| Paper Abstract and Keywords |
| Presentation |
2022-12-07 13:35
Skeleton Estimation for Swing based on Deep Model by Introducing Articulation Constraints into Training
Atsuki Sakata, Shogo Kihira, Nobutaka Shimada (Ritsumeikan Univ.), Yuki Nagano, Masahiki Ueda (SRI) |
| Abstract |
(in Japanese) |
(See Japanese page) |
| (in English) |
For the purpose of developing the automatic diagnosis system of players' golf swing, we propose a method for estimating 3-D coordinates of human skeleton during swing using a deep learning model trained with mass CG images generated from mo-cap time-series data. In order to introduce the geometric constraint that the skeletal length is invariant for the same person's swing sequence into the training of the deep model, each single training sample pair randomly taken from the same person's swing sequence is used as training data, and the difference in skeletal lengths between them is additionally considered as loss function. By using CG images augmented by motion embedding in CG for 31 players' sequences, a deep neural model was trained and achieved higher consistency of skeletal lengths than unconstrained learning. |
| Keyword |
(in Japanese) |
(See Japanese page) |
| (in English) |
Golf swing measurement / Deep learning / Depth image, / Constraint satisfaction / / / / |
| Reference Info. |
ITE Tech. Rep., vol. 46, no. 39, SIP2022-10, pp. 23-26, Dec. 2022. |
| Paper # |
SIP2022-10 |
| Date of Issue |
2022-11-30 (ME, SIP) |
| ISSN |
Print edition: ISSN 1342-6893 Online edition: ISSN 2424-1970 |
| Download PDF |
|
| Conference Information |
| Committee |
ME SIP TOKAI |
| Conference Date |
2022-12-07 - 2022-12-07 |
| Place (in Japanese) |
(See Japanese page) |
| Place (in English) |
|
| Topics (in Japanese) |
(See Japanese page) |
| Topics (in English) |
Image Processing, Sports Information Processing, and other general |
| Paper Information |
| Registration To |
SIP |
| Conference Code |
2022-12-ME-SIP-TOKAI |
| Language |
Japanese |
| Title (in Japanese) |
(See Japanese page) |
| Sub Title (in Japanese) |
(See Japanese page) |
| Title (in English) |
Skeleton Estimation for Swing based on Deep Model by Introducing Articulation Constraints into Training |
| Sub Title (in English) |
|
| Keyword(1) |
Golf swing measurement |
| Keyword(2) |
Deep learning |
| Keyword(3) |
Depth image, |
| Keyword(4) |
Constraint satisfaction |
| Keyword(5) |
|
| Keyword(6) |
|
| Keyword(7) |
|
| Keyword(8) |
|
| 1st Author's Name |
Atsuki Sakata |
| 1st Author's Affiliation |
Ritsumeikan University (Ritsumeikan Univ.) |
| 2nd Author's Name |
Shogo Kihira |
| 2nd Author's Affiliation |
Ritsumeikan University (Ritsumeikan Univ.) |
| 3rd Author's Name |
Nobutaka Shimada |
| 3rd Author's Affiliation |
Ritsumeikan University (Ritsumeikan Univ.) |
| 4th Author's Name |
Yuki Nagano |
| 4th Author's Affiliation |
Sumitomo Rubber Industries (SRI) |
| 5th Author's Name |
Masahiki Ueda |
| 5th Author's Affiliation |
Sumitomo Rubber Industries (SRI) |
| 6th Author's Name |
|
| 6th Author's Affiliation |
() |
| 7th Author's Name |
|
| 7th Author's Affiliation |
() |
| 8th Author's Name |
|
| 8th Author's Affiliation |
() |
| 9th Author's Name |
|
| 9th Author's Affiliation |
() |
| 10th Author's Name |
|
| 10th Author's Affiliation |
() |
| 11th Author's Name |
|
| 11th Author's Affiliation |
() |
| 12th Author's Name |
|
| 12th Author's Affiliation |
() |
| 13th Author's Name |
|
| 13th Author's Affiliation |
() |
| 14th Author's Name |
|
| 14th Author's Affiliation |
() |
| 15th Author's Name |
|
| 15th Author's Affiliation |
() |
| 16th Author's Name |
|
| 16th Author's Affiliation |
() |
| 17th Author's Name |
|
| 17th Author's Affiliation |
() |
| 18th Author's Name |
|
| 18th Author's Affiliation |
() |
| 19th Author's Name |
|
| 19th Author's Affiliation |
() |
| 20th Author's Name |
|
| 20th Author's Affiliation |
() |
| Speaker |
Author-1 |
| Date Time |
2022-12-07 13:35:00 |
| Presentation Time |
25 minutes |
| Registration for |
SIP |
| Paper # |
ME2022-91, SIP2022-10 |
| Volume (vol) |
vol.46 |
| Number (no) |
no.39 |
| Page |
pp.23-26 |
| #Pages |
4 |
| Date of Issue |
2022-11-30 (ME, SIP) |
|