DTA-Head: Dynamic Task Alignment Head for Regression and Classification in Small Object Detection
Source abstract
Abstract
Detecting small targets poses significant challenges due to the limited feature information and the aggregation of features in deep feature maps. Existing single-stage detectors handle classification and regression separately, leading to inconsistent predictions and potential filtering by Non-Maximum Suppression (NMS). To address these issues, we propose the Dynamic Task Alignment (DTA) Head. This novel approach comprises two branches: regression and classification. The regression branch computes offsets and masks for feature alignment adopted by Deformable ConvNets v2 (DCNv2), while the classification branch enhances feature interaction through dynamic selection. The task-decomposition module separates features for each branch. Additionally, we introduce Diverse-Scale Channel-Specific Convolution (DSCSC) to apply diverse convolutions across specific channels and exchange channel information. Our methods achieved an AP@.5 of 30.9% on the TinyPerson dataset, a 3.3% improvement over the original model’s 27.6% and outperforming other common models.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Neural Network Applications · Machine Learning and Data Classification · Neural Networks and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
The Ocean College, Zhejiang University, Zhoushan 316021, China
The College of Electrical Engineering, Zhejiang University, Hangzhou 310058, China
The Hainan Institute, Zhejiang University, Sanya 572025, China
Funding
National Natural Science Foundation of China · State Grid Corporation of China Technology Project · National Natural Science Foundation of Zhejiang Province · Research Startup Funding from Hainan Institute of Zhejiang University · Sanya Science and Technology Innovation Project
References (39)
Show source-provided references
1. Duan, K., Bai, S., Xie, L., Qi, H., Huang, Q., and Tian, Q. (November, January 27). Centernet: Keypoint triplets for object detection. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea.
2. Yu, X., Gong, Y., Jiang, N., Ye, Q., and Han, Z. (2020, January 1–5). Scale match for tiny person detection. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Snowmass, CO, USA.10.1109/WACV45572.2020.9093394 ↗
3. 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 ↗
4. Ren, S., He, K., Girshick, R., and Sun, J. (2016). Faster r-cnn: Towards real-time object detection with region proposal networks. arXiv.10.1109/TPAMI.2016.2577031 ↗
5. Spatial pyramid pooling in deep convolutional networks for visual recognition10.1109/TPAMI.2015.2389824 ↗
6. Dai, J., Li, Y., He, K., and Sun, J. (2016). R-fcn: Object detection via region-based fully convolutional networks. arXiv.
7. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., and Berg, A.C. (2016, January 11–14). Ssd: Single shot multibox detector. Proceedings of the Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands. Proceedings, Part I 14.10.1007/978-3-319-46448-0_2 ↗
8. Cai, Z., Fan, Q., Feris, R.S., and Vasconcelos, N. (2016, January 11–14). A unified multi-scale deep convolutional neural network for fast object detection. Proceedings of the Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands. Proceedings, Part IV 14.10.1007/978-3-319-46493-0_22 ↗
9. Bell, S., Zitnick, C.L., Bala, K., and Girshick, R. (2016, January 27–30). Inside-outside net: Detecting objects in context with skip pooling and recurrent neural networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.314 ↗
10. Lin, T.Y., Dollár, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017, January 21–26). Feature pyramid networks for object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.106 ↗
11. Liang, Z., Shao, J., Zhang, D., and Gao, L. (2018, January 21–22). Small object detection using deep feature pyramid networks. Proceedings of the Advances in Multimedia Information Processing–PCM 2018: 19th Pacific-Rim Conference on Multimedia, Hefei, China. Proceedings, Part III 19.10.1007/978-3-030-00764-5_51 ↗
12. FCOS: A simple and strong anchor-free object detector
13. Zhang, S., Chi, C., Yao, Y., Lei, Z., and Li, S.Z. (2020, January 13–19). Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00978 ↗
14. Foveabox: Beyound anchor-based object detection10.1109/TIP.2020.3002345 ↗
15. Zhu, C., He, Y., and Savvides, M. (2019, January 15–20). Feature selective anchor-free module for single-shot object detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.10.1109/CVPR.2019.00093 ↗
