公开(公告)号：US20250053865A1

公开(公告)日：2025-02-13

申请号：US18561104

申请日：2022-12-14

Applicant: Google LLC

Inventor： Wei Qiao ,  Chun-Ta Lu ,  Yinatao Liu ,  Ariel Fuxman ,  Mehmet Nejat Tek ,  Dongjin Kwon ,  Florian Nils Stimberg

IPC: G06N20/00 ,  G06V20/70

Abstract: The technology is generally directed to the training and execution of a model to identify policy violating content that has been obfuscated. The model may be trained using obfuscated training images. The obfuscated training images may be associated with one or more labels, such as a policy, obfuscation label, etc. The obfuscated training images and associated labels may be input into the model. During training, the output of the model may be a policy prediction as to whether the obfuscated input images violate the content policy of a host or are approved content for publishing. During implementation, the model may receive content as input and provide as output a policy prediction for the content. The host may use the policy prediction provided by the model to determine whether or not to publish the content.

公开(公告)号：US20240428573A1

公开(公告)日：2024-12-26

申请号：US18341218

申请日：2023-06-26

Applicant: Google LLC

Inventor： Ariel Fuxman ,  Alexander Kenji Hata ,  Edward Benjamin Vendrow ,  Otilia Stretcu ,  Wenlei Zhou ,  Krishnamurthy Viswanathan ,  Aditya Avinash ,  Gabriel Berger ,  Andrew Ames Bunner ,  Javier Alejandro Rey ,  Wei Qiao ,  Yintao Liu ,  Guanzhong Wang ,  Thomas Nathan Denby ,  Mehmet Nejat Tek ,  Neil Gordon Alldrin ,  Enming Luo ,  Chun-Ta Lu

IPC: G06V10/778 ,  G06V10/764 ,  G06V10/774 ,  G06V10/82 ,  G06V10/94

Abstract: A computer-implemented method includes receiving an input from a user relating to a concept, automatically obtaining a first set of images from an unlabeled dataset of images based on the input, and obtaining a first rating via the user for each image from the first set of images. The method further includes training a classifier model relating to the concept based on the first set of images rated by the user, automatically obtaining a second set of images from the unlabeled dataset of images based on the classifier model trained based on the first set of images, and obtaining a second rating via the user for each image from the second set of images. The classifier model relating to the concept is retrained based on the first set of images rated by the user and the second set of images rated by the user to obtain an updated classifier model.