enhancement,
  Junjie Hu, Xiyue Guo, Junfeng Chen, Guanqi Liang, Fuqin Deng and Tin Lun Lam,
  IEEE Robotics and Automation Letters (RAL), 2021.
  paper, code
• Semantic Histogram Based Graph Matching for Real-Time Multi-Robot Global Localization in Large Scale Environment,
  Xiyue Guo, Junjie Hu, Junfeng Chen, Fuqin Deng and Tin Lun Lam,
  IEEE Robotics and Automation Letters (RAL), 2021.
  paper, demo, code
• Extending Information Maximization from a Rate--Distortion Perspective,
  Yan Zhang, Junjie Hu and Takayuki Okatani,
  Neurocomputing , 2020,
  paper, code
• PPM-SSNet: An Pyramid Pooling Module-based Semi-Siamese Network for End-to-end Building Damage Assessment from Satellite Imagery,
  Yanbing Bai, Junjie Hu, Jinhua Su, Xing Liu, Haoyu Liu, Xianwen He, Shengwang Meng, Erick Mas, Shunichi Koshimura,
  Remote Sensing, 2020.
  paper

Conference Publications

• From Satellite to Ground: Satellite Assisted Visual Localization with Cross-view Semantic Matching,
  Xiyue Guo, Haocheng Peng, Junjie Hu, Hujun Bao, Guofeng Zhang,
  IEEE International Conference on Robotics and Automation (ICRA), 2024.
  
• Meta-Reinforcement Learning Based Cooperative Surface Inspection of 3D Uncertain Structures using Multi-robot Systems,
  Junfeng Chen, Yuan Gao, Junjie Hu, Fuqing Deng, Tin Lun Lam,
  IEEE International Conference on Robotics and Automation (ICRA), 2024.
  paper
• MultiRoboLearn: An open-source Framework for Multi-robot Deep Reinforcement Learning,
  Junfeng Chen, Fuqin Deng, Yuan Gao, Junjie Hu, Xiyue Guo, Guanqi Liang, Tin Lun Lam,
  IEEE International Conference on Robotics and Biomimetics (ROBIO), 2023.
  paper, code
• Trajectory prediction with contrastive pre-training and social rank fine-tuning,
  Chenyou Fan, Haiqi Jiang, Aimin Huang, and Junjie Hu,
  International Conference on Neural Information Processing (ICONIP), 2023.
  paper
• Boosting Light-Weight Depth Estimation Via Knowledge Distillation,
  Junjie Hu, Chenyou Fan, Hualie Jiang, Xiyue Guo, Yuan Gao, Xiangyong Lu, Tin Lun Lam,
  The 16th International Conference on Knowledge Science, Engineering and Management (KSEM), 2023.
  paper, code
• Descriptor Distillation for Efficient Multi-robot SLAM,
  Xiyue Guo, Junjie Hu, Hujun Bao, Guofeng Zhang,
  IEEE International Conference on Robotics and Automation (ICRA), 2023.
  paper
• Private Semi-Supervised Federated Learning,
  Chenyou Fan, Junjie Hu, Jianwei Huang,
  International Joint Conferences on Artificial Intelligence (IJCAI), 2022.
  paper
• PLNet: Plane and Line Priors for Unsupervised Indoor Depth Estimation,
  Hualie Jiang, Laiyan Ding, Junjie Hu and Rui Huang,
  IEEE International Conference on 3D Vision (3DV), 2021.
  paper, code
• Few-Shot Multi-Agent Perception,
  Chenyou Fan, Junjie Hu and Jianwei Huang,
  ACM Multimedia (ACM MM), 2021.
  paper, code
• FEANet: Feature-Enhanced Attention Network for RGB-Thermal Real-Time Semantic Segmentation,
  Fuqin Deng, Hua Feng, Mingjian Liang, Hongmin Wang, Yong Yang, Yuan Gao, Junfeng Chen, Junjie Hu, Xiyue Guo, Tin Lun Lam,
  IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021.
  paper, code
• Visualization of Convolutional Neural Networks for Monocular Depth Estimation,
  Junjie Hu, Yan Zhang and Takayuki Okatani,
  IEEE International Conference on Computer Vision (ICCV), 2019.
  paper, supplementary file , code
• Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundaries,
  Junjie Hu, Mete Ozay, Yan Zhang and Takayuki Okatani,
  IEEE Winter Conference on Applications of Computer Vision (WACV), 2019.
  paper, code
• Non-rigid structure from motion via sparse self-expressive representation,
  Junjie Hu and Terumasa Aoki,
  IEEE International Conference on Image Processing (ICIP), 2017.
  paper
• A Convex Approach for Non-rigid Structure from Motion Via Sparse Representation,
  Junjie Hu and Terumasa Aoki,
  International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP), 2017.
  paper
• Improving the performance of non-rigid 3D shape recovery by points classification,
  Junjie Hu and Terumasa Aoki,
  IEEE International Conference on Machine Vision Applications (MVA), 2017.
  paper

Preprints

• Peer Learning for Unbiased Scene Graph Generation,
  Liguang Zhou, Junjie Hu, Yuhongze Zhou, Tin Lun Lam, Yangsheng Xu,
  arXiv preprint arXiv:2301.00146, 2022.
  paper
• Progressive Self-Distillation for Ground-to-Aerial Perception Knowledge Transfer,
  Junjie Hu, Chenyou Fan, Mete Ozay, Hua Feng, Yuan Gao, Tin Lun Lam,
  arXiv preprint arXiv:2208.13404, 2022.
  paper,code
• Analysis of Deep Networks for Monocular Depth Estimation Through Adversarial Attacks with Proposal of a Defense Method,
  Junjie Hu and Takayuki Okatani,
  arXiv preprint arXiv:1911.08790, 2019.
  paper

Honors & Awards

• Employee of the year, Shenzhen Institute of Artificial Intelligence and Robotics for Society, 2023
• Shenlong High-level Talent, Level C, Longgang district, Shenzhen city, 2021
• Overseas High-Caliber Personnel, Level C, Shenzhen City, 2020
• Best presentation award, Graduate School of Information Sciences, Tohoku University, 2019
• Global Hagi Scholarship, Tohoku University, 2018 - 2020
• Rotary Yoneyama Scholarship, 2016 - 2017

Professional Activities

• Conference Reviews:
  IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
  IEEE International Conference on Robotics and Automation (ICRA)
  European Conference on Computer Vision (ECCV)
  
• Journal Reviews:
  International Journal of Computer Vision (IJCV)
  IEEE Robotics and Automation Letters (RAL)
  IEEE Transactions on Instrumentation and Measurement (TIM)
  IEEE Transactions on Image Processing (TIP)
  IEEE Transactions on Intelligent Vehicles (TIV)
  IEEE Transactions on Mechatronics (TMECH)
  Knowledge-Based Systems (KBS)
  Neurocomputing
  Machine Learning