针对PointNet模型只针对孤立点提取特征信息,而对邻域结构的信息提取能力不足的问题,提出基于图卷积网络的三维点云分类分割模型Gr......

Reading text in images automatically has become an attractive research topic in computer vision.Specifically,end-to-end ......

机坪场景下包含丰富的空间位置关系上下文信息。传统目标检测器往往只关注单一的视觉外观而忽略上下文信息;此外机坪数据集中部分......

Video summarization has established itself as a fundamental technique for generating compact and concise video, which al......

Detecting pedestrians, especially under heavy occlusion, is a challenging computer vision problem with numerous real-wor......