TY - GEN
T1 - Human motion data refinement unitizing structural sparsity and spatial-temporal information
AU - Wang, Zhao
AU - Liu, Shuang
AU - Qian, Rongqiang
AU - Jiang, Tao
AU - Yang, Xiaosong
AU - Zhang, Jian J.
N1 - Funding Information:
The authors would like to thanks the supported from the grant of the Sino-UK project, National Natural Science Foundation of China (Grant No.61572431 and Grant No.51475394), AniNex project (FP7-IRSES-612627), Dr.Inventor project(FP7-ICT-2013.8.1 611383) and Santander PGR Mobility Award
Publisher Copyright:
© 2016 IEEE.
PY - 2016/7/2
Y1 - 2016/7/2
N2 - Human motion capture techniques (MOCAP) are widely applied in many areas such as computer vision, computer animation, digital effect and virtual reality. Even with professional MOCAP system, the acquired motion data still always contains noise and outliers, which highlights the need for the essential motion refinement methods. In recent years, many approaches for motion refinement have been developed, including signal processing based methods, sparse coding based methods and low-rank matrix completion based methods. However, motion refinement is still a challenging task due to the complexity and diversity of human motion. In this paper, we propose a data-driven-based human motion refinement approach by exploiting the structural sparsity and spatio-temporal information embedded in motion data. First of all, a human partial model is applied to replace the entire pose model for a better feature representation to exploit the abundant local body posture. Then, a dictionary learning which is for special task of motion refinement is designed and applied in parallel. Meanwhile, the objective function is derived by taking the statistical and locality property of motion data into account. Compared with several state-of-art motion refine methods, the experimental result demonstrates that our approach outperforms the competitors.
AB - Human motion capture techniques (MOCAP) are widely applied in many areas such as computer vision, computer animation, digital effect and virtual reality. Even with professional MOCAP system, the acquired motion data still always contains noise and outliers, which highlights the need for the essential motion refinement methods. In recent years, many approaches for motion refinement have been developed, including signal processing based methods, sparse coding based methods and low-rank matrix completion based methods. However, motion refinement is still a challenging task due to the complexity and diversity of human motion. In this paper, we propose a data-driven-based human motion refinement approach by exploiting the structural sparsity and spatio-temporal information embedded in motion data. First of all, a human partial model is applied to replace the entire pose model for a better feature representation to exploit the abundant local body posture. Then, a dictionary learning which is for special task of motion refinement is designed and applied in parallel. Meanwhile, the objective function is derived by taking the statistical and locality property of motion data into account. Compared with several state-of-art motion refine methods, the experimental result demonstrates that our approach outperforms the competitors.
KW - Motion Capture Data
KW - Motion Refinement
UR - http://www.scopus.com/inward/record.url?scp=85016243832&partnerID=8YFLogxK
U2 - 10.1109/ICSP.2016.7877975
DO - 10.1109/ICSP.2016.7877975
M3 - Conference Proceeding
AN - SCOPUS:85016243832
T3 - International Conference on Signal Processing Proceedings, ICSP
SP - 975
EP - 982
BT - ICSP 2016 - 2016 IEEE 13th International Conference on Signal Processing, Proceedings
A2 - Baozong, Yuan
A2 - Qiuqi, Ruan
A2 - Yao, Zhao
A2 - Gaoyun, An
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 13th IEEE International Conference on Signal Processing, ICSP 2016
Y2 - 6 November 2016 through 10 November 2016
ER -