Coordinate attention for efficient mobile network design

Q Hou, D Zhou, J Feng - … of the IEEE/CVF conference on …, 2021 - openaccess.thecvf.com
Recent studies on mobile network design have demonstrated the remarkable effectiveness
of channel attention (eg, the Squeeze-and-Excitation attention) for lifting model performance …

Rethinking bottleneck structure for efficient mobile network design

D Zhou, Q Hou, Y Chen, J Feng, S Yan - … 23–28, 2020, Proceedings, Part III …, 2020 - Springer
The inverted residual block is dominating architecture design for mobile networks recently. It
changes the classic residual bottleneck by introducing two design rules: learning inverted …

Micronet: Improving image recognition with extremely low flops

Y Li, Y Chen, X Dai, D Chen, M Liu… - Proceedings of the …, 2021 - openaccess.thecvf.com
This paper aims at addressing the problem of substantial performance degradation at
extremely low computational cost (eg 5M FLOPs on ImageNet classification). We found that …

MPRNet: Multi-path residual network for lightweight image super resolution

A Mehri, PB Ardakani… - Proceedings of the IEEE …, 2021 - openaccess.thecvf.com
Lightweight super resolution networks have extremely importance for real-world
applications. In recent years several SR deep learning approaches with outstanding …

Refconv: Re-parameterized refocusing convolution for powerful convnets

Z Cai, X Ding, Q Shen, X Cao - arXiv preprint arXiv:2310.10563, 2023 - arxiv.org
We propose Re-parameterized Refocusing Convolution (RefConv) as a replacement for
regular convolutional layers, which is a plug-and-play module to improve the performance …

Sinet: Extreme lightweight portrait segmentation networks with spatial squeeze module and information blocking decoder

H Park, L Sjosund, YJ Yoo, N Monet… - Proceedings of the …, 2020 - openaccess.thecvf.com
Designing a lightweight and robust portrait segmentation algorithm is an important task for a
wide range of face applications. However, the problem has been considered as a subset of …

Resolution switchable networks for runtime efficient image recognition

Y Wang, F Sun, D Li, A Yao - … Conference, Glasgow, UK, August 23–28 …, 2020 - Springer
We propose a general method to train a single convolutional neural network which is
capable of switching image resolutions at inference. Thus the running speed can be …

Falconnet: factorization for the light-weight convnets

Z Cai, Q Shen - International Conference on Neural Information …, 2023 - Springer
Designing light-weight CNN models with little parameters and Flops is a prominent research
concern. However, three significant issues persist in the current light-weight CNNs: i) the …

A lightweight deep neural network with higher accuracy

L Zhao, L Wang, Y Jia, Y Cui - Plos one, 2022 - journals.plos.org
To improve accuracy of the MobileNet network, a new lightweight deep neural network is
designed based on the MobileNetV2 network. Firstly, it modifies the network depth of …

A novel method for trajectory recognition and working condition diagnosis of sucker rod pumping systems based on high-resolution representation learning

Q Wang, K Zhang, H Zhao, H Zhang, L Zhang… - Journal of Petroleum …, 2022 - Elsevier
The pumping unit is the most important lifting equipment in the rod pumping system, which
transmits ground power to the downhole pump through its sucker rod. Trajectory …