US 12,249,077 B2
Consistency measure for image segmentation processes
Yizhe Zhang, San Diego, CA (US); and Fatih Murat Porikli, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Nov. 11, 2021, as Appl. No. 17/524,253.
Claims priority of provisional application 63/113,725, filed on Nov. 13, 2020.
Prior Publication US 2022/0156943 A1, May 19, 2022
Int. Cl. G06T 7/174 (2017.01); G06N 3/04 (2023.01); G06T 7/11 (2017.01); G06T 7/194 (2017.01)
CPC G06T 7/174 (2017.01) [G06N 3/04 (2013.01); G06T 7/11 (2017.01); G06T 7/194 (2017.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 30 Claims
OG exemplary drawing
 
1. An apparatus for determining consistency measures for image segmentation processes, the apparatus comprising:
at least one memory; and
at least one processor coupled to the at least one memory, the at least one processor configured to:
determine a first segmentation feature associated with a first segmentation mask of a first image frame of a plurality of image frames;
determine a second segmentation feature associated with a second segmentation mask of a second image frame of the plurality of image frames;
determine the second segmentation feature corresponds to the first segmentation feature;
determine a first location of the first segmentation feature associated with the first segmentation mask;
determine a second location of the second segmentation feature associated with the second segmentation mask;
determine a first location within the first image frame or a first feature map associated with the first image frame that corresponds to the first location of the first segmentation feature;
determine a second location within the second image frame or a second feature map associated with the second image frame that corresponds to the second location of the second segmentation feature;
determine a first image feature at the first location within the first image frame or the first feature map and determine a second image feature at the second location within the second image frame or the second feature map;
based on the determination that the second segmentation feature corresponds to the first segmentation feature, determine a first similarity measurement between the first image feature and the second image feature; and
determine a temporal consistency measurement associated with the first image frame and the second image frame based at least in part on the first similarity measurement.