Cited By
View all- Kamath VRenuka A(2023)Deep learning based object detection for resource constrained devicesNeurocomputing10.1016/j.neucom.2023.02.006531:C(34-60)Online publication date: 28-Apr-2023
PokerNet: Expanding Features Cheaply via Depthwise Convolutions
Authors: 
Wei Tang, Yan Huang, Liang WangAuthors Info & Claims
Pages 432 - 442
Abstract
Pointwise convolution is usually utilized to expand or squeeze features in modern lightweight deep models. However, it takes up most of the overall computational cost (usually more than 90%). This paper proposes a novel Poker module to expand features by taking advantage of cheap depthwise convolution. As a result, the Poker module can greatly reduce the computational cost, and meanwhile generate a large number of effective features to guarantee the performance. The proposed module is standardized and can be employed wherever the feature expansion is needed. By varying the stride and the number of channels, different kinds of bottlenecks are designed to plug the proposed Poker module into the network. Thus, a lightweight model can be easily assembled. Experiments conducted on benchmarks reveal the effectiveness of our proposed Poker module. And our PokerNet models can reduce the computational cost by 7.1%–15.6%. PokerNet models achieve comparable or even higher recognition accuracy than previous state-of-the-art (SOTA) models on the ImageNet ILSVRC2012 classification dataset. Code is available at https://github.com/diaomin/pokernet.
References
[1]
Krizhevsky A, Sutskever I, and Hinton G E Imagenet classification with deep convolutional neural networks Proceedings of the 25th International Conference on Neural Information Processing Systems 2012 Lake Tahoe, USA NIPS 1097-1105
[2]
K. Simonyan, A. Zisserman. Very deep convolutional networks for large-scale image recognition. [Online], Available: https://arxiv.org/abs/1409.1556, 2015.
[3]
He K M, Zhang X Y, Ren S Q, and Sun J Deep residual learning for image recognition Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2016 Las Vegas, USA IEEE 770-778
[4]
Girshick R Fast R-CNN Proceedings of IEEE International Conference on Computer Vision 2015 Santiago, Chile IEEE 1440-1448
[5]
He K M, Gkioxari G, Dollár P, and Girshick R Mask R-CNN Proceedings of IEEE International Conference on Computer Vision 2017 Venice, Italy IEEE 2980-2988
[6]
Lin T Y, Dollár P, Girshick R, He K M, Hariharan B, and Belongie S Feature pyramid networks for object detection Proceedings of IEEE Conference On computer Vision and Pattern Recognition 2017 Honolulu, USA IEEE 936-944
[7]
Long J, Shelhamer E, and Darrell T Fully convolutional networks for semantic segmentation Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2015 Boston, USA IEEE 3431-3440
[8]
L. C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, A. L. Yuille. Semantic image segmentation with deep convolutional nets and fully connected CRFs. [Online], Available: https://arxiv.org/abs/1412.7062, 2016.
[9]
W. Y. Chen, X. Y. Gong, X. M. Liu, Q. Zhang, Y. Li, Z. Y. Wang. FasterSeg: Searching for faster real-time semantic segmentation. In Proceedings of the 4th International Conference on Learning Representations, OpenReview. net, Addis Ababa, Ethiopia, 2020.
[10]
H. Li, A. Kadav, I. Durdanovic, H. Samet, H. P. Graf. Pruning filters for efficient convNets. In Proceedings of the 5th International Conference on Learning Representations, OpenReview.net, Toulon, France, 2017.
[11]
He Y H, Zhang X Y, and Sun J Channel pruning for accelerating very deep neural networks Proceedings of IEEE International Conference on Computer Vision 2017 Venice, Italy IEEE 1398-1406
[12]
I. Hubara, M. Courbariaux, D. Soudry, R. El-Yaniv, Y. Bengio. Binarized neural networks. In Proceedings of the 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, pp. 4114–4122, 2016.
[13]
W. Tang, G. Hua, L. Wang. How to train a compact binary neural network with high accuracy? In Proceedings of the 31st AAAI Conference on Artificial Intelligence, San Francisco, USA, pp. 2625–2631, 2017.
[14]
A. G. Howard, M. L. Zhu, B. Chen, D. Kalenichenko, W. J. Wang, T. Weyand, M. Andreetto, H. Adam. Mobilenets: Efficient convolutional neural networks for mobile vision applications. [Online], Available: https://arxiv.org/abs/1704.14861, 2017.
[15]
Zhang X Y, Zhou X Y, Lin M X, and Sun J Shufflenet: An extremely efficient convolutional neural network for mobile devices Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2018 Salt Lake City, USA IEEE 6848-6856
[16]
G. Hinton, O. Vinyals, J. Dean. Distilling the knowledge in a neural network. [Online], Available: https://arxiv.org/abs/1503.02531, 2015.
