arXiv · 机器人与计算机视觉· Abhishek Tyagi, Tanmay Jain, Vilas Biradar, Krithi Krishna Koduri, Misba Hul Bashir Masoodi, Swetha Rapolu, Hardik Rughwani, Rachana Kotha, Anjan Kumar Kaipa, Anudeep Katrevula, Rakesh Garlapati, Aniruddha Pratap Singh, Rajendra Patel, Abhay Tyagi, Sundeep Lakhtakia, Mohan Ramchandani, Rakesh Kalapala, D. Nageshwar Reddy·· 9 天前AI 评分68
EndoSCOP-V:面向内镜报告与临床推理的多轮视频理解评估框架
EndoSCOP-V: A Multi-Turn Video Understanding Evaluation Framework for Multimodal Models in Endoscopy Reporting and Clinical Reasoning
AI 导读
针对现有内镜基准仅测试静态图像或单病种视频的局限,作者提出 EndoSCOP-V 多轮视频理解评估框架。该框架包含 1,307 个真实内镜及结肠镜视频(共 63.2 小时),其 20,274 问答对源自医生填写的结构化电子病历,覆盖 54 种疾病及 9 大任务分类。
来源:arXiv · 机器人与计算机视觉 · arxiv.org