
Research Article
Pose+Context: A Model for Recognizing Non-verbal Teaching Behavior of Normal College Student
@INPROCEEDINGS{10.1007/978-3-031-33614-0_17, author={Yonghe Zhang and Bing Li and Xiaoming Cao}, title={Pose+Context: A Model for Recognizing Non-verbal Teaching Behavior of Normal College Student}, proceedings={Big Data Technologies and Applications. 11th and 12th EAI International Conference, BDTA 2021 and BDTA 2022, Virtual Event, December 2021 and 2022, Proceedings}, proceedings_a={BDTA}, year={2023}, month={5}, keywords={deep learning computer vision pose estimation classroom behavior analysis}, doi={10.1007/978-3-031-33614-0_17} }
- Yonghe Zhang
Bing Li
Xiaoming Cao
Year: 2023
Pose+Context: A Model for Recognizing Non-verbal Teaching Behavior of Normal College Student
BDTA
Springer
DOI: 10.1007/978-3-031-33614-0_17
Abstract
Normal college students practice their teaching skills to prepare for their teacher career. However, due to the complexity of teaching skills and the overburdening of instructors, their needs for instructional feedback are often not met. To server automatic feedbacks to normal college students, this paper proposes a deep learning model, called Pose+Contex, to recognize three types of non-verbal teaching behaviors (NVTB). The model includes three parts: (1) context detection, (2) pose estimation, and (3) behavior recognition. Our model is featured by the context detection component, and experiments show that it performs better than a similar model without this component.
Copyright © 2021–2025 ICST