文章目录 一、背景二、方法2.1 框架总览2.1.1 Text-based classifiers from language descriptions2.1.2 Vision-based Classifiers from Image Exemplars2.1.3 Constructing Classifiers via Multi-Modal Fusion 三、效果3.1 数据集3.2 实现细节3.3 开集目标检测结果 论文：Multi-Modal Cl 继续阅读