Global-and-local attention networks for visual recognition

By Drew Linsley, Dan Scheibler, Sven Eberhardt, Thomas Serre. 2018

33 citations recorded by the source corpus.

State-of-the-art deep convolutional networks (DCNs) such as squeeze-and- excitation (SE) residual networks implement a form of attention, also known as contextual guidance, which is derived from global image features. Here, we explore a complementary form of attention, known as visual saliency, which is derived from local image features. We extend the SE module with a novel global-and-local attention (GALA) module which combines both forms of attention -- resulting in state-of-the-art accuracy on ILSVRC. We further describe ClickMe.ai, a large-scale online experiment designed for human participants to identify diagnostic image regions to co-train a GALA network. Adding humans-in-the-loop is shown to significantly improve network accuracy, while also yielding visual features that are more interpretable and more similar to those used by human observers.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

1. Jonathon Shlens
2. Quoc V. Le
3. Ashish Vaswani

In Computer Vision and Pattern Recognition

1. Quoc V. Le
2. Luc Van Gool
3. Alan L. Yuille
4. Dacheng Tao
5. Trevor Darrell
6. Jiashi Feng
7. Xiaogang Wang
8. Ming-Hsuan Yang

In Electrical and Electronic Engineering

1. Jonathon Shlens
2. Barret Zoph
3. Quoc V. Le
4. Ashish Vaswani
5. Irwan Bello

Citations in the corpus

• Attention Augmented Convolutional Networks (2019, background): “All attention augmented networks use κ = 2 υ =0 Table 2 benchmarks Attention Augmentation against channel and spatial attention mechanisms BAM [31], CBAM [46] and GALA [28] with channel reduction ratio σ = 16 on the ResNet50 architecture.”