Skip to main content
SpringerLink
Log in

Small, Accurate, and Fast Re-ID on the Edge: The SAFR Approach

  • Conference paper
  • First Online:
Edge Computing – EDGE 2020 (EDGE 2020)

Part of the book series: Lecture Notes in Computer Science ((LNISA,volume 12407))

Included in the following conference series:

  • 594 Accesses

Abstract

We propose a Small, Accurate, and Fast Re-ID (SAFR) design for flexible vehicle re-id under a variety of compute environments such as cloud, mobile, edge, or embedded devices by only changing the re-id model backbone. Through best-fit design choices, feature extraction, training tricks, global attention, and local attention, we create a re-id model design that optimizes multi-dimensionally along model size, speed, & accuracy for deployment under various memory and compute constraints. We present several variations of our flexible SAFR model: SAFR-Large for cloud-type environments with large compute resources, SAFR-Small for mobile devices with some compute constraints, and SAFR-Micro for edge devices with severe memory & compute constraints.

SAFR-Large delivers state-of-the-art results with mAP 81.34 on the VeRi-776 vehicle re-id dataset (15% better than related work). SAFR-Small trades a 5.2% drop in performance (mAP 77.14 on VeRi-776) for over 60% model compression and 150% speedup. SAFR-Micro, at only 6 MB and 130 MFLOPS, trades 6.8% drop in accuracy (mAP 75.80 on VeRi-776) for 95% compression and 33x speedup compared to SAFR-Large .

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

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    convolutional.

  2. 2.

    normalization.

  3. 3.

    github.com/megvii-model/ShuffleNet-Series/.

References

  1. Ananthanarayanan, G., et al.: Real-time video analytics: the killer app for edge computing. Computer 50(10), 58–67 (2017)

    Article  Google Scholar 

  2. Wan, Y., Huang, Y., Buckles, B.: Camera calibration and vehicle tracking: highway traffic video analytics. Transp. Res. Part C Emerg. Technol. 44, 202–213 (2014)

    Article  Google Scholar 

  3. Chang, M.C., Wei, Y., Song, N., Lyu, S.: Video analytics in smart transportation for the AIC’18 challenge. In: CVPR Workshops (2018)

    Google Scholar 

  4. Liu, X., Zhang, S., Huang, Q., Gao, W.: Ram: a region-aware deep model for vehicle re-identification. In: IEEE International Conference on Multimedia and Expo, pp. 1–6. IEEE (2018)

    Google Scholar 

  5. Wang, Z., et al.: Orientation invariant feature embedding and spatial temporal regularization for vehicle re-identification. In: ICCV

    Google Scholar 

  6. Zhou, Y., Shao, L.: Aware attentive multi-view inference for vehicle re-ID. In: CVPR

    Google Scholar 

  7. Lou, Y., Bai, Y., Liu, J., Wang, S., Duan, L.Y.: Embedding adversarial learning for vehicle re-ID. IEEE Trans. Image Process. (2019)

    Google Scholar 

  8. Bai, Y., Lou, Y., Gao, F., Wang, S., Wu, Y., Duan, L.Y.: Group-sensitive triplet embedding for vehicle reidentification. IEEE Trans. Multimedia 20(9), 2385–2399 (2018)

    Article  Google Scholar 

  9. Jiang, J., Ananthanarayanan, G., Bodik, P., Sen, S., Stoica, I.: Chameleon: scalable adaptation of video analytics. In: ACM SIG Data Communication, pp. 253–266 (2018)

    Google Scholar 

  10. Chen, T., et al.: MXNet: a flexible and efficient machine learning library for heterogeneous distributed systems. arXiv:1512.01274 (2015)

  11. Ma, N., Zhang, X., Zheng, H.-T., Sun, J.: ShuffleNet V2: practical guidelines for efficient CNN architecture design. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) Computer Vision – ECCV 2018. LNCS, vol. 11218, pp. 122–138. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01264-9_8

    Chapter  Google Scholar 

  12. Howard, A.G., et al.: Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861 (2017)

  13. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR, pp. 770–778 (2016)

    Google Scholar 

  14. Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: CVPR (2016)

    Google Scholar 

  15. Cheng, Y., Wang, D., Zhou, P., Zhang, T.: Survey of model compression and acceleration for deep neural networks. arXiv:1710.09282

  16. Shen, Y., Xiao, T., Li, H., Yi, S., Wang, X.: Learning DNNs for vehicle re-id with visual-spatio-temporal path proposals. In: ICCV

