Hi-GaTA: Hierarchical Gated Temporal Aggregation Adapter for Surgical Video Report Generation
Hi-GaTA:用于外科视频报告生成的分层门控时间聚合适配器
机构 * School of Engineering, College of Engineering and Physical Sciences, University of Birmingham, Birmingham, UK(英国伯明翰大学工程学院) ; School of Computer Science, University of Birmingham, Birmingham, UK(英国伯明翰大学计算机科学学院) ; Department of Applied Health Sciences, University of Birmingham, Birmingham, UK(英国伯明翰大学应用健康科学系)
专题命中 推理评测 :reasoning(abstract)
AI总结 本文提出Hi-GaTA框架,通过时间聚合压缩长视频序列生成LLM兼容的视觉前缀令牌,结合预训练的外科专用视频编码器和LoRA微调,实现高质量外科报告生成。
Comments 11 pages, 2 figures