Scholarlink

This paper introduces a novel methodology that integrates a Vision Transformer with a loss function based on Complete Intersection over Union (CIOU) to achieve precise object localization and bounding box regression. Experimental results on a brain tumor dataset demonstrate that our approach improves the performance of ViT compared to alternative loss functions such as Mean Square Error (MSE), IoU, GIoU, and DGIoU.

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

Brain Tumor Localization with ViT and a Loss Function Based on Complete Intersection Over Union

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hong Cheng,
  • Rupesh Konduru

摘要

This paper introduces a novel methodology that integrates a Vision Transformer with a loss function based on Complete Intersection over Union (CIOU) to achieve precise object localization and bounding box regression. Experimental results on a brain tumor dataset demonstrate that our approach improves the performance of ViT compared to alternative loss functions such as Mean Square Error (MSE), IoU, GIoU, and DGIoU.