Motorola Solutions, Inc.

Aravind Anantha

Mahesh Saptharishi

Yanyan Hu

Patrick Beaulieu

Daniel Booth

Jason Verelli

Reducing false negatives and finding new classes in object detectors is disclosed. Also disclosed is a method that includes grouping together a plurality of cropped image portions from region proposals based on image properties. The method also includes receiving first user input that either establishes that the grouped together, cropped image portions relate to uninteresting objects or establishes that the grouped together, cropped image portions relate to an object class of interesting objects. The method also includes obtaining, only when the grouped together, cropped image portions relate to the object class of interesting objects, second user input that includes an object label corresponding to the object class of interesting objects.

Reducing false negatives and finding new classes in object detectors

Methods, systems, and techniques for facilitating improved training of a supervised machine learning process, such as a decision tree. First and second object detections of an object depicted in a video are respectively generated using first and second object detectors, with the second object detector requiring more computational resources than the first object detector to detect the object. Whether a similarity and a difference between the first and second object detections respectively satisfy a similarity threshold and a difference threshold is determined. When the similarity threshold is satisfied, the first object detection is stored as a positive example for the machine learning training. When the difference threshold is satisfied, the first object detection is stored as a negative example for the machine learning training.

Growing community of inventors

Malden, MA, United States of America

Aravind Anantha