AFINet: Attentive Feature Integration Networks for Image Classification
This work addresses the need for more efficient and effective feature transfer in CNNs, particularly for image classification tasks, but it is incremental as it builds upon existing architectures like ResNets.
The paper tackles the problem of improving feature integration in convolutional neural networks for image classification by introducing Attentive Feature Integration (AFI) modules, resulting in AFI-ResNet-152 achieving a 1.24% relative improvement on ImageNet while reducing FLOPs by about 10% and parameters by about 9.2% compared to ResNet-152.
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.