作者: Roux , McKeown
关键词: Pattern recognition 、 Artificial intelligence 、 Mathematics 、 Computational geometry 、 Feature extraction 、 Computer vision 、 Feature matching 、 Aerial imagery 、 Multiple view 、 Process (computing) 、 Matching (statistics) 、 Oblique case
摘要: We present a method to generate three-dimensional building hypotheses starting from sparse features, hypothesized corners, extracted multiple views of the scene. This technique relies on knowledge about imaging geometry and acquisition parameters provide rigorous geometric constraints for matching process. The effectiveness this approach is demonstrated using complex aerial imagery taken with highly oblique containing buildings flat or peaked roofs. >