Categorization of actions in soccer videos using a combination of transfer learning and Gated Recurrent Unit
Anik Sen, Kaushik Deb
Abstract
Extraction of knowledge from soccer videos has enormous applications like context-based advertisement, content-based video retrieval, match summarization, and highlight extraction. Overlapping soccer actions and uncontrolled video capturing conditions make it challenging to detect action accurately. For overcoming these problems, Convolutional Neural Network and Recurrent Neural Network are used jointly to classify different lengths of soccer actions. Initially, transfer learning from pre-trained VGG network extracts characteristic spatial features. Afterwards, Gated Recurrent Unit deals with temporal dependency and solves the vanishing gradient problem. Finally, softmax layer assigns decimal probabilities to each class. Experimental results demystify the significance of the proposed architecture.