Using machine learning to detect a constituent image within a composite image

摘要

A system and methods are disclosed for training a machine learning model to identify constituent images within composite images. A combined image 203 is generated, where the compound image 202 comprises a first portion containing pixel data of a first constituent image, and a second region containing pixel information related to a second picture component. A first training input comprising pixel data of the composite image and a first target output for the first training input are created, where the first target output 204 identifies a position of the first portion within the composite image. The training dataset is provided to train the machine learning model on (i) a set of training inputs comprising the first training input and (ii) a set of target outputs comprising the first target output. The first portion of the composite image may surround the second region and said images may be video frames. The method is subsequently used to analyze further images to extract confidence level related to the presence and/or localization of constituent images within composite pictures (Figure 4).