	US 11,854,211 B2
	Training multi-object tracking models using simulation
Ishani Chakraborty, Seattle, WA (US); Jonathan C. Hanzelka, Kenmore, WA (US); Lu Yuan, Redmond, WA (US); Pedro Urbina Escos, Seattle, WA (US); and Thomas M. Soemo, Redmond, WA (US)
Assigned to Microsoft Technology Licensing, LLC., Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Jan. 26, 2022, as Appl. No. 17/585,053.
Application 17/585,053 is a continuation of application No. 17/026,084, filed on Sep. 18, 2020, granted, now 11,335,008.
Prior Publication US 2022/0148197 A1, May 12, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/20 (2017.01); G06N 20/00 (2019.01); G06T 17/00 (2006.01); G06T 19/20 (2011.01)

CPC G06T 7/20 (2013.01) [G06N 20/00 (2019.01); G06T 17/00 (2013.01); G06T 19/20 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30196 (2013.01); G06T 2207/30241 (2013.01); G06T 2219/2004 (2013.01)]

20 Claims

1. A system for multi-object tracking, the system comprising:

a processor; and

a computer-readable medium storing instructions that are operative upon execution by the processor to:

generate a plurality of training images based at least on scene generation information, each training image comprising a plurality of objects to be tracked;

generate, for each training image, original simulated data based at least on the scene generation information, the original simulated data comprising tag data for a first object of the plurality of objects;

locate, within the original simulated data for a first training image of the plurality of training images, tag data for the first object, based at least on an anomaly alert associated with the first object in the first training image; and

based at least on locating the tag data for the first object, modify at least a portion of the tag data for the first object from the original simulated data for the first training image, thereby generating preprocessed training data from the original simulated data.