AFINet: Attentive Feature Integration Networks for Image Classification

TL;DR

This paper introduces AFI-Nets, a new CNN architecture with attentive feature integration modules that explicitly model feature correlations, improving accuracy and efficiency in image classification tasks.

Contribution

The paper proposes AFI modules and AFI-Nets, a novel architecture that enhances feature transfer and correlation modeling in CNNs, leading to better performance and reduced computational costs.

Findings

AFI-ResNet-152 improves accuracy by 1.24% on ImageNet.

AFI-Nets reduce FLOPs by about 10%.

AFI-ResNet-152 decreases parameters by about 9.2%.

Abstract

Convolutional Neural Networks (CNNs) have achieved tremendous success in a number of learning tasks including image classification. Recent advanced models in CNNs, such as ResNets, mainly focus on the skip connection to avoid gradient vanishing. DenseNet designs suggest creating additional bypasses to transfer features as an alternative strategy in network design. In this paper, we design Attentive Feature Integration (AFI) modules, which are widely applicable to most recent network architectures, leading to new architectures named AFI-Nets. AFI-Nets explicitly model the correlations among different levels of features and selectively transfer features with a little overhead.AFI-ResNet-152 obtains a 1.24% relative improvement on the ImageNet dataset while decreases the FLOPs by about 10% and the number of parameters by about 9.2% compared to ResNet-152.