公开(公告)号：US20220391704A1

公开(公告)日：2022-12-08

申请号：US17824571

申请日：2022-05-25

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Efthymia TSAMOURA ,  Davide BUFFELLI

IPC: G06N3/08 ,  G06N3/04 ,  G06V10/75 ,  G06V10/771 ,  G06V10/774 ,  G06V10/82 ,  G06V10/84

Abstract: Broadly speaking, the disclosure generally relates to relates to a computer-implemented methods and systems for scene graph generation, and in particular for training a machine learning, ML, model to generate a scene graph. The method includes inputting training a training image into a machine learning model, outputting a predicted label for at least two objects in the training image and a predicted label for a relationship between the at least two objects. The training method includes calculating a loss, which takes into account both a supervised loss calculated by comparing the predicted labels to the actual labels for the training image, and a logic-based loss calculated by comparing the predicted labels to stored integrity constraints comprising common-sense knowledge. Advantageously, this means that the performance of the model is improved without increasing processing at inference-time.