Key Considerations for Real-Time Object Recognition on Edge Computing Devices
Source abstract
Abstract
The rapid growth of the Internet of Things (IoT) and smart devices has led to an increasing demand for real-time data processing at the edge of networks closer to the source of data generation. This review paper introduces how artificial intelligence (AI) can be integrated with edge computing to enable efficient and scalable object recognition applications. It covers the key considerations of employing deep learning on edge computing devices, such as selecting edge devices, deep learning frameworks, lightweight deep learning models, hardware optimization, and performance metrics. An example of an application is also presented in this article, which is about real-time power transmission line detection using edge computing devices. The evaluation results show the significance of implementing lightweight models and model compression techniques such as quantized Tiny YOLOv7. It also shows the hardware performance on some edge devices, such as Raspberry Pi and Jetson platforms. Through practical examples, readers will gain insights into designing and implementing AI-powered edge solutions for various object recognition use cases, including smart surveillance, autonomous vehicles, and industrial automation. The review concludes by addressing emerging trends, such as federated learning and hardware accelerators, which are set to shape the future of AI on edge computing for object recognition.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Neural Network Applications · Visual Attention and Saliency Detection · IoT and Edge/Fog Computing
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electrical, Electronics and Communication Engineering, Faculty Science and Engineering, Tokyo City University, Setagaya-ku, Tokyo 158-8557, Japan
School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung 40132, Indonesia
Funding
Tokyo City University Prioritized Studies
References (100+)
Show source-provided references
1. Edge AI: A survey10.1016/j.iotcps.2023.02.004 ↗
2. A Survey of Recent Advances in Edge-Computing-Powered Artificial Intelligence of Things10.1109/JIOT.2021.3088875 ↗
3. Wang, X., Han, Y., Leung, V.C.M., Niyato, D., Yan, X., and Chen, X. (2020). Edge AI: Convergence of Edge Computing and Artificial Intelligence, Springer Nature.10.1007/978-981-15-6186-3 ↗
4. The emergence of edge computing10.1109/MC.2017.9 ↗
5. Bringing AI to edge: From deep learning’s perspective10.1016/j.neucom.2021.04.141 ↗
6. Edge Intelligence Empowered Vehicle Detection and Image Segmentation for Autonomous Vehicles10.1109/TITS.2022.3232153 ↗
7. Deep Learning Inferencing with High-performance Hardware Accelerators10.1145/3594221 ↗
8. Custom Hardware Inference Accelerator for TensorFlow Lite for Microcontrollers10.1109/ACCESS.2022.3189776 ↗
9. TensorRT-Based Framework and Optimization Methodology for Deep Learning Inference on Jetson Boards10.1145/3508391 ↗
10. Jajal, P., Jiang, W., Tewari, A., Kocinare, E., Woo, J., Sarraf, A., Lu, Y.H., Thiruvathukal, G.K., and Davis, J.C. (2024, January 16–20). Interoperability in Deep Learning: A User Survey and Failure Analysis of ONNX Model Converters. Proceedings of the ISSTA 2024: 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis, Vienna, Austria.10.1145/3650212.3680374 ↗
11. Prerad, D., Ivanovic, Z., and Avramovic, A. (2024, January 6–8). Edge AI acceleration using OpenCL framework on NXP i.MX 8M Nano SoM. Proceedings of the 2024 XV International Symposium on Industrial Electronics and Applications (INDEL), Banja Luka, Bosnia and Herzegovina.10.1109/INDEL62640.2024.10772692 ↗
12. Wang, J., and Gu, S. (2021, January 21–23). FPGA Implementation of Object Detection Accelerator Based on Vitis-AI. Proceedings of the 2021 11th International Conference on Information Science and Technology (ICIST), Chengdu, China.10.1109/ICIST52614.2021.9440554 ↗
13. Edge Artificial Intelligence for Industrial Internet of Things Applications: An Industrial Edge Intelligence Solution10.1109/MIE.2020.3026837 ↗
14. Deep Learning with Edge Computing: A Review10.1109/JPROC.2019.2921977 ↗
