DL-DEIM: An Efficient and Lightweight Detection Framework with Enhanced Feature Fusion for UAV Object Detection
Source abstract
Abstract
UAV object detection is still difficult to achieve due to large-scale variation, dense small objects, a complicated background, and resource constraints from onboard computing. To solve these problems, we develop a diffusion-enhanced detection network, DL-DEIM, tailored for aerial images. The proposed scheme generalizes the DEIM baseline across three orthogonal axes. First, we propose a lightweight backbone network called DCFNet, which utilizes a DRFD module and a FasterC3k2 module to maintain spatial information and reduce computational complexity. Second, we propose a LFDPN module, which can conduct bidirectional multi-scale fusion via frequency-spatial self-attention and deep feature refinement and largely enhance cross-scale contextual propagation for small objects. Third, we propose LAWDown, an adaptive-content-aware downsampling to preserve the discriminative representation with higher accuracy at lower resolutions, which can effectively capture the spatially-variant weights and group channel interactions. On the VisDrone2019 dataset, DL-DEIM achieves a mAP@0.5 of 34.9% and a mAP@0.5:0.95 of 20.0%, outperforming the DEIM baseline by +4.6% and +2.9%, respectively. The model maintains real-time inference speed (356 FPS) with only 4.64 M parameters and 11.73 GFLOPs. Ablation studies validate the fact that DCFNet, LFDPN, and LAWDown collaboratively contribute to the accuracy and efficiency. Visualizations also display clustered and better localized activation in crowded scenes. These results show that DL-DEIM achieves a good tradeoff between detection probability and computation burden and it can be used in practice on resource-limited UAV systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Neural Network Applications · UAV Applications and Optimization · Domain Adaptation and Few-Shot Learning
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Aviation, Inner Mongolia University of Technology, Hohhot 010051, China
Funding
Department of Education Project · Inner Mongolia “14th Five-Year Plan” Key Research and Development and Achievement Transformation Plan Project in the Field of Social Welfare · Inner Mongolia University Scientifc Research Project
References (42)
Show source-provided references
1. Real-time disaster response using UAV-based object detection with lightweight neural networks
2. Urban traffic monitoring and analysis using drone-captured imagery with deep learning
3. Vision meets drones: Past, present and future
4. E3-UAV: An Edge-Based Energy-Efficient Object Detection System for Unmanned Aerial Vehicles
5. Object Detection in Aerial Images: A Large-Scale Benchmark and Challenges10.1109/TPAMI.2021.3117983 ↗
6. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23–28). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.81 ↗
7. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition10.1109/TPAMI.2015.2389824 ↗
8. Girshick, R. (2015, January 7–13). Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile.10.1109/ICCV.2015.169 ↗
9. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks10.1109/TPAMI.2016.2577031 ↗
10. He, K., Gkioxari, G., Dollár, P., and Girshick, R. (2017, January 22–29). Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.10.1109/ICCV.2017.322 ↗
11. Redmon, J., Divvala, S., Girshick, R., 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, Las Vegas, NV, USA.10.1109/CVPR.2016.91 ↗
12. Redmon, J., and Farhadi, A. (2017, January 21–26). YOLO9000: Better, Faster, Stronger. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.690 ↗
13. Redmon, J., and Farhadi, A. (2018, April 08). YOLOv3: An Incremental Improvement. Available online: https://arxiv.org/abs/1804.02767.
14. Bochkovskiy, A., Wang, C.Y., and Liao, H.Y.M. (2020, April 23). YOLOv4: Optimal Speed and Accuracy of Object Detection. Available online: https://arxiv.org/abs/2004.10934.
15. Jocher, G. (2020, June 25). YOLOv5. Available online: https://github.com/ultralytics/yolov5.
16. Tripathi, R., Patel, I., Tushar, M., and Khandelwal, V. (2020, August 21). ASAP-NMS: Accelerating Non-Maximum Suppression Using Spatially Aware Priors. Available online: https://arxiv.org/abs/2007.09785.
17. Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., and Zagoruyko, S. (2020, January 23–28). End-to-End Object Detection with Transformers. Proceedings of the European Conference on Computer Vision, Glasgow, UK.10.1007/978-3-030-58452-8_13 ↗
18. Liu, S., Ren, T., Chen, J., Zeng, Z., Zhang, H., Li, F., Li, H., Huang, J., Su, H., and Zhu, J. (2023, January 2–6). Detection Transformer with Stable Matching. Proceedings of the IEEE/CVF International Conference on Computer Vision, Paris, France.10.1109/ICCV51070.2023.00597 ↗
19. Huang, S., Yu, X., Liu, L., Xie, H., Li, F., and Wang, Z. (2024, December 05). DEIM: DETR with Improved Matching for Fast Convergence. Available online: https://arxiv.org/abs/2412.04234.
