US 11,816,854 B2
Image processing apparatus and image processing method
Hiroki Mizuno, Tokyo (JP)
Assigned to SONY GROUP CORPORATION, Tokyo (JP)
Appl. No. 17/598,540
Filed by SONY GROUP CORPORATION, Tokyo (JP)
PCT Filed Mar. 4, 2020, PCT No. PCT/JP2020/009154
§ 371(c)(1), (2) Date Sep. 27, 2021,
PCT Pub. No. WO2020/208976, PCT Pub. Date Oct. 15, 2020.
Claims priority of application No. 2019-073298 (JP), filed on Apr. 8, 2019.
Prior Publication US 2022/0180545 A1, Jun. 9, 2022
Int. Cl. G06T 17/20 (2006.01); G06T 7/579 (2017.01)
CPC G06T 7/579 (2017.01) [G06T 17/20 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/10028 (2013.01)] 15 Claims
OG exemplary drawing
 
1. An image processing apparatus, comprising:
circuitry configured to:
acquire a depth map based on a captured image of a depth camera;
generate, based on the depth map, a depth-map-based TSDF space (TSDF Volume);
determine, based on a reliability value, a low-reliability depth region and a high-reliability depth region of the depth-map-based TSDF space, wherein
the reliability value corresponds to a pixel value associated with each pixel unit of the captured image of the depth camera, and
the reliability value of the low-reliability depth region is lower than the reliability value of the high-reliability depth region;
acquire a visible light image from a visible light camera;
partially generate, based on the determination of the low-reliability depth region and the high-reliability depth region, a visible-light-image-based TSDF space for the low-reliability depth region by a visual volume intersection method (Visual Hull);
generate an integrated TSDF space by integration of the depth-map-based TSDF space and the visible-light-image-based TSDF space corresponding to the low-reliability depth region; and
analyze a three-dimensional shape of a subject in the integrated TSDF space.