15. Edge YOLO: Real-Time Intelligent Object Detection System Based on Edge-Cloud Cooperation in Autonomous Vehicles10.1109/TITS.2022.3158253 ↗
16. Gupta, H., Kotlyar, O., Andreasson, H., and Lilienthal, A.J. (2024, January 3–8). Robust Object Detection in Challenging Weather Conditions. Proceedings of the 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, USA.10.1109/WACV57701.2024.00735 ↗
17. A small object detection algorithm based on feature interaction and guided learning10.1016/j.jvcir.2023.104011 ↗
18. Chen, S., Ye, M., and Du, B. (2022, January 10–14). Rotation Invariant Transformer for Recognizing Object in UAVs. Proceedings of the MM ’22: 30th ACM International Conference on Multimedia, Lisbon, Portugal.10.1145/3503161.3547799 ↗
19. Khoa, T.A., Van Nguyen, D., Dao, M.S., and Zettsu, K. (2021, January 15–18). Fed xData: A Federated Learning Framework for Enabling Contextual Health Monitoring in a Cloud-Edge Network. Proceedings of the 2021 IEEE International Conference on Big Data (Big Data), Orlando, FL, USA.10.1109/BigData52589.2021.9671536 ↗
20. Liang, H., He, E., Zhao, Y., Jia, Z., and Li, H. (2022). Adversarial Attack and Defense: A Survey. Electronics, 11.10.3390/electronics11081283 ↗
21. Chen, E.-C., Chen, P.-Y., Chung, I.-H., and Lee, C. (2023). Overload: Latency Attacks on Object Detection for Edge Devices. arXiv.10.1109/CVPR52733.2024.02334 ↗
22. Surantha, N., Yamashina, E., Sato, Y., and Iwao, T. (2024, January 24–27). Power Transmission Line Component Detection Using YOLOv7 on Single-Board Computer Platforms. Proceedings of the 2024 IEEE 99th Vehicular Technology Conference (VTC2024-Spring), Singapore.10.1109/VTC2024-Spring62846.2024.10683056 ↗
23. Real-time strawberry detection using deep neural networks on embedded system (rtsd-net): An edge AI application10.1016/j.compag.2021.106586 ↗
24. AI-Assisted Edge Vision for Violence Detection in IoT-Based Industrial Surveillance Networks10.1109/TII.2021.3116377 ↗
25. Edge Computing with Artificial Intelligence: A Machine Learning Perspective10.1145/3555802 ↗
26. Edge Computing Driven Low-Light Image Dynamic Enhancement for Object Detection10.1109/TNSE.2022.3151502 ↗
27. Kalyani, Y., and Collier, R. (2021). A systematic survey on the role of cloud, fog, and edge computing combination in smart agriculture. Sensors, 21.10.3390/s21175922 ↗
28. Dynamic task offloading for resource allocation and privacy-preserving framework in Kubeedge-based edge computing using machine learning10.1007/s10586-024-04420-8 ↗
29. Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing10.1109/TWC.2019.2946140 ↗
30. Leveraging the edge and cloud for V2X-based real-time object detection in autonomous driving10.1016/j.comcom.2023.11.025 ↗
31. Rouf, A., Arslan, E., and Charyyev, B. (2024, January 20). Latency and Bandwidth Benefits of Edge Computing for Scientific Applications. Proceedings of the IEEE INFOCOM 2024—IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Vancouver, BC, Canada.10.1109/INFOCOMWKSHPS61880.2024.10620811 ↗
32. Privacy-Preserving Uncertainty Calibration using Perceptual Encryption in Cloud-Edge Collaborative Artificial Intelligence of Things10.1109/JIOT.2025.3558289 ↗
33. A privacy-preserving learning framework for face recognition in edge and cloud networks10.1109/ACCESS.2020.3011112 ↗
34. Learning Spatiotemporal Failure Dependencies for Resilient Edge Computing Services10.1109/TPDS.2020.3046188 ↗
35. FPGA-Based Real-Time Object Detection and Classification System Using YOLO for Edge Computing10.1109/ACCESS.2024.3404623 ↗
36. Deep Reinforcement Learning-Based Energy-Efficient Edge Computing for Internet of Vehicles10.1109/TII.2022.3155162 ↗
37. A novel middleware for adaptive and efficient split computing for real-time object detection10.1016/j.pmcj.2025.102028 ↗
38. Delay-Aware Cooperative Task Offloading for Multi-UAV Enabled Edge-Cloud Computing
39. Near-Edge Computing Aware Object Detection: A Review10.1109/ACCESS.2023.3347548 ↗
40. Netscore: Towards universal metrics for large-scale performance analysis of deep neural networks for practical on-device edge usage
41. Thermal-Aware Scheduling for Deep Learning on Mobile Devices With NPU10.1109/TMC.2024.3379501 ↗