20. Inference latency prediction for CNNs on heterogeneous mobile devices and ML frameworks
21. Multi-scale feature extraction for energy-efficient object detection in remote sensing images10.1049/cvi2.12317 ↗
22. Shi, W., Zhang, J., Fu, Y., Chen, D., Zhu, J., and Lv, C. (2025). Lightweight detection and segmentation of crayfish parts using an improved YOLOv11n segmentation model. Sci. Rep., 15.10.1038/s41598-025-11201-9 ↗
23. Weng, T., and Niu, X. (2025). Enhancing UAV object detection in low-light conditions with ELS-YOLO: A lightweight model based on improved YOLOv11. Sensors, 25.10.20944/preprints202506.2235.v1 ↗
24. Yang, Y., Yang, S., and Chan, Q. (2025). LEAD-YOLO: A Lightweight and Accurate Network for Small Object Detection in Autonomous Driving. Sensors, 25.10.3390/s25154800 ↗
25. HCRP-YOLO: A lightweight algorithm for potato defect detection10.1016/j.atech.2025.100849 ↗
26. Chen, J., Kao, S.-H., He, H., Zhuo, W., Wen, S., Lee, C.-H., and Chan, S.-H.G. (2023, January 17–24). Run, Don’t walk: Chasing higher FLOPS for faster neural networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.10.1109/CVPR52729.2023.01157 ↗
27. Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., Ke, Z., Li, Q., Cheng, M., and Wei, X. (2022). YOLOv6: A single-stage object detection framework for industrial applications. arXiv.
28. A Robust Feature Downsampling Module for Remote-Sensing Visual Tasks10.1109/TGRS.2023.3282048 ↗
29. Sun, Y., Xu, C., and Yang, J. (October, January 29). Frequency-spatial entanglement learning for camouflaged object detection. Proceedings of the European Conference on Computer Vision, Milan, Italy.
30. Zhang, Y., Zhou, S., and Li, H. (2024, January 16–22). Depth information assisted collaborative mutual promotion network for single image dehazing. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.00275 ↗
31. Yang, Z., Guan, Q., Yu, Z., Xu, X., Long, H., Lian, S., Hu, H., and Tang, Y. (2025). Mhaf-yolo: Multi-branch heterogeneous auxiliary fusion yolo for accurate object detection. arXiv.10.2139/ssrn.5231601 ↗
32. Du, D., Zhu, P., Wen, L., Bian, X., Lin, H., Hu, Q., Peng, T., Zheng, J., Wang, X., and Zhang, Y. (November, January 27). VisDrone-DET2019: The Vision Meets Drone Object Detection in Image Challenge Results. Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops, Seoul, Korea.
33. Howard, A., Sandler, M., Chen, B., Wang, W., Chen, L.C., Tan, M., Chu, G., Vasudevan, V., Zhu, Y., and Pang, R. (November, January 27). Searching for MobileNetV3. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Korea.
34. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
35. Tan, M., and Le, Q. (2019, January 9–15). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the 36th International Conference on Machine Learning, Long Beach, CA, USA.
36. 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 European Conference on Computer Vision (ECCV), Munich, Germany.10.1007/978-3-030-01264-9_8 ↗
37. Jocher, G., Chaurasia, A., and Qiu, J. (2025, August 10). Ultralytics YOLO (Version 8.0.0). Available online: https://github.com/ultralytics/ultralytics.
38. Zhao, Y., Lv, W., Xu, S., Wei, J., Wang, G., Dang, Q., Liu, Y., and Chen, J. (2024, January 17–21). DETRs Beat YOLOs on Real-time Object Detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.01605 ↗
39. Wang, C.Y., Yeh, I.H., and Liao, H.Y.M. (2024). YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information. arXiv.10.1007/978-3-031-72751-1_1 ↗
40. 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.
41. Ultralytics (2025, August 10). YOLO11: The Latest Advancement in State-of-the-Art Object Detection. Available online: https://docs.ultralytics.com/models/yolo11/.
42. Peng, Y., Li, H., Wu, P., Zhang, Y., Sun, X., and Wu, F. (2024). D-FINE: Redefine Regression Task in DETRs as Fine-Grained Distribution Refinement. arXiv.