Multi-path Convolutional Neural Networks for Complex Image Classification

TL;DR

This paper introduces a multi-path convolutional neural network that processes different image versions separately, leading to improved classification performance on complex images compared to traditional single-path models.

Contribution

The paper proposes a novel multi-path CNN architecture that enhances feature learning for complex images, outperforming traditional models on challenging validation sets.

Findings

Achieved higher top-1 and top-5 accuracy on complex validation images.

Identified weaknesses of standard CNNs through detailed performance analysis.

Demonstrated the effectiveness of multi-path architecture over single-path models.

Abstract

Convolutional Neural Networks demonstrate high performance on ImageNet Large-Scale Visual Recognition Challenges contest. Nevertheless, the published results only show the overall performance for all image classes. There is no further analysis why certain images get worse results and how they could be improved. In this paper, we provide deep performance analysis based on different types of images and point out the weaknesses of convolutional neural networks through experiment. We design 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 image than the traditional single path model. We acquire better classification results on complex validation set on both top 1 and top 5 scores than the best ILSVRC 2013 classification model.