42. Darian, G., and Surantha, N. (2025). Performance Evaluation of Convolutional Neural Network (CNN) for Skin Cancer Detection on Edge Computing Devices. Appl. Sci., 15.10.3390/app15063077 ↗
43. Lightweight face recognition-based portable attendance system with liveness detection10.1016/j.iot.2024.101089 ↗
44. Low Latency YOLOv3-Tiny Accelerator for Low-Cost FPGA Using General Matrix Multiplication Principle10.1109/ACCESS.2021.3120629 ↗
45. Fast and Scalable Multicore YOLOv3-Tiny Accelerator Using Input Stationary Systolic Architecture10.1109/TVLSI.2023.3305937 ↗
46. Design of Smart Home Security System using Object Recognition and PIR Sensor10.1016/j.procs.2018.08.198 ↗
47. Deep learning for object recognition: A comprehensive review of models and algorithms
48. Vandendriessche, J., Wouters, N., da Silva, B., Lamrini, M., Chkouri, M.Y., and Touhafi, A. (2021). Environmental sound recognition on embedded systems: From fpgas to tpus. Electronics, 10.10.3390/electronics10212622 ↗
49. Correction: An implementation of real-time traffic signs and road objects detection based on mobile GPU platforms10.1109/ACCESS.2022.3198954 ↗
50. Shahid, A., and Mushtaq, M. (2020, January 5–7). A Survey Comparing Specialized Hardware and Evolution in TPUs for Neural Networks. Proceedings of the 2020 IEEE 23rd International Multitopic Conference (INMIC), Bahawalpur, Pakistan.10.1109/INMIC50486.2020.9318136 ↗
51. Hu, Y., Liu, Y., and Liu, Z. (2022, January 7–9). A Survey on Convolutional Neural Network Accelerators: GPU, FPGA and ASIC. Proceedings of the 2022 14th International Conference on Computer Research and Development (ICCRD), Shenzhen, China.10.1109/ICCRD54409.2022.9730377 ↗
52. Benchmarking Deep Learning Models on NVIDIA Jetson Nano for Real-Time Systems: An Empirical Investigation10.1016/j.procs.2025.03.273 ↗
53. Scalcon, F.P., Tahal, R., Ahrabi, M., Huangfu, Y., Ahmed, R., Nahid-Mobarakeh, B., Shirani, S., Vidal, C., and Emadi, A. (2024, January 19–21). AI-Powered Video Monitoring: Assessing the NVIDIA Jetson Orin Devices for Edge Computing Applications. Proceedings of the 2024 IEEE Transportation Electrification Conference and Expo (ITEC), Chicago, IL, USA.10.1109/ITEC60657.2024.10598994 ↗
54. Winzig, J., Almanza, J.C.A., Mendoza, M.G., and Schumann, T. (2022, January 14–16). Edge AI—Use Case on Google Coral Dev Board Mini. Proceedings of the 2022 IET International Conference on Engineering Technologies and Applications (IET-ICETA), Changhua, Taiwan.10.1109/IET-ICETA56553.2022.9971614 ↗
55. Surantha, N., Sukizaki, Y., Yamashina, E., and Iwao, T. (2023, January 16–18). Power Transmission Line Component Detection using YOLO V3 on Raspberry Pi. Proceedings of the 2023 22nd International Symposium on Communications and Information Technologies (ISCIT), Sydney, Australia.10.1109/ISCIT57293.2023.10376097 ↗
56. Ramesh, S.N., Sarkar, M., Audette, M., and Paolini, C. (2023, January 19–21). Efficient Real-time Fall Prediction and Detection using Privacy-Centric Vision-based Human Pose Estimation on the Xilinx® KriaTM K26 SOM. Proceedings of the 2023 IEEE Biomedical Circuits and Systems Conference (BioCAS), Toronto, ON, Canada.10.1109/BioCAS58349.2023.10388590 ↗
57. Choi, P., Kim, J., and Kwak, J. (2024, January 3–7). Impact of Joint Heat and Memory Constraints of Mobile Device in Edge-Assisted On-Device Artificial Intelligence. Proceedings of the NetAISys ’24: 2nd International Workshop on Networked AI Systems, Tokyo, Japan.10.1145/3662004.3663555 ↗
58. Lightweight convolutional neural network architecture implementation using TensorFlow lite
59. Iqbal, U., Davies, T., and Perez, P. (2024). A Review of Recent Hardware and Software Advances in GPU-Accelerated Edge-Computing Single-Board Computers (SBCs) for Computer Vision. Sensors, 24.10.3390/s24154830 ↗
60. Zunin, V.V. (2021, January 5–11). Intel OpenVINO Toolkit for Computer Vision: Object Detection and Semantic Segmentation. Proceedings of the 2021 International Russian Automation Conference (RusAutoCon), Sochi, Russia.10.1109/RusAutoCon52004.2021.9537452 ↗
61. guk OpenCL-Darknet: Implementation and optimization of OpenCL-based deep learning object detection framework10.1007/s11280-020-00778-y ↗
