Name
Abstract | ||
|---|---|---|
This paper proposes a method of using multiple depth sensors vulnerable to joint occlusion and rotation to capture the dynamic motions of two users. The proposed method captures a series of multi-user motions by tracking user IDs, sorting out ID specific joints, and incorporating the joint movement data based on weighting. The ellipses accelerate the process of sorting out the depth data per user. Then, the sorted depth data add to the accuracy of joint positions for the adjustment of lower-limb joints. The proposed method enables an accurate restoration of 3D poses even in dynamic motions, and is applicable to experiential and training programs.
|
Year | DOI | Venue |
|---|---|---|
2018 | 10.1145/3301326.3301365 | Proceedings of the 2018 VII International Conference on Network, Communication and Computing |
Keywords | Field | DocType |
3D pose, ID tracking, Joints Estimation, Multi-Kinects | Computer vision,Weighting,Computer science,Sorting,Artificial intelligence,Ellipse | Conference |
ISBN | Citations | PageRank |
978-1-4503-6553-6 | 0 | 0.34 |
References | Authors | |
3 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Seongmin Baek | 1 | 0 | 1.69 |
| Myunggyu Kim | 2 | 6 | 5.62 |