Squeeze aggregated excitation network
This work addresses the challenge of enhancing representational power in vision tasks, but it appears incremental as it builds on existing squeeze-and-excitation methods.
The authors tackled the problem of improving channel-wise representation in convolutional neural networks by proposing SaEnet, which integrates global representations within channels using aggregated excitation and a multibranch linear layer. The result is that the model achieves performance comparable to or better than state-of-the-art architectures on ImageNet and CIFAR100 datasets.
Convolutional neural networks have spatial representations which read patterns in the vision tasks. Squeeze and excitation links the channel wise representations by explicitly modeling on channel level. Multi layer perceptrons learn global representations and in most of the models it is used often at the end after all convolutional layers to gather all the information learned before classification. We propose a method of inducing the global representations within channels to have better performance of the model. We propose SaEnet, Squeeze aggregated excitation network, for learning global channelwise representation in between layers. The proposed module takes advantage of passing important information after squeeze by having aggregated excitation before regaining its shape. We also introduce a new idea of having a multibranch linear(dense) layer in the network. This learns global representations from the condensed information which enhances the representational power of the network. The proposed module have undergone extensive experiments by using Imagenet and CIFAR100 datasets and compared with closely related architectures. The analyzes results that proposed models outputs are comparable and in some cases better than existing state of the art architectures.