The Community for Technology Leaders
2016 International Conference on Frontiers of Information Technology (FIT) (2016)
Islamabad, Pakistan
Dec. 19, 2016 to Dec. 21, 2016
ISBN: 978-1-5090-5300-1
pp: 247-251
ABSTRACT
Deep learning tools such as the convolutional neural network (CNN) are extensively used for image analysis and interpretation tasks but they become relatively expensive to use for a corresponding analysis in videos by requiring memory provision for the additional temporal information. Crowd video analysis is one of the subareas in video analysis that has recently gained notoriety. In this paper we have shown that a 2D CNN can be used to classify videos by using 3-channel image map input for each video computed using spatial and temporal information and this reduces space and time complexity over a classical 3D CNN usually used for video analysis. We test the model developed with the state-of-the-art method of [1] using their proposed dataset, and without any additional processing steps, improve upon their reported accuracy.
INDEX TERMS
Machine learning, Two dimensional displays, Neural networks, Testing, Stability analysis, Three-dimensional displays, Image analysis,crowd video analysis, crowd, video analytics, deep learning, cnn, convolutional neural network, classification
CITATION
Atika Burney, Tahir Q. Syed, "Crowd Video Classification Using Convolutional Neural Networks", 2016 International Conference on Frontiers of Information Technology (FIT), vol. 00, no. , pp. 247-251, 2016, doi:10.1109/FIT.2016.052
90 ms
(Ver 3.3 (11022016))