公开(公告)号：US12183056B2

公开(公告)日：2024-12-31

申请号：US17573041

申请日：2022-01-11

Applicant: Adobe Inc.

Inventor： Maksym Andriushchenko ,  John Collomosse ,  Xiaoyang Li ,  Geoffrey Oxholm

IPC: G06V10/75 ,  G06F16/58 ,  G06F16/583 ,  G06N3/084 ,  G06V10/72 ,  G06V10/771

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize a deep visual fingerprinting model with parameters learned from robust contrastive learning to identify matching digital images and image provenance information. For example, the disclosed systems utilize an efficient learning procedure that leverages training on bounded adversarial examples to more accurately identify digital images (including adversarial images) with a small computational overhead. To illustrate, the disclosed systems utilize a first objective function that iteratively identifies augmentations to increase contrastive loss. Moreover, the disclosed systems utilize a second objective function that iteratively learns parameters of a deep visual fingerprinting model to reduce the contrastive loss. With these learned parameters, the disclosed systems utilize the deep visual fingerprinting model to generate visual fingerprints for digital images, retrieve and match digital images, and provide digital image provenance information.

公开(公告)号：US20230222762A1

公开(公告)日：2023-07-13

申请号：US17573041

申请日：2022-01-11

Applicant: Adobe Inc.

Inventor： Maksym Andriushchenko ,  John Collomosse ,  Xiaoyang Li ,  Geoffrey Oxholm

IPC: G06V10/75 ,  G06F16/583 ,  G06F16/58 ,  G06V10/72 ,  G06V10/771 ,  G06N3/08

CPC classification number: G06V10/751 ,  G06F16/583 ,  G06F16/5866 ,  G06V10/72 ,  G06V10/771 ,  G06N3/084

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize a deep visual fingerprinting model with parameters learned from robust contrastive learning to identify matching digital images and image provenance information. For example, the disclosed systems utilize an efficient learning procedure that leverages training on bounded adversarial examples to more accurately identify digital images (including adversarial images) with a small computational overhead. To illustrate, the disclosed systems utilize a first objective function that iteratively identifies augmentations to increase contrastive loss. Moreover, the disclosed systems utilize a second objective function that iteratively learns parameters of a deep visual fingerprinting model to reduce the contrastive loss. With these learned parameters, the disclosed systems utilize the deep visual fingerprinting model to generate visual fingerprints for digital images, retrieve and match digital images, and provide digital image provenance information.