A fine-grained attention and geometric correspondence model for musculoskeletal risk classification in athletes using multimodal visual and skeletal features
基于多模态视觉和骨骼特征的运动员肌肉骨骼风险分类的细粒度注意力与几何对应模型
机构 * Department of Computer Science and Engineering, United International University(计算机科学与工程系,国际联合大学) ; Department of Data Science and Artificial Intelligence, Monash University(数据科学与人工智能系,墨尔本大学) ; Faculty of Science and Technology, Charles Darwin University(科学与技术学院,查尔斯达尔文大学) ; Applied Artificial Intelligence and Intelligent Systems (AAIINS) Laboratory, Dhaka(应用人工智能与智能系统实验室,达卡)
专题命中 多模态评测 :multimodal(title,abstract);cross-modal(abstract);分类 cs.CV
AI总结 提出ViSK-GAT多模态框架,融合图像和骨骼坐标特征,通过细粒度注意力模块和几何对应模块实现运动员肌肉骨骼风险八级分类,关键指标超93%。
Comments Published in Computers and Electrical Engineering
Journal ref Computers and Electrical Engineering, Vol. 138, 111281, 2026