International Journal of New Developments in Education, 2025, 7(1); doi: 10.25236/IJNDE.2025.070116.
Chao Pan1, Yifeng Wang1, Haihong Zheng1, Fei Wang2
1School of Computer Science and Technology, Xidian University, Xi’an, China
2School of Mechano-Electronic Engineering, Xidian University, Xi’an, China
The development of information technology has made ‘Artificial Intelligence (AI)+ Education’ widespread, generating a massive amount of classroom teaching videos and courseware text data. Knowledge-based multimodal teaching video retrieval and localization refers to the process of retrieving videos corresponding to specific knowledge points from teaching videos and locating fine-grained video moments based on query text related to those knowledge points. This approach can enhance teaching efficiency and support auxiliary teaching. However, current video retrieval and localization technologies are based on single-scene datasets and lack video-text datasets specifically for educational contexts. This research aims to develop knowledge-based teaching video retrieval and localization technologies. It involves constructing a video-text offline teaching dataset tailored for educational scenes, used for video retrieval and localization in educational contexts. A pre-trained model for the educational scene is employed to process data and extract features. By utilizing video-text retrieval and moment localization techniques, this study builds a smart education video retrieval and localization system. The system retrieves corresponding videos and locates the relevant knowledge point moments in the teaching videos quickly and accurately based on the input knowledge point query text. This contributes to both ‘convenient teaching for teachers’ and ‘efficient learning for students.’
AI+Education, Video Retrieval and Localization, Teaching Video
Chao Pan, Yifeng Wang, Haihong Zheng, Fei Wang. Knowledge-Based Teaching Video Retrieval and Localization Technology. International Journal of New Developments in Education (2025), Vol. 7, Issue 1: 110-115. https://doi.org/10.25236/IJNDE.2025.070116.
[1] Chen L, Chen P, Lin Z. Artificial intelligence in education: A review[J]. Ieee Access, 2020, 8: 75264-75278.
[2] Cantú-Ortiz F J, Galeano Sánchez N, Garrido L, et al. An artificial intelligence educational strategy for the digital transformation[J]. International Journal on Interactive Design and Manufacturing (IJIDeM), 2020, 14: 1195-1209.
[3] George B, Wooden O. Managing the strategic transformation of higher education through artificial intelligence[J]. Administrative Sciences, 2023, 13(9): 196.
[4] Chun S, Oh S J, De Rezende R S, et al. Probabilistic embeddings for cross-modal retrieval[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 8415-8424.
[5] Wei J, Yang Y, Xu X, et al. Universal weighting metric learning for cross-modal retrieval [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 44(10): 6534-6545.
[6] Lu H, Fei N, Huo Y, et al. COTS: Collaborative two-stream vision-language pre-training model for cross-modal retrieval [C] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 15692-15701.