专利检索 ap:("TOYOTA JIDOSHA KABUSHIKI KAISHA" OR "Katholieke Universiteit Leuven KU Leuven Research & Development") AND inv:"DE WOLF, Mark" 第 1 页

1.

发明公开
OBJECT POSE ESTIMATION FROM AN RGB IMAGE BY A NEURAL NETWORK OUTPUTTING A PARAMETER MAP 审中-公开

公开(公告)号：EP3869392A1

公开(公告)日：2021-08-25

申请号：EP20158613.8

申请日：2020-02-20

申请人： TOYOTA JIDOSHA KABUSHIKI KAISHA , Katholieke Universiteit Leuven KU Leuven Research & Development

发明人： ABBELOOS, Wim , OLMEDA REINO, Daniel , ABDELKAWY, Hazem , HEYLEN, Jonas , DE WOLF, Mark , DAWAGNE, Bruno , BARNES, Michael , LEMKENS, Wim , PROESMANS, Marc , VAN GOOL, Luc

IPC分类号： G06K9/00 , G06K9/46 , G06T7/70

摘要： A method for processing at least one image comprising inputting the image to at least one neural network (ANN1, ANN2), the at least one network being configured to deliver, for each pixel of a group of pixels belonging to an object of a given type visible on the image, an estimation of parameters of the object,
the method further comprising processing the estimations of the parameters of the object using an instance segmentation mask identifying instances of objects having the given type.