US 12,293,544 B2
Information processing system, information processing method, and recording medium
Hiroaki Sugawara, Tokyo (JP)
Assigned to Safie Inc., Tokyo (JP)
Filed by Safie Inc., Tokyo (JP)
Filed on Mar. 10, 2023, as Appl. No. 18/120,107.
Claims priority of application No. 2022-133355 (JP), filed on Aug. 24, 2022.
Prior Publication US 2024/0070907 A1, Feb. 29, 2024
Int. Cl. H04N 19/46 (2014.01); G06T 7/73 (2017.01); G06V 20/40 (2022.01); G06V 20/52 (2022.01); H04N 7/18 (2006.01)
CPC G06T 7/74 (2017.01) [G06V 20/40 (2022.01); G06V 20/52 (2022.01); H04N 7/181 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/30244 (2013.01); G06V 2201/07 (2022.01)] 12 Claims
OG exemplary drawing
 
1. An information processing system which provides information about a physical object shown in a video captured by a fixed network camera, wherein the fixed network camera is fixedly placed in a certain position relative to a ground surface, wherein the information processing system is configured to:
acquire first video data from the fixed network camera which is fixedly placed in the certain position; and
estimate an installation angle of the fixed network camera in the certain position based on the first video data,
when it is determined the installation angle of the fixed network camera is changed from an original installation angle of the fixed network camera to a changed installation angle different from the original installation angle due to an external factor, the information processing system is configured to:
acquire second video data from the fixed network camera which is fixedly positioned at the changed installation angle different from the original installation angle;
generate a detection model configured to detect and output the information about the physical object from the video, based on the second video data captured at the changed installation angle different from the original installation angle; and
output, by the detection model, the information about the physical object by inputting the second video data captured at the changed installation angle different from the original installation angle to the detection model,
wherein the detection model is a learning model trained to output the information about the physical object when the first or second video data is input.