Multi-path Convolutional Neural Networks for Image Classification

Abstract

Convolutional Neural Networks have demonstrated high performance in the ImageNet Large-Scale Visual Recognition Challenges contest. Nevertheless, the published results only show the overall performance for all image classes. These models do not include further analysis why certain image are misclassified and how they could be improved. In this thesis, we provide deep performance analysis based on different types of images and point out weaknesses of CNN (Convolutional Neural Networks) through experiment. We designed a novel multiple paths convolutional neural network, which feeds different versions of images into separated paths to learn more comprehensive features. This model has better presentation for images than the traditional single path model. We design different experiments to show that our model gets better classification results in top 1 and top 5 scores on the 100 categories dataset by comparing with the model, which was the best one for the image classification task in ILSVRC 2013.