Scholarlink

<p>Object detection is a technology that automatically identifies and locates specific objects in images or videos and plays a core role in various fields, such as autonomous driving, security surveillance, and medical imaging. You Only Look Once (YOLO) has gained attention for achieving both accuracy and detection speed in real-time applications; however, during the resolution reduction process, detailed information is lost, and unnecessary signals are mixed in the multi-scale feature fusion stage, resulting in limited detection performance for small objects and complex background scenes. To alleviate these limitations, we propose YOLO-RECAP, which integrates a Content-Aware ReAssembly of FEatures (CARAFE) and Efficient Channel Attention (ECA) modules based on YOLOv11. CARAFE precisely restores boundaries and shapes during the upsampling stage by utilizing position-specific content information, whereas ECA effectively models interchannel interactions to emphasize important signals. For performance verification, VisDrone2019, Store Keeping Unit-110&#xa0;K (SKU-110&#xa0;K), Pascal Visual Object Classes (VOC), and Dataset for Object Detection in Aerial Images (DOTA)v1 were used. In addition, Latency is reported under an end-to-end setting that includes pre-processing, inference, and post-processing, to reflect practical deployment conditions. As a result, YOLO-RECAP achieved mAP50 of 0.316, mAP50@95 of 0.184, Latency of 16.5&#xa0;ms, and 60.6 FPS on VisDrone2019; mAP50 of 0.895, mAP50@95 of 0.572, Latency of 18.2&#xa0;ms, and 55.0 FPS on SKU-110&#xa0;K; mAP50 of 0.770, mAP50@95 of 0.561, Latency of 15.6&#xa0;ms, and 63.9 FPS on Pascal VOC; and mAP50 of 0.281, mAP50@95 of 0.157, Latency of 18.6&#xa0;ms, and 53.9 FPS on DOTAv1. Qualitative bounding-box visualizations further indicate reduced missed detections and more stable predictions in cluttered and densely populated scenes. As a result, YOLO-RECAP provided a more stable and balanced detection performance than the existing YOLOv11 and recent detection models, especially for small objects and complex backgrounds. This code is available at <a href="https://github.com/Heon-ju/YOLO-RECAP.git">https://github.com/Heon-ju/YOLO-RECAP.git</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

YOLO-RECAP: reassembly with channel attention for perception

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月19日

查看全文
  • Heon-Ju Kim,
  • Sung-Wook Park,
  • Chun-Bo Sim,
  • Se-Hoon Jung

摘要

Object detection is a technology that automatically identifies and locates specific objects in images or videos and plays a core role in various fields, such as autonomous driving, security surveillance, and medical imaging. You Only Look Once (YOLO) has gained attention for achieving both accuracy and detection speed in real-time applications; however, during the resolution reduction process, detailed information is lost, and unnecessary signals are mixed in the multi-scale feature fusion stage, resulting in limited detection performance for small objects and complex background scenes. To alleviate these limitations, we propose YOLO-RECAP, which integrates a Content-Aware ReAssembly of FEatures (CARAFE) and Efficient Channel Attention (ECA) modules based on YOLOv11. CARAFE precisely restores boundaries and shapes during the upsampling stage by utilizing position-specific content information, whereas ECA effectively models interchannel interactions to emphasize important signals. For performance verification, VisDrone2019, Store Keeping Unit-110 K (SKU-110 K), Pascal Visual Object Classes (VOC), and Dataset for Object Detection in Aerial Images (DOTA)v1 were used. In addition, Latency is reported under an end-to-end setting that includes pre-processing, inference, and post-processing, to reflect practical deployment conditions. As a result, YOLO-RECAP achieved mAP50 of 0.316, mAP50@95 of 0.184, Latency of 16.5 ms, and 60.6 FPS on VisDrone2019; mAP50 of 0.895, mAP50@95 of 0.572, Latency of 18.2 ms, and 55.0 FPS on SKU-110 K; mAP50 of 0.770, mAP50@95 of 0.561, Latency of 15.6 ms, and 63.9 FPS on Pascal VOC; and mAP50 of 0.281, mAP50@95 of 0.157, Latency of 18.6 ms, and 53.9 FPS on DOTAv1. Qualitative bounding-box visualizations further indicate reduced missed detections and more stable predictions in cluttered and densely populated scenes. As a result, YOLO-RECAP provided a more stable and balanced detection performance than the existing YOLOv11 and recent detection models, especially for small objects and complex backgrounds. This code is available at https://github.com/Heon-ju/YOLO-RECAP.git.