Comparative analysis of 3D convolutional and LSTM neural networks in the action recognition task by video data
Ruslan J. Portsev, Andrey V. Makarenko · Journal of Physics Conference Series · 2021
Abstract In the present paper a comparative analysis of two architectural neural network approaches (based on 3D convolutional and LSTM) in the recognition of actions on video is made. The problem was being solved on 10 behavior classes separated from the UCF50 dataset. The original neural network architectures were developed and pre-trained. It was found that the network based on 3D convolutions has better generalization ability and is more stable in the training.