[17]
You S, Xu C, Xu C, and Tao D C Learning from multiple teacher networks Proceedings of the 23rd ACM SIGK-DD International Conference on Knowledge Discovery and Data Mining 2017 Halifax, Canada ACM 1285-1294
[18]
Chen H T, Wang Y H, Xu C, Yang Z H, Liu C J, Shi B X, Xu C J, Xu C, and Tian Q Data-free learning of student networks Proceedings of IEEE/CVF International Conference on Computer Vision 2019 Seoul, Korea IEEE 3513-3521
[19]
Han K, Wang Y H, Tian Q, Guo J Y, Xu C J, and Xu C Ghostnet: More features from cheap operations Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2020 Seattle, USA IEEE 1577-1586
[20]
Sandler M, Howard A, Zhu M L, Zhmoginov A, and Chen L C MobileNetV2: Inverted residuals and linear bottle-necks Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2018 Salt Lake City, USA IEEE 4510-4520
[21]
Howard A, Sandler M, Chen B, Wang W J, Chen L C, Tan M X, Chu G, Vasudevan V, Zhu Y K, Pang R M, Adam H, and Le Q Searching for mobileNetV3 Proceedings of IEEE/CVF International Conference on Computer Vision 2019 Seoul, Korea IEEE 1314-1324
[22]
Ma N N, Zhang X Y, Zheng H T, and Sun J Shufflenet V2: Practical guidelines for efficient CNN architecture design Proceedings of the 15th European Conference on Computer Vision 2018 Munich, Germany Springer 122-138
[23]
M. Lin, Q. Chen, S. C. Yan. Network in network. [Online], Available: https://arxiv.org/abs/1312.4400, 2014.
[24]
F. N. Iandola, S. Han, M. W. Moskewicz, K. Ashraf, W. J. Dally, K. Keutzer. SqueezeNet: Alexnet-level accuracy with 50x fewer parameters and < 0.5 MB model size. [Online], Available: https://arxiv.org/abs/1602.07360, 2016.
[25]
Wu B C, Wan A, Yue X Y, Jin P, Zhao S C, Golmant N, Gholaminejad A, Gonzalez J, and Keutzer K Shift: A zero FLOP, zero parameter alternative to spatial convolutions Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2018 Salt Lake City, USA IEEE 9127-9135
[26]
H. Cai, L. G. Zhu, S. Han. ProxylessNAS: Direct neural architecture search on target task and hardware. In Proceedings of the 7th International Conference on Learning Representations, OpenReview.net, New Orleans, USA, 2019.
[27]
Tan M X, Chen B, Pang R M, Vasudevan V, Sandler M, Howard A, and Le Q V MnasNet: Platform-aware neural architecture search for mobile Proceedings IEEE/CVF Conference on Computer Vision and Pattern Recognition 2019 Long Beach, USA IEEE 2815-2823
[28]
Liu C X, Zoph B, Neumann M, Shlens J, Hua W, Li L J, Li F F, Yuille A, Huang J, and Murphy K Progressive neural architecture search Proceedings of the 15th European Conference on Computer Vision 2018 Munich, Germany Springer 19-35
[29]
B. Zoph, Q. V. Le. Neural architecture search with reinforcement learning. In Proceedings of the 5th International Conference on Learning Representations, OpenReview.net, Toulon, France, 2017.
[30]
B. Baker, O. Gupta, N. Naik, R. Raskar. Designing neural network architectures using reinforcement learning. In Proceedings of the 5th International Conference on Learning Representations, OpenReview.net, Toulon, France, 2017.
[31]
Zoph B, Vasudevan V, Shlens J, and Le Q V Learning transferable architectures for scalable image recognition Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2018 Salt Lake City, USA IEEE 8697-8710
[32]
H. Pham, M. Y. Guan, B. Zoph, Q. V. Le, J. Dean. Efficient neural architecture search via parameters sharing. In Proceedings of the 35th International Conference on Machine Learning, Stockholm, Sweden, pp. 4095–4104, 2018.
[33]
H. X. Liu, K. Simonyan, Y. M. Yang. Darts: Differentiable architecture search. In Proceedings of the 7th International Conference on Learning Representations, OpenReview.net, New Orleans, USA, 2019.
[34]
Wu B C, Dai X L, Zhang P Z, Wang Y H, Sun F, Wu Y M, Tian Y D, Vajda P, Jia Y Q, and Keutzer K Fb-Net: Hardware-aware efficient convNet design via differentiable neural architecture search Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2019 Long Beach, USA IEEE 10726-10734
[35]
He Y H, Lin J, Liu Z J, Wang H R, Li L J, and Han S AMC: AutoML for model compression and acceleration on mobile devices Proceedings of the 15th European Conference on Computer Vision 2018 Munich, Germany Springer 815-832
[36]
Dai X L, Zhang P Z, Wu B C, Yin H X, Sun F, Wang Y H, Dukhan M, Hu Y Q, Wu Y M, Jia Y Q, Vajda P, Uyttendaele M, and Jha N K ChamNet: Towards efficient network design through platform-aware model adaptation Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2019 Long Beach, USA IEEE 11390-11399
[37]
Wan A, Dai X L, Zhang P Z, He Z J, Tian Y D, Xie S N, Wu B C, Yu M, Xu T, Chen K, Vajda P, and Gonzalez J E FbNetV2: Differentiable neural architecture search for spatial and channel dimensions Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2020 Seattle, USA IEEE 12962-12971
[38]
X. L. Dai, A. Wan, P. Z. Zhang, B. C. Wu, Z. J. He, Z. Wei, K. Chen, Y. D. Tian, M. Yu, P. Vajda, J. E. Gonzalez. FBNetV3: Joint architecture-recipe search using neural acquisition function. [Online], Available: https://arxiv.org/abs/2006.02049, 2020.
