US 12,272,119 B2
Adaptive image classification
Victor Wang, San Diego, CA (US); Yumin Shen, Fremont, CA (US); and Zayra Lobo, Sunnyvale, CA (US)
Assigned to GM Cruise Holdings LLC, San Francisco, CA (US)
Filed by GM Cruise Holdings LLC, San Francisco, CA (US)
Filed on Nov. 29, 2022, as Appl. No. 18/071,255.
Prior Publication US 2024/0177452 A1, May 30, 2024
Int. Cl. G06V 20/58 (2022.01); G06T 7/11 (2017.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 20/56 (2022.01)
CPC G06V 10/764 (2022.01) [G06V 10/774 (2022.01); G06V 20/56 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
accessing raw image data of an image gathered by a sensor associated with an autonomous vehicle (AV) during operation of the AV;
applying a first stage of a two-stage classifier to the raw image data to generate a classification output of the first stage of the two-stage classifier, wherein the first stage of the two-stage classifier is trained by first raw AV data captured at varying values of one or more capture parameters associated with operating one or more sensors of the AV in capturing the first raw AV data; and
applying a second stage of the two-stage classifier to the raw image data to generate a final classification output of the raw image data, wherein:
the second stage of the two-stage classifier is formed by a plurality of image calibration classifiers trained by second raw AV data at varying values of one or more image calibration parameters; and
an image calibration classifier of the plurality of image calibration classifiers is selected based on the classification output of the first stage of the two-stage classifier and applied to the raw image data as part of applying the second stage of the two-stage classifier to generate the final classification output of the raw image data.