Songpengcheng. Xia
Affiliations. Ph.D. Candidate @ Shanghai Jiao Tong University.
I am Xia(夏宋鹏程- in Chinese, Pronounce), currently pursuing a Ph.D. degree at Shanghai Jiao Tong University (Ph.D. National Scholarship), focusing on the research of wearable-based human motion reconstruction and human activity recognition. My doctoral journey is under the guidance of Prof.Ling Pei, and meanwhile, I worked closely with Dr.Lei Chu. Previously, I obtained the Bachelor’s degree in Navigation Engineering at Wuhan University (National Scholarship,Outstanding graduates, Top 1% Award of Wuhan University).
Most recently, I am currently deeply engaged in three areas of interest: Motion Capture from Sparse Observations (utilizing IMUs/VR/Radar technologies), Weakly-supervised Activity Segmentation, and Motion Analysis for Health purposes. If you share a passion for these fields or have expertise in related areas, I would love to connect. Please feel free to reach out to me via email at songpengchengxia@sjtu.edu.cn. Looking forward to potential collaboration opportunities. I am also actively seeking suitable postdoctoral positions.
news
Mar 19, 2024 | Our work has been accepted by IEEE Transactions on Mobile Computing (TMC). ! |
---|---|
Feb 27, 2024 | Our work has been accepted by the CVPR 2024 (Co-First Author). ! |
Dec 14, 2023 | Two works have been accepted by the ICASSP 2024. ! |
Jul 14, 2023 | Our work has been accepted by the ICCV(International Conference on Computer Vision). ! |
Oct 01, 2022 | I honored the PhD. National Scholarship (2022). ! |
selected publications
- Timestamp-supervised Wearable-based Activity Segmentation and Recognition with Contrastive Learning and Order-Preserving Optimal TransportIEEE Transactions on Mobile Computing (TMC), 2024
- Dynamic Inertial Poser (DynaIP): Part-Based Motion Dynamics Learning for Enhanced Human Pose Estimation with Sparse Inertial SensorsIEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR), 2024
- MMBaT: A Multi-task Framework for mmWave-based Human Body Reconstruction and Translation Prediction2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
- Nerf-loam: Neural implicit representation for large-scale incremental lidar odometry and mappingIn iccv , 2023
- A Boundary Consistency-Aware Multitask Learning Framework for Joint Activity Segmentation and Recognition With Wearable SensorsIEEE Transactions on Industrial Informatics (TII), 2022
- Learning Disentangled Representation for Mixed-Reality Human Activity Recognition with a Single IMU SensorIEEE Transactions on Instrumentation and Measurement (TIM), 2021
- MARS: Mixed Virtual and Real Wearable Sensors for Human Activity Recognition with Multi-Domain Deep Learning ModelIEEE Internet of Things Journal, 2021