Squeeze-and-Excitation Networks

TL;DR

Squeeze-and-Excitation Networks introduce SE blocks that recalibrate channel-wise features in CNNs, significantly improving performance with minimal additional computation, and achieved top results in ImageNet 2017 classification.

Contribution

The paper proposes the SE block, a novel architectural unit for adaptively recalibrating channel features in CNNs, enhancing their representational power and performance.

Findings

SE blocks improve CNN accuracy across multiple datasets.

SE-enhanced models outperform state-of-the-art CNNs with slight computational overhead.

SENet achieved first place in ILSVRC 2017 classification challenge.

Abstract

The central building block of convolutional neural networks (CNNs) is the convolution operator, which enables networks to construct informative features by fusing both spatial and channel-wise information within local receptive fields at each layer. A broad range of prior research has investigated the spatial component of this relationship, seeking to strengthen the representational power of a CNN by enhancing the quality of spatial encodings throughout its feature hierarchy. In this work, we focus instead on the channel relationship and propose a novel architectural unit, which we term the "Squeeze-and-Excitation" (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels. We show that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets. We further demonstrate that SE blocks bring significant improvements in performance for existing state-of-the-art CNNs at slight additional computational cost. Squeeze-and-Excitation Networks formed the foundation of our ILSVRC 2017 classification submission which won first place and reduced the top-5 error to 2.251%, surpassing the winning entry of 2016 by a relative improvement of ~25%. Models and code are available at https://github.com/hujie-frank/SENet.