    Google Scholar 

  17. Kanaci, A., Li, M., Gong, S., Rajamanoharan, G.: Multi-task mutual learning for vehicle re-ID. In: CVPR Workshops, pp. 62–70

    Google Scholar 

  18. Zhu, J., et al.: Vehicle re-identification using quadruple directional deep learning features. IEEE Trans. Intell. Transp. Syst. (2019)

    Google Scholar 

  19. Luo, H., Gu, Y., Liao, X., Lai, S., Jiang, W.: Bag of tricks and a strong baseline for deep person re-identification. In: CVPR Workshops

    Google Scholar 

  20. Liu, H., Tian, Y., Yang, Y., Pang, L., Huang, T.: Deep relative distance learning: tell the difference between similar vehicles. In: CVPR, pp. 2167–2175 (2016)

    Google Scholar 

  21. Liu, X., Liu, W., Mei, T., Ma, H.: A deep learning-based approach to progressive vehicle re-identification for urban surveillance. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9906, pp. 869–884. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46475-6_53

    Chapter  Google Scholar 

  22. Lou, Y., Bai, Y., Liu, J., Wang, S., Duan, L.: VERI-wild: a large dataset and a new method for vehicle re-identification in the wild. In: CVPR

    Google Scholar 

  23. Kanacı, A., Zhu, X., Gong, S.: Vehicle re-identification in context. In: Brox, T., Bruhn, A., Fritz, M. (eds.) GCPR 2018. LNCS, vol. 11269, pp. 377–390. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-12939-2_26

    Chapter  Google Scholar 

  24. Gale, T., Elsen, E., Hooker, S.: The state of sparsity in deep neural networks. arXiv:1902.09574 (2019)

  25. Narang, S., Elsen, E., Diamos, G., Sengupta, S.: Exploring sparsity in recurrent neural networks. arXiv:1704.05119

  26. Woo, S., Park, J., Lee, J.-Y., Kweon, I.S.: CBAM: convolutional block attention module. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11211, pp. 3–19. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01234-2_1

    Chapter  Google Scholar 

  27. He, B., Li, J., Zhao, Y., Tian, Y.: Part-regularized near-duplicate vehicle re-identification. In: CVPR, pp. 3997–4005

    Google Scholar 

  28. Basha, S., Dubey, S.R., Pulabaigari, V., Mukherjee, S.: Impact of fully connected layers on performance of CNNs for image classification. arXiv:1902.02771 (2019)

  29. Alfasly, S., Hu, Y., Li, H., Liang, T., Jin, X., Liu, B., Zhao, Q.: Multi-label-based similarity learning for vehicle re-identification. IEEE Access 7, 162605–162616 (2019)

    Article  Google Scholar 

  30. Zhang, X., Zhou, X., Lin, M., Sun, J.: Shufflenet: an extremely efficient CNN for mobile devices. In: CVPR (2018)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. School of Computer Science, Georgia Institute of Technology, Atlanta, USA

    Abhijit Suprem & Calton Pu

  2. Department of Mathematics and Statistics, University of Sao Paulo, Sao Paulo, Brazil

    Joao Eduardo Ferreira

Authors
  1. Abhijit Suprem
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Calton Pu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Joao Eduardo Ferreira
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Abhijit Suprem .

Editor information

Editors and Affiliations

  1. Missouri State University, Springfield, IL, USA

    Ajay Katangur

  2. North Carolina State University, Raleigh, NC, USA

    Shih-Chun Lin

  3. Department of Software and Information System, University of North Carolina, Charlotte, NC, USA

    Jinpeng Wei

  4. Purdue University Northwest, Hammond, IN, USA

    Shuhui Yang

  5. Kingdee International Software Group Co., Ltd., Shenzhen, China

    Liang-Jie Zhang

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Suprem, A., Pu, C., Ferreira, J.E. (2020). Small, Accurate, and Fast Re-ID on the Edge: The SAFR Approach. In: Katangur, A., Lin, SC., Wei, J., Yang, S., Zhang, LJ. (eds) Edge Computing – EDGE 2020. EDGE 2020. Lecture Notes in Computer Science(), vol 12407. Springer, Cham. https://doi.org/10.1007/978-3-030-59824-2_5

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-59824-2_5

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-59823-5

  • Online ISBN: 978-3-030-59824-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation