Junyi (Kevin) Ma

About Me

• Ph.D. candidate at IRMV Lab, Shanghai Jiao Tong University | Supervisor: Prof. Hesheng Wang
• M.S. in Mechanical Engineering, Beijing Institute of Technology (2023) | Supervisors: Prof. Guangming Xiong & Prof. Xieyuanli Chen (NUDT)
• B.S. in Mechanical Engineering, Beijing Institute of Technology (2020) | Bachelor Thesis Advisor: Prof. Oliver Dürr (Hochschule Konstanz​)

News

• [May 2026] Our paper Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views is accepted by T-PAMI.
• [Nov. 2025] Our paper MADiff: Motion-Aware Mamba Diffusion Models for Hand Trajectory Prediction on Egocentric Videos is accepted by T-PAMI.
• [Jun. 2025] Four papers are accepted by IROS 2025.
• [Feb. 2025] Our paper Spatiotemporal Decoupling for Efficient Vision-Based Occupancy Forecasting is accepted by CVPR 2025. It is a follow-up work of our Cam4DOcc (CVPR 2024).

SELECTED Publications

Learning from Human Videos

• Robot Learning from Human Videos: A Survey
  Junyi Ma, Erhang Zhang, Haoran Yang, Ditao Li, Chenyang Xu, Guangming Wang, Hesheng Wang*
  [PDF] [Paper list]
  

• Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views
  Junyi Ma, Wentao Bao, Jingyi Xu, Guanzhong Sun, Yu Zheng, Erhang Zhang, Xieyuanli Chen, Hesheng Wang*
  IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2026.
  [PDF] [Page] [Code]
  

• MADiff: Motion-Aware Mamba Diffusion Models for Hand Trajectory Prediction on Egocentric Videos
  Junyi Ma#, Xieyuanli Chen#, Wentao Bao, Jingyi Xu, Hesheng Wang*
  IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2025.
  [PDF] [Page]
  

• Zero-Shot Temporal Interaction Localization for Egocentric Videos
  Erhang Zhang#, Junyi Ma#, Yin-Dong Zheng, Yixuan Zhou, Hesheng Wang*
  IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025.
  [PDF] [Code]
  

• EgoLoc: A Generalizable Solution for Temporal Interaction Localization in Egocentric Videos
  Junyi Ma#, Erhang Zhang#, Yin-Dong Zheng, Yuchen Xie, Yixuan Zhou, Hesheng Wang*
  [PDF] [Code]
  

HOI Prediction

• MMTwin: Novel Diffusion Models for Multimodal 3D Hand Trajectory Prediction
  Junyi Ma, Wentao Bao, Jingyi Xu, Guanzhong Sun, Xieyuanli Chen, Hesheng Wang*
  IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025.
  [PDF] [Page] [Code]
  

• Diff-IP2D: Diffusion-Based Hand-Object Interaction Prediction on Egocentric Videos
  Junyi Ma, Jingyi Xu, Xieyuanli Chen, Hesheng Wang*
  IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025.
  [PDF] [Code]
  

Place Recognition and SLAM

• OverlapTransformer: An Efficient and Yaw-Angle-Invariant Transformer Network for LiDAR-Based Place Recognition
  Junyi Ma, Jun Zhang, Jintao Xu, Rui Ai, Weihao Gu, and Xieyuanli Chen*
  IEEE Robotics and Automation Letters (RA-L), 2022, and IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022.
  [PDF] [Code]
  

• SeqOT: A Spatial-Temporal Transformer Network for Place Recognition Using Sequential LiDAR Data
  Junyi Ma, Xieyuanli Chen, Jingyi Xu, Guangming Xiong*
  IEEE Transactions on Industrial Electronics (TIE), 2022.
  [PDF] [Code]
  

• CVTNet: A Cross-View Transformer Network for Place Recognition Using LiDAR Data
  Junyi Ma, Guangming Xiong, Jingyi Xu, Xieyuanli Chen*
  IEEE Transactions on Industrial Informatics (TII), 2023.
  [PDF] [Code]
  

Point Cloud and Occupancy Forecasting

• Cam4DOcc: Benchmark for Camera-Only 4D Occupancy Forecasting in Autonomous Driving Applications
  Junyi Ma#, Xieyuanli Chen#, Jiawei Huang, Jingyi Xu, Zhen Luo, Jintao Xu, Weihao Gu, Rui Ai, Hesheng Wang*
  IEEE/CVF Conf.~on Computer Vision and Pattern Recognition (CVPR), 2024
  [PDF] [Code]
  

• PCPNet: An Efficient and Semantic-Enhanced Transformer Network for Point Cloud Prediction Mentorship
  Zhen Luo, Junyi Ma, Zijie Zhou, Guangming Xiong
  IEEE Robotics and Automation Letters (RA-L), 2023, and IEEE International Conference on Robotics and Automation (ICRA), 2024.
  [PDF] [Code]
  

Other Contributions

• Pytorch Version for OverlapNet [Code]
  
• Multiple Lidar Calibration [Code]
  
• Active SLAM Based on Information Theory [Code]
  
• Occupancy Grid Mapping [Code]
  
• Stairs Recognition [Code]
  

Datasets

• Haomo Dataset
  The dataset was collected by a mobile robot built by HAOMO.AI Technology company equipped with a HESAI PandarXT 32-beam LiDAR sensor in urban environments of Beijing.
  [Description]

• Cues-Poses Dataset
  A toy dataset about mapping multiple cues to mutual poses of robots.
  [Description]

• Cam4DOcc
  A Benchmark for Camera-Only 4D Occupancy Forecasting.
  [Description]

• CABH Benchmark
  Multiple egocentric videos capturing human hands performing simple object manipulation tasks.​​
  [Description]

Patents

• [China Utility Model] Huilong Yu, Ziang Tian, Junyi Ma, Haotian Dong, Junqiang Xi, and Guangming Xiong. A multifunctional unmanned platform for subterranean space. ZL202123083457.8
• [China Appearance Design] Huilong Yu, Ziang Tian, Junyi Ma, Haotian Dong, Junqiang Xi, and Guangming Xiong. A multifunctional unmanned caterpillar for subterranean space. ZL202130813635.4
• [China Invention Publication] Guangming Xiong, Junyi Ma, Jingyi Xu, and Jiarui Song. A reliability analysis-based multi-robot cooperative localization and mapping method. ZL202110318362.5

Awards

• Outstanding Master’s Thesis, Beijing Institute of Technology, 2023.
• National Scholarship for Graduate Students, Ministry of Education of China, 2022.
• National Scholarship for Undergraduate Students, Ministry of Education of China, 2019.
• Outstanding Master’s Graduates in Beijing, 2023.
• Outstanding Bachelor’s Graduates in Beijing, 2020.
• Best Paper Award at IEEE International Conference on Unmanned Systems (ICUS), 2021.
• Outstanding Paper Presented at the Autonomous Robotic Technology Seminar (ARTS), 2023

Services

• Reviewer of TRO, TMM, RA-L, TASE, TCSVT, ICRA, IROS.
• Student Executive Committee Member of Autonomous Robotic Technology Seminar (ARTS).

Powered by Jekyll and Minimal Light theme.