Scholarlink

The cutting-edge computer vision framework SignShapeNet presents an innovative YOLO-NAS (You Only Look One-Neural Architecture Search) method for effective and precise shape detection. SignShapeNet distinguishes itself by streamlining neural network designs and guaranteeing top performance in shape detection tasks by utilising traffic sign board images as datasets and applying the YOLO-NAS method. This new technique emphasises the significance of efficiency, a vital component in real-world applications. SignShapeNet reduces computing complexity by finding and optimising its neural architecture on its own, which makes it ideal for deployment on resource-constrained edge devices.

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

SignShapeNet: A YOLO-NAS Approach for Efficient Shape Detection in Traffic Signboard

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kirtan Matalia,
  • Muskan C. Dave,
  • Ronak R. Patel

摘要

The cutting-edge computer vision framework SignShapeNet presents an innovative YOLO-NAS (You Only Look One-Neural Architecture Search) method for effective and precise shape detection. SignShapeNet distinguishes itself by streamlining neural network designs and guaranteeing top performance in shape detection tasks by utilising traffic sign board images as datasets and applying the YOLO-NAS method. This new technique emphasises the significance of efficiency, a vital component in real-world applications. SignShapeNet reduces computing complexity by finding and optimising its neural architecture on its own, which makes it ideal for deployment on resource-constrained edge devices.