2023 Vol.6 Aug.N04

• Title: Small Object Detection Based on Multi Scale and Parallel FPN
• Name: Xinrong Wu, Fan Wang, Boyi Wang, Xiaopeng Hu
• Company: Dalian University of Technology,Dalian,116000 China
• Abstract:

  In scenes with small objects, there may be targets of various sizes. This poses a challenge to fully cover existing detection algorithms. To address this issue, we propose a multi-scale small object detection algorithm that fits the distribution of small objects. This algorithm uses multiple down-sampling modules to redesign the backbone network and extract features of different scales. Considering the inference delay problem of multi-scale feature fusion, we propose a dual-path parallel feature fusion algorithm. The algorithm groups multiple features of different scales. It designs a PAFPN structure within the group, and performs parallel inference between groups, which alleviates the inference delay of multi-scale feature fusion. Experimental results show that the proposed algorithm can effectively improve the performance of small target detection in the model.

  
  

• Keyword: Small object detection; Multi-scale; Parallel feature fusion;
• DOI: 10.12250/jpciams2023090615
• Citation form: Xinrong Wu.Small Object Detection Based on Multi Scale and Parallel FPN [J]. Computer Informatization and Mechanical System,2023,Vol.6,pp.66-68

[1] GIRSHICK ROSS. Fast R-CNN[C]. In Proceedings of the IEEE International Conference on Computer Vision, 2015: 1440-1448.

[2] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. Yolov4: Optimal speed and accuracy of object detection[J]. arXiv preprint, 2020.

[3] GE Z, LIU S, WANG F, et al. YoloX: Exceeding yolo series in 2021[J]. arXiv preprint, 2021.

[4] DU D, ZHU P, WEN L, et al. VisDrone-SOT2019: The Vision meets Drone Single Object Tracking Challenge Results[C]. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 199-212.

[5] CAI Z, VASCONCELOS N. Cascade R-CNN: Delving into high quality object detection[C]. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, 2018: 6154-6162.

[6] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]. In Proceedings of the IEEE International Conference on Computer Vision, 2017: 2980-2988.

[7] LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature Pyramid Networks for Object Detection[C]. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.

[8] LIU S, QI L, QIN H, et al. Path aggregation network for instance segmentation[C]. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8759-8768.