Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to main content
Springer Nature Link
Log in

Towards SSD accelerating for embedded environments: a compressive sensing based approach

  • Original Research Paper
  • Published:
Journal of Real-Time Image Processing Aims and scope Submit manuscript

Abstract

Since the rise of convolutional neural networks (CNN), deep learning-based computer vision has been a dynamic field of research. Nevertheless, modern CNN architectures have not given sufficient consideration to real−time applications within limited computation settings and always compromise speed and accuracy. To this end, a novel approach to CNN design, based on the emerging technology of compressive sensing (CS), is proposed. For instance, CS networks function in a compression−reconstruction approach as an encoder−decoder neural network. This approach transforms the computer vision problem into a multioutput learning problem by incorporating the CS network into a recognition network for joint training. As to the deployment phase, images are obtained from a CS−acquisition device and fed directly, without reconstruction, to the new recognition network. Following such an approach considerably improves transmission bandwidth and reduces the computational burden. Furthermore, the redesigned CNN holds fewer parameters than its original counterpart, thus reducing model complexity. To validate our findings, object detection using the Single−Shot Detector (SSD) network was redesigned to operate in our CS−based ecosystem using different datasets. The results show that the lightweight CS network offers good performance at a faster running speed. For instance, the number of FLOPS was reduced by 57% compared to the SSD baseline. Furthermore, the proposed CS_SSD achieves a compelling accuracy while being 30% faster than its original counterpart on small GPUs. Code is available at: https://github.com/Bouderbal-Imene/CS-SSD.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

References

  1. Adler, A., Elad, M., Zibulevsky, M.: Compressed learning: a deep neural network approach. arXiv:1610.09615 (2016)

  2. Agustsson, E., Timofte, R.: Ntire 2017 challenge on single image super-resolution: Dataset and study. In: Proceedings of the IEEE conference on computer vision and pattern recognition workshops, pp. 126–135 (2017)

  3. Bethi, Y.R.T., Narayanan, S., Rangan, V., Thakur, C.S.: Real-time object detection and localization in compressive sensed video on embedded hardware. arXiv:1912.08519 (2019)

  4. Bochkovskiy, A., Wang, C.Y., Liao, H.Y.M.: Yolov4: Optimal speed and accuracy of object detection. arXiv:2004.10934 (2020)

  5. Bouderbal, I., Amamra, A., Benatia, M.A.: How would image down-sampling and compression impact object detection in the context of self-driving vehicles? In: CSA, pp. 25–37 (2020)

  6. Bouderbal, I., Amamra, A., Benatia, M.A.: An analytical study of efficient cnns tuning and scaling for traffic signs recognition. In: 2021 international conference on recent advances in mathematics and informatics (ICRAMI), pp. 1–6, IEEE (2021)

  7. Cai, Z., Vasconcelos, N.: Cascade r-cnn: High quality object detection and instance segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence (2019)

  8. Deguerre, B., Chatelain, C., Gasso, G.: Fast object detection in compressed jpeg images. In: 2019 IEEE intelligent transportation systems conference (ITSC), pp. 333–338, IEEE (2019)

  9. Deguerre, B., Chatelain, C., Gasso, G.: Object detection in the dct domain: is luminance the solution? In: 2020 25th international conference on pattern recognition (ICPR), pp. 2627–2634, IEEE (2021)

  10. Ding, K., Ma, K., Wang, S., Simoncelli, E.P.: Comparison of full-reference image quality models for optimization of image processing systems. Int. J. Comput. Vis. 129(4), 1258–1281 (2021)

    Article  Google Scholar 

  11. Fowler, J.E., Mun, S., Tramel, E.W.: Multiscale block compressed sensing with smoothed projected landweber reconstruction. In: 2011 19th European signal processing conference, pp. 564–568, IEEE (2011)

  12. Girshick, R.: Fast r-cnn. In: Proceedings of the IEEE international conference on computer vision, pp. 1440–1448 (2015)

  13. Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 580–587 (2014)

  14. Gueguen, L., Sergeev, A., Kadlec, B., Liu, R., Yosinski, J.: Faster neural networks straight from jpeg. Adv. Neural Inf. Process. Syst. 31, 3933–3944 (2018)

    Google Scholar 

  15. Guo, S., Liu, Y., Ni, Y., Ni, W.: Lightweight ssd: real-time lightweight single shot detector for mobile devices. In: VISIGRAPP (5: VISAPP), pp. 25–35 (2021)

  16. Han, J., Yang, Y.: L-net: lightweight and fast object detector-based shufflenetv2. J. Real-Time Image Process. 18(6), 2527–2538 (2021)

    Article  Google Scholar 

  17. He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask r-cnn. In: Proceedings of the IEEE international conference on computer vision, pp. 2961–2969 (2017)

  18. Hore, A., Ziou, D.: Image quality metrics: Psnr vs. ssim. In: 2010 20th international conference on pattern recognition, pp. 2366–2369. IEEE (2010)

  19. Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., Adam, H.: Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861 (2017)

  20. Justus, D., Brennan, J., Bonner, S., McGough, A.S.: Predicting the computational cost of deep learning models. In: 2018 IEEE international conference on big data (Big Data), pp. 3873–3882, IEEE (2018)

