Scholarlink

Visual Question Answering (VQA) is a study domain that allows robots to understand and answer natural language inquiries regarding visual data. Natural language processing, commonsense reasoning and computer vision are all integrated in this complex multi-modal AI task. VQA can be used to power AI-based personal assistants and assist people with visual impairments, among other human–computer interaction scenarios. The problem is regarded as AI-complete. The system under discussion explores using dual attention techniques to improve VQA performance. The initial phase of VQA involves object detection, for which the YOLO and R-CNN algorithms have been employed. This study compares two advanced object detection algorithms YOLOv8 and R-CNN focusing on their applicability to VQA. The objective is to evaluate their strengths, weaknesses, and opportunities for enhancement to improve performance, precision, and accuracy in object detection tasks and to find out the best model.

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

A Comparison of YOLO V8 and R-CNN Algorithm—A VQA Object Detection Method Survey

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • K. Devaki,
  • S. Bakiyalakshmi

摘要

Visual Question Answering (VQA) is a study domain that allows robots to understand and answer natural language inquiries regarding visual data. Natural language processing, commonsense reasoning and computer vision are all integrated in this complex multi-modal AI task. VQA can be used to power AI-based personal assistants and assist people with visual impairments, among other human–computer interaction scenarios. The problem is regarded as AI-complete. The system under discussion explores using dual attention techniques to improve VQA performance. The initial phase of VQA involves object detection, for which the YOLO and R-CNN algorithms have been employed. This study compares two advanced object detection algorithms YOLOv8 and R-CNN focusing on their applicability to VQA. The objective is to evaluate their strengths, weaknesses, and opportunities for enhancement to improve performance, precision, and accuracy in object detection tasks and to find out the best model.