| Project/Area Number |
17K13180
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Sports science
|
|---|
| Research Institution | University of Tsukuba |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2017-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2017: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
|
|---|
| Keywords | 3次元関節位置 / Deep Learning / スポーツ競技映像 / カメラキャリブレーション / ３次元関節位置 / 3次元関節位置検出 / 人工知能 / 情報工学 / 機械学習 / アルゴリズム / 確率論 |
|---|
| Outline of Final Research Achievements |
We devised an algorithm for detecting the joint position of athletes in sports video utilizing artificial intelligence (Deep Learning) as training support for all sports. We applied a 2D joint position detection algorithm using artificial intelligence and devised a 3D joint position detection algorithm. The proposed algorithm has been shown to work with high accuracy with little labor. We carried out quantitative evaluation experiments of camera calibration in the proposed 3D joint position detection algorithm, and confirmed that the proposed method had high accuracy. We also conducted comparative experiments of the previous method and the proposed method, and showed the effectiveness of the proposed method.
|
| Academic Significance and Societal Importance of the Research Achievements |
コンピュータビジョンの分野において，これまでの人工知能を活用した関節位置検出アルゴリズムでは２次元画像上の位置を推定していたが，本研究では２次元関節位置検出アルゴリズムを応用し，３次元関節位置の推定を実現した．また，２台のカメラの撮影と補完画像を撮影する少ない労力で３次元関節位置を推定可能とし，３次元関節位置の推定精度は高精度であることを確認した．体育学の分野において，考案した３次元関節位置推定法は全てのスポーツ競技に応用できる．従って，この３次元関節位置の推定データは，選手の移動量の算出や，フォーム関節データ解析など，選手の技術力向上に貢献できる．
|
