Visual context is essential for point cloud semantic segmentation. The contextual information captures the semantic relationship between 3-D points, providing helpful hints for reasoning the category labels of points. Most current methods harness the internal ... ...

请注册登录后继续浏览