  21. Khosravy, M., Gupta, N., Patel, N., Duque, C.A.: Recovery in compressive sensing: a review. In: Compressive sensing in healthcare, pp. 25–42 (2020)

  22. Kulkarni, K., Lohit, S., Turaga, P., Kerviche, R., Ashok, A.: Reconnet: non-iterative reconstruction of images from compressively sensed measurements. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 449–458 (2016)

  23. Li, Z., Sun, Y., Tian, G., Xie, L., Liu, Y., Su, H., He, Y.: A compression pipeline for one-stage object detection model. J. Real-Time Image Process. 18(6), 1949–1962 (2021)

    Article  Google Scholar 

  24. Liao, L., Li, K., Yang, C., Liu, J.: Low-cost image compressive sensing with multiple measurement rates for object detection. Sensors 19(9), 2079 (2019)

    Article  Google Scholar 

  25. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., Berg, A.C.: Ssd: single shot multibox detector. In: European conference on computer vision, pp. 21–37. Springer (2016)

  26. Ophoff, T., Van Beeck, K., Goedemé, T.: Exploring rgb+ depth fusion for real-time object detection. Sensors 19(4), 866 (2019)

    Article  Google Scholar 

  27. Padilla, R., Passos, W.L., Dias, T.L., Netto, S.L., da Silva, E.A.: A comparative analysis of object detection metrics with a companion open-source toolkit. Electronics 10(3), 279 (2021)

    Article  Google Scholar 

  28. Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 779–788 (2016)

  29. Redmon, J., Farhadi, A.: Yolo9000: better, faster, stronger. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 7263–7271 (2017)

  30. Redmon, J., Farhadi, A.: Yolov3: an incremental improvement. arXiv:1804.02767 (2018)

  31. Ren, K., Huang, L., Fan, C., Han, H., Deng, H.: Real-time traffic sign detection network using ds-detnet and lite fusion fpn. J. Real-Time Image Process. 18(6), 2181–2191 (2021)

    Article  Google Scholar 

  32. Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: towards real-time object detection with region proposal networks. Adv. Neural Inf. Process. Syst. 28, 91–99 (2015)

    Google Scholar 

  33. Shi, W., Jiang, F., Liu, S., Zhao, D.: Image compressed sensing using convolutional neural network. IEEE Trans. Image Process. 29, 375–388 (2019)

    Article  MathSciNet  MATH  Google Scholar 

  34. Torfason, R., Mentzer, F., Agustsson, E., Tschannen, M., Timofte, R., Van Gool, L.: Towards image understanding from deep compression without decoding. arXiv:1803.06131 (2018)

  35. Wallace, G.K.: The jpeg still picture compression standard. IEEE Trans. Consumer Electron. 38(1), 5 (1992)

    Article  Google Scholar 

  36. Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)

    Article  Google Scholar 

  37. Xu, D., Shi, Y., Tsang, I.W., Ong, Y.S., Gong, C., Shen, X.: Survey on multi-output learning. IEEE Trans. Neural Netw. Learn. Syst. 31(7), 2409–2429 (2019)

    MathSciNet  Google Scholar 

  38. Zhang, J., Zhao, D., Gao, W.: Group-based sparse representation for image restoration. IEEE Trans. Image Process. 23(8), 3336–3351 (2014)

    Article  MathSciNet  MATH  Google Scholar 

  39. Zhang, X., Xie, H., Zhao, Y., Qian, W., Xu, X.: A fast ssd model based on parameter reduction and dilated convolution. J. Real-Time Image Process. 18(6), 2211–2224 (2021)

    Article  Google Scholar 

  40. Zhao, Z.Q., Zheng, P., Xu, S.T., Wu, X.: Object detection with deep learning: a review. IEEE Trans. Neural Netw. Learn. Syst. 2019, 5 (2019)

    Google Scholar 

  41. Zhou, X., Xu, L., Liu, S., Lin, Y., Zhang, L., Zhuo, C.: An efficient compressive convolutional network for unified object detection and image compression. In: Proceedings of the AAAI conference on artificial intelligence, vol. 33, pp. 5949–5956 (2019)

  42. Zhu, R., Zhang, S., Wang, X., Wen, L., Shi, H., Bo, L., Mei, T.: Scratchdet: Training single-shot object detectors from scratch. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 2268–2277 (2019)

Download references

Author information

Authors and Affiliations

  1. Ecole Militaire Polytechnique, Bordj El-Bahri BP 17, Algiers, Algeria

    Imene Bouderbal, Abdenour Amamra, M. El-Arbi Djebbar & M. Akrem Benatia

Authors
  1. Imene Bouderbal
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Abdenour Amamra
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. M. El-Arbi Djebbar
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. M. Akrem Benatia
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Imene Bouderbal.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 11289 kb)

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bouderbal, I., Amamra, A., Djebbar, M.EA. et al. Towards SSD accelerating for embedded environments: a compressive sensing based approach. J Real-Time Image Proc 19, 1199–1210 (2022). https://doi.org/10.1007/s11554-022-01255-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11554-022-01255-7

Keywords

Search

Navigation