Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
OBJECT FEATURE POINT DETECTION DEVICE
Document Type and Number:
WIPO Patent Application WO/2023/188908
Kind Code:
A1
Abstract:
An object feature point detection device according to an embodiment of the present invention includes a detection model (64), a learning unit (60), and a parameter optimization unit (65). The detection model (64) outputs estimation data including an estimated position for each feature point included in each of a plurality of objects in an inputted image. The learning unit (60) executes the detection model (64) and machine learning of the detection model (64). The parameter optimization unit (65) optimizes a parameter for estimating the position of each feature point of any object among the plurality of objects in the estimation data outputted from the detection model (64) to which a training image obtained by imaging the plurality of objects has been inputted. The detection model (64) uses the parameter optimized by the parameter optimization unit (65) to output estimation data including an estimated position of each feature point included in each of a plurality of objects in a newly inputted image.

Inventors:
TOBETA MASAKAZU (JP)
Application Number:
PCT/JP2023/005012
Publication Date:
October 05, 2023
Filing Date:
February 14, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
AISIN CORP (JP)
International Classes:
G06T7/00; G06T7/70
Foreign References:
JP2021086322A2021-06-03
Other References:
HARUTA, RYUSUKE; IRIE, TOSHIHIRO: " Forecasting Human Motion for Motion Planning: Optimizer Comparation and Consideration of 3D Datasets. ", PROCEEDINGS OF THE 37TH ANNUAL CONFERENCE OF THE INSTITUTE OF IMAGE ELECTRONICS ENGINEERS OF JAPAN 2009; JUNE 25, 2009 - JUNE 26, 2009, 26 June 2021 (2021-06-26) - 26 June 2021 (2021-06-26), pages 1 - 4, XP009549584, ISSN: 2436-4371, DOI: 10.11371/aiieej.49.0_28
Attorney, Agent or Firm:
SAKAI INTERNATIONAL PATENT OFFICE (JP)
Download PDF: