US 12,283,113 B2
Method and device for determining that a detected object belongs to a specific class
Ludvig Hassbring, Lund (SE)
Assigned to AXIS AB, Lund (SE)
Filed by Axis AB, Lund (SE)
Filed on Mar. 15, 2024, as Appl. No. 18/606,543.
Claims priority of application No. 23169423 (EP), filed on Apr. 24, 2023.
Prior Publication US 2024/0355127 A1, Oct. 24, 2024
Int. Cl. G06V 20/40 (2022.01); G06T 7/223 (2017.01); G06V 20/58 (2022.01)
CPC G06V 20/58 (2022.01) [G06T 7/223 (2017.01); G06V 20/40 (2022.01); G06V 2201/07 (2022.01); G06V 2201/10 (2022.01)] 15 Claims
OG exemplary drawing
 
1. A method for determining that a detected object belongs to a specific class, the method comprising:
performing first object detection, using a first object detector, the first object detection comprising:
detecting an object in a first image frame of a video sequence, and
determining a first probability score indicating a probability that the detected object in the first image frame belongs to the specific class;
performing second object detection, using a second object detector, the second object detection comprising:
detecting an object in a second image frame of the video sequence, and
determining a second probability score indicating a probability that the detected object in the second image frame belongs to the specific class, wherein the second image frame is subsequent to the first image frame in the video sequence, wherein the first object detector has a higher object detection precision as compared to the second object detector, wherein the first object detector requires a first processing time for performing the first object detection, the first processing time being longer as compared to a second processing time required by the second object detector for performing the second object detection, and wherein the second object detection is performed in parallel with the first object detection;
on condition that the determined first probability score is below a first classification threshold and the determined second probability score is above a second classification threshold, performing at least one of reducing the first classification threshold or increasing the first probability score; and
on condition that the increased first probability score is above the first classification threshold, the first probability score is above the reduced first classification threshold, or the increased first probability score is above the reduced first classification threshold, the first object detector determining that the detected object in the first image frame belongs to the specific class.