IMPROVING VIDEO CLASSIFICATION ACCURACY USING CLOUD
Loading...
Files
Date
2019-06-25T10:14:18Z
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The focus of this project is on the frame level features. One of the promising algorithms that can be used for this purpose is Deep Bag of Frame pooling (DBoF). Deep bag of frame model is a convolutional neural network (CNN). The main idea is to design two layers in the convolutional part. The approach enjoys the computational benefits of CNN, while at the same time the weights on the up-projection layer can still provide a strong representation of input features on frame level. The classification is performed at the final layer of the CNN. We will use the Youtube-8M dataset for experimentation. The Youtube-8M dataset is the largest publicly available multi-label video classification dataset, with approximately 8 Million videos annotated with 3862 classes of labels. The videos within the dataset averages 3.01 labels per video, where the number of labels per video ranges from 1 to 23. As this dataset covers over 500,000 hours of video, 2.6 billion audio and visual features have been extracted and pre-processed in advance by the Google Research Team as it would be infeasible for research teams to train hundreds of Terabytes worth of video for their mode.
Description
Keywords
1NH15IS081