Can Local Vision-Language Models improve Activity Recognition over Vision Transformers? -- Case Study on Newborn Resuscitation
局部视觉-语言模型能否在视觉转换器上提高活动识别?——新生儿复苏案例研究
机构 * University of Stavanger, Dept. of electrical eng. and computer science(斯塔万格大学电气工程与计算机科学系)
专题命中 领域大模型 :language model(title,abstract);large language model(abstract)
AI总结 本研究探讨了局部视觉-语言模型在新生儿复苏视频活动识别中的应用,通过LoRA微调提升性能,达到0.91的F1分数,优于TimeSFormer基线。
Comments Presented at the Satellite Workshop on Workshop 15: Generative AI for World Simulations and Communications & Celebrating 40 Years of Excellence in Education: Honoring Professor Aggelos Katsaggelos, IEEE International Conference on Image Processing (ICIP), 2025