作者: Phan Raymond , Rzeszutek Richard Jeffrey , Lam Joseph , Yu Yuanhao
DOI:
关键词: Point cloud 、 Field of view 、 Artificial intelligence 、 Set (abstract data type) 、 Computer vision 、 Structure (mathematical logic) 、 Image (mathematics) 、 Computer science
摘要: A method of determining a support structure depth having front and back separated by the includes: obtaining point cloud structure, mask indicating, for plurality portions an image captured from capture pose, respective confidence levels that depict structure; selecting, cloud, initial set points located within field view originating at pose; points, unoccluded subset measurements, measurements in corresponding to coordinates; retrieving, mask, level each subset; based on retrieved levels, depth.