Browse > Article
http://dx.doi.org/10.3837/tiis.2022.01.001

A new lightweight network based on MobileNetV3  

Zhao, Liquan (Key Laboratory of Modern Power System Simulation and Control and Renewable Energy Technology, Ministry of Education, Northeast Electric Power University)
Wang, Leilei (Key Laboratory of Modern Power System Simulation and Control and Renewable Energy Technology, Ministry of Education, Northeast Electric Power University)
Publication Information
KSII Transactions on Internet and Information Systems (TIIS) / v.16, no.1, 2022 , pp. 1-15 More about this Journal
Abstract
The MobileNetV3 is specially designed for mobile devices with limited memory and computing power. To reduce the network parameters and improve the network inference speed, a new lightweight network is proposed based on MobileNetV3. Firstly, to reduce the computation of residual blocks, a partial residual structure is designed by dividing the input feature maps into two parts. The designed partial residual structure is used to replace the residual block in MobileNetV3. Secondly, a dual-path feature extraction structure is designed to further reduce the computation of MobileNetV3. Different convolution kernel sizes are used in the two paths to extract feature maps with different sizes. Besides, a transition layer is also designed for fusing features to reduce the influence of the new structure on accuracy. The CIFAR-100 dataset and Image Net dataset are used to test the performance of the proposed partial residual structure. The ResNet based on the proposed partial residual structure has smaller parameters and FLOPs than the original ResNet. The performance of improved MobileNetV3 is tested on CIFAR-10, CIFAR-100 and ImageNet image classification task dataset. Comparing MobileNetV3, GhostNet and MobileNetV2, the improved MobileNetV3 has smaller parameters and FLOPs. Besides, the improved MobileNetV3 is also tested on CPU and Raspberry Pi. It is faster than other networks
Keywords
MobileNetV3; Real-time image classification; Lightweight network; Deep convolutional neural network; residual structure;
Citations & Related Records
연도 인용수 순위
  • Reference
1 A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang et al., "MobileNets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.
2 M. Sandler, A. Howard, M. Zhu, A. Zhmoginov and L. C. Chen, "MobileNetV2: Inverted residuals and linear bottlenecks," in Proc. of CVPR 2018, Salt Lake City, USA, pp. 4510-4520, June 2018.
3 J. Hu, L. Shen and G. Sun, "Squeeze-and-excitation networks," in Proc. of CVPR 2019, Salt Lake City, USA, pp. 7132-7141, June 2018.
4 F. N. Iandola, S. Han, M. W. Moskewicz, K. Ashrafet, W. J. Dally and K. Keutzer, "SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size," arXiv preprint arXiv: 1602.07360, 2016.
5 K. He, X. Zhang, S. Ren and J. Sun, "Deep residual learning for image recognition," in Proc. of CVPR 2016, Las Vegas, NV, USA, pp. 770-778, June 2016.
6 H. Yang, K. Zhan, B. Bao, Q. Yao, J. Zhang and M. Cheriet, "Automatic guarantee scheme for intent-driven network slicing and reconfiguration," J NETW COMPUT APPL, vol. 190, no. 15, 103163, September, 2021.   DOI
7 K. Wang, Z. Liu, Y. Lin, J. Lin and S. Han, "Haq: Hardware-aware automated quantization with mixed precision," in Proc. of CVPR 2019, Long Beach, CA, USA, pp. 8612-8620, June 2019.
8 K. Han, Y. Wang, Q. Tian, J. Guo, C. Xu et al., "GhostNet: More features from cheap operations," in Proc. of CVPR 2020, Seattle, WA, USA, pp. 1580-1589, June 2020.
9 Y. He, X. Zhang and J. Sun, "Channel pruning for accelerating very deep neural networks," in Proc. of ICCV 2017, Italy, pp. 1389-1397, October 2017.
10 M. Rastegari, V. Ordonez, J. Redmon and A. Farhadi, "Xnor-net: Imagenet classification using binary convolutional neural networks," in Proc. of ECCV 2016, Amsterdam, the Netherlands, pp. 525-542, October 2016.
11 H. Chen, Y. Wang, C. Xu, Z. Yang, C. Liu et al., "Data-free learning of student networks," in Proc. of ICCV 2019, Seoul, Korea, pp. 3514-3522, 2019.
