Engagement Detection using Video-based Estimation of Head Movement
| Autor: | Sudhir S. Mane, Anil Surve |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Facial expression
Local binary patterns Computer science business.industry Feature vector 010401 analytical chemistry 02 engineering and technology Movement activity 01 natural sciences 0104 chemical sciences 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Computer vision Artificial intelligence K nearest neighbour business Classifier (UML) Video based |
| Zdroj: | 2018 3rd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT). |
| DOI: | 10.1109/rteict42901.2018.9012323 |
| Popis: | We explore how computer vision methods can be utilized to identify engagement where person in video will be doing writing activity similar to the student doing writing activity in front of monitor. Person will be provided engagement annotations concurrently during the writing activity. We utilized one of computer vision method to extract features from videos that is local binary patterns in three orthogonal planes (LBP-TOP). These highlights were utilized as a part of managed learning for detection. KNN (K nearest neighbour) classifier was used to find classifying accuracy. Conventional camera was used to capture persons head movement activity to further create feature vector to produce best results. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|