US 11,721,099 B2
Cloud based active commissioning system for video analytics
Zhengwei Jiang, East Hartford, CT (US); Jianwei Zhao, Shanghai (CN); Zhen Jia, Shanghai (CN); and Jie Xi, Shanghai (CN)
Assigned to CARRIER CORPORATION, Palm Beach Gardens, FL (US)
Appl. No. 15/999,803
Filed by Carrier Corporation, Farmington, CT (US)
PCT Filed Feb. 17, 2017, PCT No. PCT/US2017/016766
§ 371(c)(1), (2) Date Aug. 20, 2018,
PCT Pub. No. WO2017/142736, PCT Pub. Date Aug. 24, 2017.
Claims priority of provisional application 62/297,551, filed on Feb. 19, 2016.
Prior Publication US 2021/0216774 A1, Jul. 15, 2021
Int. Cl. G06V 20/40 (2022.01); G06F 16/71 (2019.01); G06V 20/52 (2022.01); G06V 10/98 (2022.01)
CPC G06V 20/41 (2022.01) [G06F 16/71 (2019.01); G06V 10/987 (2022.01); G06V 20/40 (2022.01); G06V 20/52 (2022.01); G06V 20/44 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A method for an active commissioning system for video analytics (VA) of video surveillance, the method comprising:
capturing, using a video capture device, a video sequence;
receiving, at a VA processor, a parameter set from a cloud resource, wherein the parameter set is used for video analytics;
processing the video sequence, using the VA processor, by selecting subparts of the video sequence as events, wherein the events are detected based on the parameter set used by the VA processor;
generating a visualized report using the events from the VA processor and providing the visualized report to an operator;
marking, by the operator, yes/no on the events in the visualized report;
generating annotations for events based on the marking; and
transmitting the annotations and the events to the cloud resource for processing, wherein the cloud resource processes the annotations and events and provides a predicted parameter set as the parameter set based on the processing, the predicted parameter used as the parameter set in a next iteration of processing the video sequence;
the method further comprising:
receiving the annotations and the events in a video database of the cloud resource;
processing events with annotations with an exhaustive VA processor of the cloud resource;
generating feasible parameter sets based on the exhaustively processed events with annotations from the exhaustive VA processor;
processing video segments without annotations with a low-level feature extractor of the cloud resource and generating a low-level feature descriptor of the video sequence;
generating, using a feature parameter mapping model, possible parameter sets based on the low-level feature descriptor;
selecting the predicted parameter set based on the feasible parameter set and the possible parameter set; and
transmitting the predicted parameter set from the cloud resource to the VA processor.