Gaze region estimation algorithm without calibration available in a tablet
| Project/Area Number |
17K04966
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Special needs education
|
|---|
| Research Institution | National Institute of Technology, Kumamoto College |
|---|
Principal Investigator |
HAKATA Tetsuya 熊本高等専門学校, 電子情報システム工学系CIグループ, 教授 (60237899)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
柴里 弘毅 熊本高等専門学校, 電子情報システム工学系CIグループ, 教授 (60259968)
加藤 達也 熊本高等専門学校, 電子情報システム工学系CIグループ, 助教 (10707970)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2018: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2017: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
|
|---|
| Keywords | 重度重複障害 / 人間福祉工学 / 社会福祉関係 / アシスティブテクノロジー / 特別支援教育 |
|---|
| Outline of Final Research Achievements |
This study proposes gaze region estimation method that does not require an expensive device nor initial setting called calibration. Objective of this study is not only implementation of the application but also enhancement of utilization of the application by teachers in special needs education school who are not good at IT or computers, hence concept of this application is “simple, user friendly but effective”. In order for challenged people to show their intentions, it is considered that detecting gaze with high precision is not necessary and it is sufficient to detect gaze region concisely. Therefore, a typical webcam is adopted to implement for the algorithm. Generally, smartphone or tablet computer have the built-in camera and the proposed method is available for these devices. To figure out of applicability of the proposed algorithm, some trial experiments were conducted and the accuracy and usability is evaluated.
|
| Academic Significance and Societal Importance of the Research Achievements |
視線領域検出に機械学習を用いることで，キャリブレーションを不要な仕組みを構築した．高額で特殊なセンサ機器を使用することなくタブレット端末で児童生徒の意思表示の支援が可能になり有用である．例えば，食事の挨拶の場面で使用する際，児童生徒の画面注視による意思表示時に「いただきます」の音声が再生されれば，周囲の児童生徒・教諭らに自立的な動作を行えたことが自然に伝わり，その場の空間を共有することができる．日常生活における児童生徒の自発的な意思表出を周囲が理解し，達成感や自己肯定感が得られることから，自立活動の拡大に繋がることが期待される．
|
Report
(4 results)
Research Products
(5 results)