12 W. Park, D. Kim, Y. Lu, M. Cho, "Relational knowledge distillation," in Proc. of CVPR 2019, Long Beach, CA, USA, pp. 3967-3976, June 2019.
13 M, Tan, Q, Le, "EfficientNet: Rethinking model scaling for convolutional neural networks," in Proc. of ICML 2019, PMLR, Long Beach, USA, pp. 6105-6114, June 2019.
14 A. Howard, M. Sandler, G. Chu, L. C. Chen, B. Chen et al., "Searching for MobileNetV3," in Proc. of ICCV 2019, Seoul, Korea, pp. 1314-1324, 2019.
15 S. Han, H. Mao and W. J. Dally, "Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding," arXiv preprint arXiv:1510.00149, 2015.
16 M. Tan, B. Chen, R. Pang, V. Vasudevan, M. Sandler et al., "MnasNet: Platform-aware neural architecture search for mobile," in Proc. of CVPR 2019, Long Beach, CA, USA, pp. 2820-2828, June 2019.
17 G. A. Khan, J. Hu, T. Li, B Diallo and H. Wang, "Multi-view data clustering via non-negative matrix factorization with manifold regularization," Int. J. Mach. Learn. & Cyber, pp.1-13, March, 2021.
18 B. Diallo, J. Hu, T. Li, G. A. Khan, A. S. Hussein, "Multi-view document clustering based on geometrical similarity measurement," Int. J. Mach. Learn. & Cyber, pp.1-13, March, 2021.
19 G.A. Khan, J. Hu, T. Li, et al., "Multi-view low rank sparse representation method for three-way clustering," Int. J. Mach. Learn. & Cyber, vol. 13, pp. 233-253, 2022.   DOI
20 N. Ma, X. Zhang, H. T. Zheng and J. Sun, "ShuffleNetV2: Practical guidelines for efficient cnn architecture design," in Proc. of ECCV 2018, Munich, Germany, pp. 122-138, September 2018.
21 F. Chollet, "Xception: Deep learning with depthwise separable convolutions," in Proc. of CVPR 2017, Honolulu, HI, USA, pp. 1251-1258, July 2017.
22 J. F. Qiu, Q. H. Wu, G. R Ding, Y. H. Xu and S. Feng, "A survey of machine learning for big data processing," EURASIP J Adv Signal Process, vol. 2016, no. 67, pp.1-16, May, 2016.   DOI
23 Z. X. Zhong, X. B. Ma, A Wei, "A Lightweight Solar Irradiance Prediction Model Based on Ground-based Cloud Images and Meteorological Data," Journal of Northeast Electric Power University, vol. 41, no. 1, pp. 24-30, June, 2021.
24 X. Zhang, X. Zhou, M. Lin and J. Sun, "ShuffleNet: An extremely efficient convolutional neural network for mobile devices," in Proc. of CVPR 2018, Salt Lake City, USA, pp. 6848-6856, June 2018.
25 Y. J. Wang, P. P. Cao, X. S. Wang, X. Y. Yan, "Research on Insulator Self Explosion Detection Method Based on Deep Learning," Journal of Northeast Electric Power University, vol. 40, no. 3, pp. 33-40, June, 2020.
26 B. Wu, A. Wan, X. Yue, P. Jin, S. Zhao et al., "Shift: A zero flop, zero parameter alternative to spatial convolutions," in Proc. of CVPR 2018, Salt Lake City, USA, pp. 9127-9135, June 2018.
27 C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens and Z. Wojna, "Rethinking the inception architecture for computer vision," in Proc. of CVPR 2016, Las Vegas, NV, USA, pp. 2818-2826, June 2016.
28 Z. Yang, Y. Wang, X. Chen, B. Shi, C. Xu et al., "Cars: Continuous evolution for efficient neural architecture search," in Proc. of CVPR 2020, Seattle, WA, USA, pp. 1829-1838, June 2020.
29 X. Gong, S. Chang, Y. Jiang and Z. Wang, "Autogan: Neural architecture search for generative adversarial networks," in Proc. of ICCV 2019, Seoul, Korea, pp. 3224-3234, October 2019.
30 Z. Yang, Y. Wang, C. Liu, B. Shi, C. Xu et al., "LegoNet: Efficient convolutional neural networks with lego filters," in Proc. of ICML 2019, PMLR, Long Beach, USA, pp. 7005-7014, June 2019.