62. Evaluation of Model Quantization Method on Vitis-AI for Mitigating Adversarial Examples10.1109/ACCESS.2023.3305264 ↗
63. Implementation of the PointPillars Network for 3D Object Detection in Reprogrammable Heterogeneous Devices Using FINN10.1007/s11265-021-01733-4 ↗
64. Andreetto, M., Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., and Weyand, T. (2025, May 03). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications Automatic 3-D Modeling of Textured Cultural Heritage Objects View project MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. Available online: https://www.researchgate.net/publication/316184205.
65. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.C. (2018, January 18–23). MobileNetV2: Inverted Residuals and Linear Bottlenecks. Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00474 ↗
66. Chu, G., Chen, L., Chen, B., Tan, M., Howard, A., Wang, W., and Sandler, M. (November, January 27). Searching for MobileNetV3 Accuracy vs MAdds vs model size. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), Seoul, Republic of Korea.
67. YOLOv1 to v8: Unveiling Each Variant-A Comprehensive Review of YOLO10.1109/ACCESS.2024.3378568 ↗
68. YOLOv3: An Incremental Improvement
69. Adarsh, P., Rathi, P., and Kumar, M. (2020, January 6–7). YOLO v3-Tiny: Object Detection and Recognition using one stage improved model. Proceedings of the 2020 6th International Conference on Advanced Computing and Communication Systems (ICACCS), Coimbatore, India.10.1109/ICACCS48705.2020.9074315 ↗
70. Improved Traffic Sign Detection Model Based on YOLOv7-Tiny10.1109/ACCESS.2023.3331426 ↗
71. Wang, C.-Y., Bochkovskiy, A., and Liao, H.-Y.M. (2023, January 17–24). YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.10.1109/CVPR52729.2023.00721 ↗
72. Efficientnet-Lite and Hybrid CNN-KNN Implementation for Facial Expression Recognition on Raspberry Pi10.1109/ACCESS.2021.3113337 ↗
73. Mingxing Tan, Q.V. (2019, January 9–15). Le EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks Mingxing. Proceedings of the 36th International Conference on Machine Learning, Long Beach, CA, USA. Available online: https://proceedings.mlr.press/v97/tan19a/tan19a.pdf.
74. Zoph, B., Vasudevan, V., Shlens, J., and Le, Q.V. (2018, January 18–23). Learning Transferable Architectures for Scalable Image Recognition. Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00907 ↗
75. Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., and Keutzer, K. (2017). SqueezeNet: AlexNet-Level Accuracy with 50× Fewer Parameters and Less Than 0.5MB Model Size. arXiv.
76. Zhang, X., Lin, M., and Sun, J. (2018, January 18–23). ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA.10.1109/CVPR.2018.00716 ↗
77. A comprehensive review of model compression techniques in machine learning10.1007/s10489-024-05747-w ↗
78. A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classification10.1145/3623402 ↗
79. Model Pruning Enables Efficient Federated Learning on Edge Devices10.1109/TNNLS.2022.3166101 ↗
80. Knowledge Distillation: A Survey10.1007/s11263-021-01453-z ↗
81. Logarithmic Norm Regularized Low-Rank Factorization for Matrix and Tensor Completion10.1109/TIP.2021.3061908 ↗
82. COVID-19 and Non-COVID-19 Classification using Multi-layers Fusion From Lung Ultrasound Images10.1016/j.inffus.2021.02.013 ↗
83. Real-Time Multi-Learning Deep Neural Network on an MPSoC-FPGA for Intelligent Vehicles: Harnessing Hardware Acceleration With Pipeline10.1109/TIV.2024.3398215 ↗
84. On-the-Fly Parallel Processing IP-Core for Image Blur Detection, Compression, and Chaotic Encryption Based on FPGA10.1109/ACCESS.2021.3086044 ↗
85. Dataflow optimization with layer-wise design variables estimation method for enflame CNN accelerators10.1016/j.jpdc.2024.104869 ↗
86. Winograd Convolution for Deep Neural Networks: Efficient Point Selection10.1145/3524069 ↗
87. Sethi, G., Acun, B., Agarwal, N., Kozyrakis, C., Trippel, C., and Wu, C.J. (March, January 28). RecShard: Statistical Feature-Based Memory Optimization for Industry-Scale Neural Recommendation. Proceedings of the ASPLOS ’22: 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Lausanne, Switzerland.
