Training target detection system with generated images

The invention discloses training a target detection system with generated images, and particularly discloses devices, systems and techniques for identifying objects in images using self-supervised machine learning. In at least one embodiment, a machine learning system is trained to identify objects by training a first network to identify objects within an image generated by a second network. In at least one embodiment, the second network is a controllable network. 公开了用生成的图像训练目标检测系统，具体公开了使用自监督机器学习识别图像中对象的装置、系统和技术。在至少一个实施例中，机器学习系统被训练以通过训练第一网络识别由第二网络生成的图像内的对象来识别对象。在至少一个实施例中，第二网络是可控网络。