An Accelerating Method of YOLOv7 Based on Lightweight Network Architecture
Source abstract
Abstract
Object detection is a key technology in optical imaging detection systems. To address the issue of slow object detection model performance on mobile platforms, a lightweight network can be used as the backbone of the object detector, reducing the model’s parameters and increasing the inference speed. However, while using a lightweight network directly as the backbone improves inference speed, it significantly reduces detection accuracy. This paper proposes enhancing the backbone network by adding depthwise separable convolutional layers, which improves the receptive field of the model, enabling better detection performance in complex environments. The experimental results show that, compared to YOLOv7, the improved model reduces detection accuracy by 3.84%, while achieving a detection speed that is 3.77 times faster than the original model. This method significantly enhances the model’s suitability for real-time applications on resource-constrained devices, offering an effective solution for faster object detection without sacrificing critical performance in practical scenarios.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Computing and Algorithms
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
References (26)
Show source-provided references
1. Duman, B. (2025). A Real-Time Green and Lightweight Model for Detection of Liquefied Petroleum Gas Cylinder Surface Defects Based on YOLOv5. Appl. Sci., 15.10.3390/app15010458 ↗
2. Han, Z., Yue, Z., and Liu, L. (2025). 3L-YOLO: A Lightweight Low-Light Object Detection Algorithm. Appl. Sci., 15.10.3390/app15010090 ↗
3. Lv, Y., Tian, B., Guo, Q., and Zhang, D. (2025). A Lightweight Small Target Detection Algorithm for UAV Platforms. Appl. Sci., 15.10.3390/app15010012 ↗
4. The Application of Intelligent Technology in Precise Strike System
5. Redmon, J., Divvala, S.K., Girshick, R.B., and Farhadi, A. (2016, January 27–30). You Only Look Once: Unified, Real-Time Object Detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.91 ↗
6. Wang, C.-Y., Bochkovskiy, A., and Liao, H.-Y.M. (2023, January 18–22). YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.10.1109/CVPR52729.2023.00721 ↗
7. El-Dalahmeh, A., El-Dalahmeh, M., and Li, J. (July, January 30). Enhanced Vehicle Detection through Multi-Sensor Fusion Utilizing YOLO-NAS and Faster R-CNN. Proceedings of the 2024 International Joint Conference on Neural Networks (IJCNN), Yokohama, Japan.10.1109/IJCNN60899.2024.10651024 ↗
8. Wang, C.-Y., Yeh, I.-H., and Liao, H.-Y.M. (October, January 29). YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information. Proceedings of the 18th European Conference, Milan, Italy.
9. Wang, A., Chen, H., Liu, L., Chen, K., Lin, Z., Han, J., and Ding, G. (2024). YOLOv10: Real-Time End-to-End Object Detection. arXiv.
10. Xu, S., Wang, J., and Sang, Q. (2025). Semi-Supervised Method for Underwater Object Detection Algorithm Based on Improved YOLOv8. Appl. Sci., 15.10.3390/app15031065 ↗
11. Lightweight multiobject ship tracking algorithm based on trajectory association and improved YOLOv7tiny10.1016/j.eswa.2024.125129 ↗
12. Liu, Y., Ma, Y., Chen, F., Shang, E., Yao, W., Zhang, S., and Yang, J. (2024). YOLOv7oSAR: A Lightweight High-Precision Ship Detection Model for SAR Images Based on the YOLOv7 Algorithm. Remote Sens., 16.10.3390/rs16050913 ↗
13. A lightweight road crack detection algorithm based on improved YOLOv7 model10.1007/s11760-024-03197-y ↗
14. Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). Mobilenets: Efficient Convolutional Neural Networks for Mobile Vision Applocations. arXiv.
15. Iandola, F.N., Moskewicz, M.W., Ashraf, K., Han, S., Dally, W.J., and Keutzer, K. (2016). SqueezeNet: AlexNet-Level ACCURACY with 50x Fewer Parameters and <1 MB Model Size. arXiv.
16. Zhang, X., Zhou, X., Lin, M., and Sun, J. (2018, January 18–22). 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 ↗
17. MPQ-YOLO: Ultra low mixed-precision quantization of YOLO for edge devices deployment10.1016/j.neucom.2023.127210 ↗
18. Yang, Y. (2024, January 10–12). Quantization and Acceleration of YOLOv5 Vehicle Detection Based on GPU Chips. Proceedings of the GAIIS 2024: 2024 International Conference on Generative Artificial Intelligence and Information Security, Kuala Lumpur, Malaysia.10.1145/3665348.3665420 ↗
19. An improved insulator self-explosion detection method based on group-level pruning for the YOLOv7-tiny algorithm10.1007/s11554-024-01571-0 ↗
20. Han, S., Mao, H., and Dally, W.J. (2015). Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding. arXiv.
21. Securely outsourcing neural network inference to the cloud with lightweight techniques10.1109/TDSC.2022.3141391 ↗
22. Paralikas, I., Spantideas, S.T., Giannopoulos, A.E., and Trakadas, P. (2024, January 27–30). Lightweight Inference by Neural Network Pruning: Accuracy, Time and Comparison. Proceedings of the IFIP International Conference on Artificial Intelligence Applications and Innovations, Corfu, Greece.10.1007/978-3-031-63219-8_19 ↗
23. Aghapour, E., Shen, Y., Sapra, D., Pimentel, A.D., and Pathania, A. (2024, January 5–7). PiQi: Partially Quantized DNN Inference on HMPSoCs. Proceedings of the 29th ACM/IEEE International Symposium on Low Power Electronics and Design, Newport Beach, CA, USA.10.1145/3665314.3670841 ↗
24. Bert loses patience: Fast and robust inference with early exit
25. Teerapittayanon, S., McDanel, B., and Kung, H.T. (2016, January 4–8). Branchynet: Fast inference via early exiting from deep neural networks. Proceedings of the 2016 23rd International Conference on Pattern Recognition (ICPR), Cancun, Mexico.10.1109/ICPR.2016.7900006 ↗
26. Ma, N., Zhang, X., Zheng, H.-T., and Sun, J. (2018, January 8–14). ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design. Proceedings of the ECCV 2018: 15th European Conference, Munich, Germany.10.1007/978-3-030-01264-9_8 ↗