• KSII Transactions on Internet and Information Systems
    Monthly Online Journal (eISSN: 1976-7277)

Spatio-Temporal Residual Networks for Slide Transition Detection in Lecture Videos

Vol. 13, No. 8, August 30, 2019
10.3837/tiis.2019.08.011, Download Paper (Free):

Abstract

In this paper, we present an approach for detecting slide transitions in lecture videos by introducing the spatio-temporal residual networks. Given a lecture video which records the digital slides, the speaker, and the audience by multiple cameras, our goal is to find keyframes where slide content changes. Since temporal dependency among video frames is important for detecting slide changes, 3D Convolutional Networks has been regarded as an efficient approach to learn the spatio-temporal features in videos. However, 3D ConvNet will cost much training time and need lots of memory. Hence, we utilize ResNet to ease the training of network, which is easy to optimize. Consequently, we present a novel ConvNet architecture based on 3D ConvNet and ResNet for slide transition detection in lecture videos. Experimental results show that the proposed novel ConvNet architecture achieves the better accuracy than other slide progression detection approaches.


Statistics

Show / Hide Statistics

Statistics (Cumulative Counts from December 1st, 2015)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article

[IEEE Style]
Z. Liu, K. Li, L. Shen, R. Ma and P. An, "Spatio-Temporal Residual Networks for Slide Transition Detection in Lecture Videos," KSII Transactions on Internet and Information Systems, vol. 13, no. 8, pp. 4026-4040, 2019. DOI: 10.3837/tiis.2019.08.011.

[ACM Style]
Zhijin Liu, Kai Li, Liquan Shen, Ran Ma, and Ping An. 2019. Spatio-Temporal Residual Networks for Slide Transition Detection in Lecture Videos. KSII Transactions on Internet and Information Systems, 13, 8, (2019), 4026-4040. DOI: 10.3837/tiis.2019.08.011.

[BibTeX Style]
@article{tiis:22180, title="Spatio-Temporal Residual Networks for Slide Transition Detection in Lecture Videos", author="Zhijin Liu and Kai Li and Liquan Shen and Ran Ma and Ping An and ", journal="KSII Transactions on Internet and Information Systems", DOI={10.3837/tiis.2019.08.011}, volume={13}, number={8}, year="2019", month={August}, pages={4026-4040}}