US 12,322,123 B2
Visual object detection in a sequence of images
Volodya Grancharov, Solna (SE); Sigurdur Sverrisson, Kungsängen (SE); Alfredo Fanghella, Stockholm (SE); and Manish Sonal, Sollentuna (SE)
Assigned to TELEFONAKTIEBOLAGET LM ERICSSON (PUBL), Stockholm (SE)
Appl. No. 17/774,455
Filed by Telefonaktiebolaget LM Ericsson (publ), Stockholm (SE)
PCT Filed Nov. 7, 2019, PCT No. PCT/EP2019/080498
§ 371(c)(1), (2) Date May 4, 2022,
PCT Pub. No. WO2021/089147, PCT Pub. Date May 14, 2021.
Prior Publication US 2022/0392195 A1, Dec. 8, 2022
Int. Cl. G06T 7/33 (2017.01); G06V 10/20 (2022.01); G06V 10/46 (2022.01); G06V 10/764 (2022.01); G06V 20/40 (2022.01)
CPC G06T 7/33 (2017.01) [G06V 10/255 (2022.01); G06V 10/462 (2022.01); G06V 10/764 (2022.01); G06V 20/46 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method for visual object detection in a sequence of images, the method being performed by a visual object detector, the method comprising:
obtaining a sequence of images of a scene, the sequence of images at least comprising a current image of the scene and a previous image of the scene;
extracting a set of objects from the sequence of images, wherein the extracting comprises:
obtaining an image-wise descriptor classification score for the previous image or for the current image and pertaining to which type of content the scene comprises;
obtaining an image overlapping score pertaining to how much overlap in image area there is between the previous image and the current image;
based on the image-wise descriptor classification score and the image overlapping score, determining whether a criteria is met; and
as a result of determining that the criteria is met, performing visual object detection in at least part of the current image; and
constructing an image representation of the scene using the extracted set of objects.