16. Zhu, C., Chen, F., Shen, Z., and Savvides, M. (2020, January 23–28). Soft anchor-point object detection. Proceedings of the Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK. Proceedings, Part IX 16.10.1007/978-3-030-58545-7_6 ↗
17. Zand, M., Etemad, A., and Greenspan, M. (2022, January 23–27). Objectbox: From centers to boxes for anchor-free object detection. Proceedings of the European Conference on Computer Vision, Tel Aviv, Israel.10.1007/978-3-031-20080-9_23 ↗
18. Hu, J., Shen, L., and Sun, G. (2018, January 18–23). Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00745 ↗
19. Woo, S., Park, J., Lee, J.Y., and Kweon, I.S. (2018, January 8–14). Cbam: Convolutional block attention module. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.10.1007/978-3-030-01234-2_1 ↗
20. Wang, Q., Wu, B., Zhu, P., Li, P., Zuo, W., and Hu, Q. (2020, January 13–19). ECA-Net: Efficient channel attention for deep convolutional neural networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.01155 ↗
21. Hou, Q., Zhou, D., and Feng, J. (2021, January 20–25). Coordinate attention for efficient mobile network design. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA.10.1109/CVPR46437.2021.01350 ↗
22. Cai, X., Lai, Q., Wang, Y., Wang, W., Sun, Z., and Yao, Y. (2024, January 16–22). Poly kernel inception network for remote sensing detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.02617 ↗
23. Wang, C., He, W., Nie, Y., Guo, J., Liu, C., Wang, Y., and Han, K. (2023). Gold-YOLO: Efficient object detector via gather-and-distribute mechanism. arXiv.
24. Dai, X., Chen, Y., Xiao, B., Chen, D., Liu, M., Yuan, L., and Zhang, L. (2021, January 20–25). Dynamic head: Unifying object detection heads with attentions. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA.10.1109/CVPR46437.2021.00729 ↗
25. Zhu, X., Hu, H., Lin, S., and Dai, J. (2019, January 15–20). Deformable convnets v2: More deformable, better results. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.10.1109/CVPR.2019.00953 ↗
26. Zhu, X., Su, W., Lu, L., Li, B., Wang, X., and Dai, J. (2020). Deformable detr: Deformable transformers for end-to-end object detection. arXiv.
27. 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 applications. arXiv.
28. Han, K., Wang, Y., Tian, Q., Guo, J., Xu, C., and Xu, C. (2020, January 13–19). Ghostnet: More features from cheap operations. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00165 ↗
29. Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., Ke, Z., Li, Q., Cheng, M., and Nie, W. (2022). YOLOv6: A single-stage object detection framework for industrial applications. arXiv.
30. 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 ↗
31. Ma, X., Dai, X., Bai, Y., Wang, Y., and Fu, Y. (2024, January 16–24). Rewrite the Stars. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.00544 ↗
32. Fan, Q., Huang, H., Chen, M., Liu, H., and He, R. (2024, January 16–24). Rmt: Retentive networks meet vision transformers. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.00539 ↗
33. Zhao, Y., Lv, W., Xu, S., Wei, J., Wang, G., Dang, Q., Liu, Y., and Chen, J. (2024, January 16–24). 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 ↗
34. Chen, L., Gu, L., Zheng, D., and Fu, Y. (2024, January 16–24). Frequency-Adaptive Dilated Convolution for Semantic Segmentation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.00328 ↗
35. Wang, A., Chen, H., Lin, Z., Han, J., and Ding, G. (2024, January 16–24). Repvit: Revisiting mobile cnn from vit perspective. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.01506 ↗
36. Han, K., Wang, Y., Guo, J., and Wu, E. (2024, January 16–24). ParameterNet: Parameters Are All You Need for Large-scale Visual Pretraining of Mobile Networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.01491 ↗
37. Zheng, M., Sun, L., Dong, J., and Pan, J. (2024, January 16–24). SMFANet: A lightweight self-modulation feature aggregation network for efficient image super-resolution. Proceedings of the European Conference on Computer Vision, Seattle, WA, USA.10.1007/978-3-031-72973-7_21 ↗
38. Ding, X., Zhang, X., Han, J., and Ding, G. (2022, January 18–24). Scaling up your kernels to 31x31: Revisiting large kernel design in cnns. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.01166 ↗
39. Draelos, R.L., and Carin, L. (2020). Use HiResCAM instead of Grad-CAM for faithful explanations of convolutional neural networks. arXiv.