公开(公告)号：US20230267764A1

公开(公告)日：2023-08-24

申请号：US17652026

申请日：2022-02-22

Applicant: ADOBE INC.

Inventor： Md Mehrab Tanjim ,  Ritwik Sinha ,  Moumita Sinha ,  David Thomas Arbour ,  Sridhar Mahadevan

IPC: G06V40/16

CPC classification number: G06V40/172

Abstract: Systems and methods for diversity auditing are described. The systems and methods include identifying a plurality of images; detecting a face in each of the plurality of images using a face detection network; classifying the face in each of the plurality of images based on a sensitive attribute using an image classification network; generating a distribution of the sensitive attribute in the plurality of images based on the classification; and computing a diversity score for the plurality of images based on the distribution.

公开(公告)号：US12001520B2

公开(公告)日：2024-06-04

申请号：US17485780

申请日：2021-09-27

Applicant: Adobe Inc.

Inventor： Ritwik Sinha ,  Sridhar Mahadevan ,  Moumita Sinha ,  Md Mehrab Tanjim ,  Krishna Kumar Singh ,  David Arbour

IPC: G06K9/00 ,  G06F18/214 ,  G06F18/28 ,  G06N3/045

CPC classification number: G06F18/28 ,  G06F18/2148 ,  G06N3/045

Abstract: Methods and systems disclosed herein relate generally to systems and methods for generating simulated images for enhancing socio-demographic diversity. An image-generating application receives a request that includes a set of target socio-demographic attributes. The set of target socio-demographic attributes can define a gender, age, and/or race of a subject that are non-stereotypical for a particular occupation. The image-generating application applies the a machine-learning model to the set of target socio-demographic attributes. The machine-learning model generates a simulated image depicts a subject having visual characteristics that are defined by the set of target socio-demographic attributes.

公开(公告)号：US20230094954A1

公开(公告)日：2023-03-30

申请号：US17485780

申请日：2021-09-27

Applicant: Adobe Inc.

Inventor： Ritwik Sinha ,  Sridhar Mahadevan ,  Moumita Sinha ,  Md Mehrab Tanjim ,  Krishna Kumar Singh ,  David Arbour

IPC: G06K9/62 ,  G06N3/04

Abstract: Methods and systems disclosed herein relate generally to systems and methods for generating simulated images for enhancing socio-demographic diversity. An image-generating application receives a request that includes a set of target socio-demographic attributes. The set of target socio-demographic attributes can define a gender, age, and/or race of a subject that are non-stereotypical for a particular occupation. The image-generating application applies the a machine-learning model to the set of target socio-demographic attributes. The machine-learning model generates a simulated image depicts a subject having visual characteristics that are defined by the set of target socio-demographic attributes.

公开(公告)号：US12159482B2

公开(公告)日：2024-12-03

申请号：US17652026

申请日：2022-02-22

Applicant: ADOBE INC.

Inventor： Md Mehrab Tanjim ,  Ritwik Sinha ,  Moumita Sinha ,  David Thomas Arbour ,  Sridhar Mahadevan

IPC: G06V40/16

Abstract: Systems and methods for diversity auditing are described. The systems and methods include identifying a plurality of images; detecting a face in each of the plurality of images using a face detection network; classifying the face in each of the plurality of images based on a sensitive attribute using an image classification network; generating a distribution of the sensitive attribute in the plurality of images based on the classification; and computing a diversity score for the plurality of images based on the distribution.

公开(公告)号：US20240046412A1

公开(公告)日：2024-02-08

申请号：US17880120

申请日：2022-08-03

Applicant: ADOBE INC.

Inventor： Md Mehrab Tanjim ,  Krishna Kumar Singh ,  Kushal Kafle ,  Ritwik Sinha

IPC: G06T3/40

CPC classification number: G06T3/4046 ,  G06T3/4053

Abstract: A system debiases image translation models to produce generated images that contain minority attributes. A balanced batch for a minority attribute is created by over-sampling images having the minority attribute from an image dataset. An image translation model is trained using images from the balanced batch by applying supervised contrastive loss to output of an encoder of the image translation model and an auxiliary classifier loss based on predicted attributes in images generated by a decoder of the image translation model. Once trained, the image translation model is used to generate images with the minority image when given an input image having the minority attribute.