公开(公告)号：US20210216829A1

公开(公告)日：2021-07-15

申请号：US15733883

申请日：2019-05-31

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Takashi HOSONO ,  Shuhei TARASHIMA ,  Jun SHIMAMURA ,  Tetsuya KINEBUCHI

IPC: G06K9/62 ,  G06T7/13 ,  G06K9/46

Abstract: Objectness indicating a degree of accuracy of a single object is accurately estimated. An edge detection unit 30 detects an edge for a depth image, an edge density/uniformity calculation unit 40 calculates an edge density on the periphery of a candidate region, an edge density inside the candidate region, and edge uniformity on the periphery of the candidate region. An objectness calculation unit 42 calculates the objectness of the candidate region based on the edge density on the periphery of the candidate region, the edge density inside the candidate region, and the edge uniformity on the periphery of the candidate region.

公开(公告)号：US20210089827A1

公开(公告)日：2021-03-25

申请号：US16971635

申请日：2019-02-25

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Kaori KUMAGAI ,  Yukito WATANABE ,  Jun SHIMAMURA ,  Tetsuya KINEBUCHI

IPC: G06K9/62 ,  G06K9/20 ,  G06N3/04 ,  G06N3/08

Abstract: It is made possible that features in an input image used by a CNN to identify the input image are efficiently represented. With respect to a plurality of filters used on a specific layer of an CNN, an agreement rate indicating a degree of relevance of a feature map output from a filter of interest when a plurality of images in which regions representing a plurality of concepts defined in advance are already obtained are used as inputs, with a region representing a concept is calculated for each concept; an integration map in which feature maps are integrated by using the agreement rates with the concept of the filter outputting the feature maps is calculated for each concept and for each of a plurality of clusters, based on the agreement rates and a result of classification of feature maps output from the plurality of filters when an input image is input into the CNN, into each cluster based on values in the feature maps; and feature representation visualization images visualized by using threshold values that are based on values in the integration maps and information on the respective concepts corresponding the integration maps are output in pairs.