CLASSIFICATION PARALLELIZATION ARCHITECTURE

A system and method for generating composite images that include object identifiers by processing images through multiple machine learning models. A plurality of images corresponding to a video stream are captured at an unmanned vehicle. A plurality of machine learning models for processing the plur...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SUMMER, Matthew D, COOPER, Jeffrey S, TRUESDELL, Brad, WAGONER, Sean, MAKOVY, Kevin, FALENDYSZ, Andrew D, BOWMAN, William S
Format: Patent
Sprache:eng ; fre ; ger
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A system and method for generating composite images that include object identifiers by processing images through multiple machine learning models. A plurality of images corresponding to a video stream are captured at an unmanned vehicle. A plurality of machine learning models for processing the plurality of images are provided, where each machine learning model identifies an object type different from object types identified by other machine learning models of the plurality of machine learning models. A first image is input into each of the plurality of machine learning models. A subset of images are input into a second machine learning model. A last image of the subset of images is input into the first machine learning model while refraining from inputting other images of the subset of images into the first machine learning mode. Coordinates within each image of the plurality of images corresponding to each object identified in a corresponding image of the plurality of images are received from each of the machine learning models, and a set of composite images are generated from the plurality of images with a plurality of identifiers corresponding to each identified object.