US 12,437,556 B1
Accelerated non-maximum suppression in machine learning applications
Zhimeng Fan, Shanghai (CN)
Assigned to NVIDIA Corporation, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Apr. 16, 2024, as Appl. No. 18/636,504.
Application 18/636,504 is a continuation of application No. 17/500,167, filed on Oct. 13, 2021, granted, now 11,989,948.
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/58 (2022.01); G06F 7/24 (2006.01); G06F 7/499 (2006.01); G06F 18/2431 (2023.01)
CPC G06V 20/58 (2022.01) [G06F 7/24 (2013.01); G06F 7/49915 (2013.01); G06F 18/2431 (2023.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus comprising:
one or more processing units to:
identify a first set of confidence scores associated with a plurality of bounding boxes corresponding to one or more objects at least partially depicted in one or more images;
transform the confidences scores of the first set of confidence scores into a second set of confidence scores associated with the plurality of bounding boxes so the confidence scores of the second set of confidence scores are sortable with less than all bits;
sort the confidence scores of the second set of confidence scores with less than all bits and in a first order; and
perform a first suppression operation on the second set of confidence scores in the first order to remove one or more redundant bounding boxes from the plurality of bounding boxes.