88. Approximate Computing: Concepts, Architectures, Challenges, Applications, and Future Directions10.1109/ACCESS.2024.3467375 ↗
89. Tunable Energy-Efficient Approximate Circuits for Self-Powered AI and Autonomous Edge Computing Systems10.1109/ACCESS.2025.3548458 ↗
90. Rashidi, B., Gao, C., Lu, S., Wang, Z., Zhou, C., Niu, D., and Sun, F. (November, January 28). UNICO: Unified Hardware Software Co-Optimization for Robust Neural Network Acceleration. Proceedings of the MICRO ’23: 56th Annual IEEE/ACM International Symposium on Microarchitecture, Toronto, ON, Canada.
91. Marchisio, A., Hanif, M.A., Putra, R.V.W., and Shafique, M. (2018, January 4). HW/SW co-design and co-optimizations for deep learning. Proceedings of the INTESA ’18: Workshop on INTelligent Embedded Systems Architectures and Applications, Turin, Italy.10.1145/3285017.3285022 ↗
92. Shafique, M., Marchisio, A., Putra, R.V.W., and Hanif, M.A. (2021, January 1–4). Towards Energy-Efficient and Secure Edge AI: A Cross-Layer Framework. Proceedings of the 2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD), Munich, Germany.10.1109/ICCAD51958.2021.9643539 ↗
93. Hanif, M.A., and Shafique, M. (2024). Cross-Layer Optimizations for Efficient Deep Learning Inference at the Edge. Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing, Springer. Available online: https://link.springer.com/chapter/10.1007/978-3-031-39932-9_9.10.1007/978-3-031-39932-9_9 ↗
94. Colin, J., and Surantha, N. (2025). Interpretable Deep Learning for Pneumonia Detection Using Chest X-Ray Images. Information, 16.10.3390/info16010053 ↗
95. Padilla, R., Passos, W.L., Dias, T.L.B., Netto, S.L., and Da Silva, E.A.B. (2021). A comparative analysis of object detection metrics with a companion open-source toolkit. Electronics, 10.10.3390/electronics10030279 ↗
96. Yokomine, Y., and Surantha, N. (2025, January 18–21). Evaluation of Real-Time Train Overhead Line Component Detection on Edge Device. Proceedings of the 2025 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), Fukuoka, Japan.10.1109/ICAIIC64266.2025.10920705 ↗
97. Dong, W., Lv, J., Chen, G., Wang, Y., Li, H., Gao, Y., and Bharadia, D. (July, January 27). TinyNet: A Lightweight, Modular, and Unified Network Architecture for the Internet of Things. Proceedings of the MobiSys ’22: 20th Annual International Conference on Mobile Systems, Applications and Services, Portland, Oregon.
98. Moosmann, J., Bonazzi, P., Li, Y., Bian, S., Mayer, P., Benini, L., and Magno, M. (2023). Ultra-Efficient On-Device Object Detection on AI-Integrated Smart Glasses with TinyissimoYOLO, Springer Nature.
99. Yoo, Y., Niu, Z., Yoo, C., Cheng, P., and Xiong, Y. (2023, January 29–30). SegaNet: An Advanced IoT Cloud Gateway for Performant and Priority-Oriented Message Delivery. Proceedings of the APNet ’23: 7th Asia-Pacific Workshop on Networking, Hong Kong, China.10.1145/3600061.3600072 ↗
100. Chen, C., Min, H., Peng, Y., Yang, Y., and Wang, Z. (2022). An Intelligent Real-Time Object Detection System on Drones. Appl. Sci., 12.10.3390/app122010227 ↗