Abstract
To facilitate intelligent vision applications such as traffic surveillance and satellite monitoring, resource-constrained cameras are widely deployed in remote locations to capture and transmit images to edge servers for object detection. However, the network environment in Internet of Things scenarios experiences extremely constrained bandwidth and interference during long-distance transmission, leading to intolerable prolonged end-to-end latency and ineffective detection. Inspired by recently emerging generative vision models such as Masked AutoEncoder, we proposed GEN, an edge-assisted object detection system expedited by a novel image transmission scheme based on MAE. Specifically, GEN incorporates MAE by adopting patch-based image transmission. Only a small portion of image patches are required to transmit, while the complete image is reconstructed at the edge before detection. For transmitting each image patch, a blockwise patch packetization method aided by interpolation-based patch restoration is designed to cope with pixel loss incurred by interference. In order to prioritize the transmission of patches potentially containing more object features, a lightweight content-aware patch prioritization approach is adopted to ensure that enough object features are received before initiating image reconstruction. Evaluation results show that GEN can achieve \(2.6\times \) speedup while maintaining similar detection accuracy in bandwidth-constrained network environment compared with the existing edge-assisted object detection systems that adopt the common compression-based image transmission scheme.
This work was supported in part by the National Science Foundation of China (No. U20A20159); Guangdong Basic and Applied Basic Research Foundation (No. 2023B1515120058, No. 2021B151520008); Guangzhou Basic and Applied Basic Research Program (No. 2024A04J6367).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Jetson nx (2022). https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-xavier-series/
Raspberry pi 3bp (2022). https://www.raspberrypi.com/products/raspberry-pi-3-model-b-plus/
Sx1268 (2022). https://www.semtech.com/products/wireless-rf/lora-connect/sx1268
Prometheus (2023). https://prometheus.io/
Tensorflow compression library (2023). https://github.com/tensorflow/compression
Ballé, J.: Efficient nonlinear transforms for lossy image compression. In: 2018 Picture Coding Symposium (PCS), pp. 248–252. IEEE (2018)
Ballé, J., Minnen, D., Singh, S., Hwang, S.J., Johnston, N.: Variational image compression with a scale hyperprior. arXiv preprint arXiv:1802.01436 (2018)
Bradski, G.: The OpenCV library. Dr. Dobb’s J. Softw. Tools Prof. Program. 25(11), 120–123 (2000)
Chen, T., Eager, D., Makaroff, D.: Efficient image transmission using Lora technology in agricultural monitoring IoT systems. In: 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), pp. 937–944. IEEE (2019)
Desai, H., Nardello, M., Brunelli, D., Lucia, B.: Camaroptera: a long-range image sensor with local inference for remote sensing applications. ACM Tran. Embed. Comput. Syst. (TECS) 21(3), 1–25 (2022)
Everingham, M., Van Gool, L., Williams, C.K.I., Winn, J., Zisserman, A.: The PASCAL Visual Object Classes Challenge 2012 (VOC2012) Results. http://www.pascal-network.org/challenges/VOC/voc2012/workshop/index.html
Jocher, G., et al.: YOLOv5 (2020). https://doi.org/10.5281/zenodo.5563715
Gadre, A., Kumar, S., Manchester, Z.: Low-latency imaging and inference from Lora-enabled cubesats. arXiv preprint arXiv:2206.10703 (2022)
Giordano, M., Mayer, P., Magno, M.: A battery-free long-range wireless smart camera for face detection. In: Proceedings of the 8th International Workshop on Energy Harvesting and Energy-Neutral Sensing Systems, pp. 29–35 (2020)
He, K., Chen, X., Xie, S., Li, Y., Dollár, P., Girshick, R.: Masked autoencoders are scalable vision learners. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16000–16009 (2022)
Liando, J.C., Gamage, A., Tengourtius, A.W., Li, M.: Known and unknown facts of Lora: experiences from a large-scale measurement study. ACM Trans. Sens. Netw. (TOSN) 15(2), 1–35 (2019)
Lin, T.-Y., et al.: Microsoft COCO: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part V. LNCS, vol. 8693, pp. 740–755. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10602-1_48
Mentzer, F., Toderici, G.D., Tschannen, M., Agustsson, E.: High-fidelity generative image compression. In: Advances in Neural Information Processing Systems, vol. 33, pp. 11913–11924 (2020)
Minnen, D., Ballé, J., Toderici, G.D.: Joint autoregressive and hierarchical priors for learned image compression. In: Advances in Neural Information Processing Systems, vol. 31 (2018)
Minnen, D., Singh, S.: Channel-wise autoregressive entropy models for learned image compression. In: 2020 IEEE International Conference on Image Processing (ICIP), pp. 3339–3343. IEEE (2020)
Qiao, J., et al.: Communication resources limited decentralized learning with privacy guarantee through over-the-air computation. In: Proceedings of the Twenty-Fourth International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing, pp. 201–210 (2023)
Ren, J., Guo, Y., Zhang, D., Liu, Q., Zhang, Y.: Distributed and efficient object detection in edge computing: challenges and solutions. IEEE Netw. 32(6), 137–143 (2018)
Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, vol. 28 (2015)
Rublee, E., Rabaud, V., Konolige, K., Bradski, G.: ORB: an efficient alternative to sift or surf. In: 2011 International Conference on Computer Vision, pp. 2564–2571. IEEE (2011)
Sethian, J.A.: Fast marching methods. SIAM Rev. 41(2), 199–235 (1999)
Sun, Z., Yang, H., Liu, K., Yin, Z., Li, Z., Xu, W.: Recent advances in Lora: a comprehensive survey. ACM Trans. Sens. Netw. 18(4), 1–44 (2022)
Wallace, G.K.: The jpeg still picture compression standard. Commun. ACM 34(4), 30–44 (1991)
Wang, Y., et al.: Theoretical convergence guaranteed resource-adaptive federated learning with mixed heterogeneity. In: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pp. 2444–2455 (2023)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2025 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, J., Luo, K., Wang, H., Chen, X. (2025). Generative Model-Based Edge-Assisted Object Detection in Bandwidth-Constrained Network. In: Cai, Z., Takabi, D., Guo, S., Zou, Y. (eds) Wireless Artificial Intelligent Computing Systems and Applications. WASA 2024. Lecture Notes in Computer Science, vol 14997. Springer, Cham. https://doi.org/10.1007/978-3-031-71464-1_26
Download citation
DOI: https://doi.org/10.1007/978-3-031-71464-1_26
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-71463-4
Online ISBN: 978-3-031-71464-1
eBook Packages: Computer ScienceComputer Science (R0)