GhostNetV2: Enhance Cheap Operation with Long-Range Attention

By Yehui Tang, Kai Han, Jianyuan Guo, Chang Xu, Chaoting Xu, Yunhe Wang. Neural Information Processing Systems, 2022

579 citations recorded by the source corpus.

Light-weight convolutional neural networks (CNNs) are specially designed for applications on mobile devices with faster inference speed. The convolutional operation can only capture local information in a window region, which prevents performance from being further improved. Introducing self-attention into convolution can capture global information well, but it will largely encumber the actual speed. In this paper, we propose a hardware-friendly attention mechanism (dubbed DFC attention) and then present a new GhostNetV2 architecture for mobile applications. The proposed DFC attention is constructed based on fully-connected layers, which can not only execute fast on common hardware but also capture the dependence between long-range pixels. We further revisit the expressiveness bottleneck in previous GhostNet and propose to enhance expanded features produced by cheap operations with DFC attention, so that a GhostNetV2 block can aggregate local and long-range information simultaneously. Extensive experiments demonstrate the superiority of GhostNetV2 over existing architectures. For example, it achieves 75.3% top-1 accuracy on ImageNet with 167M FLOPs, significantly suppressing GhostNetV1 (74.5%) with a similar computational cost. The source code will be available at https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/ghostnetv2_pytorch and https://gitee.com/mindspore/models/tree/master/research/cv/ghostnetv2.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Neurology

View Paper: https://www.semanticscholar.org/paper/3e420beb7f5d1bc370470b31908dd766ba35eedd

How researchers see this paper

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

In Artificial Intelligence

1. Xinhao Cai
2. Yazhou Yao
3. Qiuxia Lai
4. Wenguan Wang
5. Yuwei Wang
6. Zeren Sun

In Computer Vision and Pattern Recognition

1. Xinhao Cai
2. Yazhou Yao
3. Qiuxia Lai
4. Wenguan Wang
5. Yuwei Wang
6. Zeren Sun

Citations in the corpus

• MobileNetV4 - Universal Models for the Mobile Ecosystem (2024, unknown): “GhostNetV2 [44] uses FC layers to capture long-range dependencies.”
• Poly Kernel Inception Network for Remote Sensing Detection (2024, methodology): “To capture long-range contextual information, inspired by [32, 58], we further integrated a Context Anchor Attention (CAA) module into the PKI Block.”