[39]
Shen M Z, Han K, Xu C J, and Wang Y H Searching for accurate binary neural architectures Proceedings of 2019 IEEE/CVF International Conference on Computer Vision Workshop 2019 Seoul, Korea IEEE 2041-2044
[40]
S. Han, H. Z. Mao, W. J. Dally. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. [Online], Available: https://arxiv.org/abs/1510.00149, 2016.
[41]
Gui S P, Wang H N, Yang H C, Yu C, Wang Z Y, and Liu J Model compression with adversarial robustness: A unified optimization framework Proceedings of Advances in Neural Information Processing Systems 2019 Vancouver, Canada Neur-IPS 1283-1294
[42]
Luo J H, Wu J X, and Lin W Y Thinet: A filter level pruning method for deep neural network compression Proceedings of IEEE International Conference On Computer Vision 2017 Venice, Italy IEEE 5068-5076
[43]
C. J. Liu, Y. H. Wang, K. Han, C. J. Xu, C. Xu. Learning instance-wise sparsity for accelerating deep models. In Proceedings of the 24th International Joint Conference on Artificial Intelligence, Macao, China, pp. 3001–3007, 2019.
[44]
W. Wen, C. P. Wu, Y. D. Wang, Y. R. Chen, H. Li. Learning structured sparsity in deep neural networks. In Proceedings of the 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, pp. 2082–2090, 2016.
[45]
Liu Z C, Mu H Y, Zhang X Y, Guo Z C, Yang X, Cheng K T, and Sun J Metapruning: Meta learning for automatic neural network channel pruning Proceedings of IEEE/CVF International Conference on Computer Vision 2019 Seoul, Korea IEEE 3295-3304
[46]
Hu J, Shen L, and Sun G Squeeze-and-excitation networks Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition 2018 Salt Lake City, USA IEEE 7132-7141
[47]
S. Ioffe, C. Szegedy. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In Proceedings of the 32nd International Conference on Machine Learning, Lille, France, pp. 448–456, 2015.
[48]
Deng J, Dong W, Socher R, Li L J, Li K, and Li F F ImageNet: A large-scale hierarchical image database Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2009 Miami, USA IEEE 248-255
[49]
Fu X B, Yue S L, and Pan D Y Camera-based basketball scoring detection using convolutional neural network International Journal of Automation and Computing 2018 18 2 266-276
[50]
Aukkapinyo K, Sawangwong S, Pooyoi P, and Kusakunniran W Localization and classification of rice-grain images using region proposals-based convolutional neural network International Journal of Automation and Computing 2020 17 2 233-246
Index Terms
- PokerNet: Expanding Features Cheaply via Depthwise Convolutions
Index terms have been assigned to the content through auto-classification.
Recommendations
PAC-Bayesian framework based drop-path method for 2D discriminative convolutional network pruning
AbstractDeep convolutional neural networks (CNNs) have demonstrated its extraordinary power on various visual tasks like object detection and classification. However, it is still challenging to deploy state-of-the-art models into real-world applications, ...
RiSA: A Reinforced Systolic Array for Depthwise Convolutions and Embedded Tensor Reshaping
Special Issue ESWEEK 2021, CASES 2021, CODES+ISSS 2021 and EMSOFT 2021Depthwise convolutions are widely used in convolutional neural networks (CNNs) targeting mobile and embedded systems. Depthwise convolution layers reduce the computation loads and the number of parameters compared to the conventional convolution layers. ...
Fusing Depthwise and Pointwise Convolutions for Efficient Inference on GPUs
ICPP Workshops '24: Workshop Proceedings of the 53rd International Conference on Parallel ProcessingDepthwise and pointwise convolutions have fewer parameters and perform fewer operations than standard convolutions. As a result, they have become increasingly used in various compact DNNs, including convolutional neural networks (CNNs) and vision ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
© Institute of Automation, Chinese Academy of Sciences and Springer-Verlag GmbH Germany, part of Springer Nature 2021.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 01 June 2021
Accepted: 01 February 2021
Received: 23 December 2020
Author Tags
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 17 Jan 2025
Other Metrics
Citations
Cited By
View all- Kamath VRenuka A(2023)Deep learning based object detection for resource constrained devicesNeurocomputing10.1016/j.neucom.2023.02.006531:C(34-60)Online publication date: 28-Apr-2023