公开(公告)号：US20220375205A1

公开(公告)日：2022-11-24

申请号：US17664402

申请日：2022-05-20

Applicant: Google LLC

Inventor： Zizhao Zhang ,  Han Zhang ,  Long Zhao ,  Tomas Pfister

IPC: G06V10/77 ,  G06V10/764 ,  G06V10/22 ,  G06V10/44

Abstract: A method includes receiving image data including a series of image patches of an image. The method includes generating, using a first set of transformers of a vision transformer (V-T) model, a first set of higher order feature representations based on the series of image patches and aggregating the first set of higher order feature representations into a second set of higher order feature representations that is smaller than the first set. The method includes generating, using a second set of transformers of the V-T model, a third set of higher order feature representations based on the second set of higher order feature representations and aggregating the third set of higher order feature representations into a fourth set of higher order feature representations that is smaller than the third set. The method includes generating, using the V-T model, an image classification of the image based on the fourth set.

公开(公告)号：US20240265586A1

公开(公告)日：2024-08-08

申请号：US18564841

申请日：2022-05-27

Applicant: Google LLC

Inventor： Long Zhao ,  Han Zhang ,  Zizhao Zhang ,  Ting Chen

IPC: G06T11/00 ,  G06T3/4046

CPC classification number: G06T11/00 ,  G06T3/4046

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating high-resolution images using self-attention based neural networks. One of the systems includes a neural network configured to generate images, the neural network comprising a sequence of one or more first network blocks followed by a sequence of one or more second network blocks, wherein: each first network block is configured to perform operations comprising: applying a self-attention mechanism over at least a subset of first elements of a first block input to generate an updated first block input; and upsampling the updated first block input to generate a first block output; and each second network block is configured to perform operations comprising: processing a second block input using one or more neural network layers to generate an updated second block input; and upsampling the updated second block input to generate a second block output.

公开(公告)号：US20250111675A1

公开(公告)日：2025-04-03

申请号：US18900467

申请日：2024-09-27

Applicant: Google LLC

Inventor： Hui Miao ,  Chun-Te Chu ,  Mingyan Gao ,  Huanfen Yao ,  Ting Liu ,  Long Zhao ,  Liangzhe Yuan ,  Yukun Zhu ,  Vinay Kumar Bettadapura ,  Ye Jin

IPC: G06V20/40 ,  G06V10/74 ,  G06V10/75 ,  G06V10/762 ,  G06V10/80

Abstract: Methods and systems for media trend detection and maintenance are provided herein. A set of media items each having common media characteristics is identified. A set of pose values is determined for each respective media item of the set of media items. Each pose value is associated with a particular predefined pose for objects depicted by the set of media items. A set of distance scores is calculated. Each distance score represents a distance between the respective set of pose values determined for a media item and a respective set of pose values determined for an additional media item. A coherence score is determined for the set of media items based on the calculated set of distance scores. Responsive to a determination that the coherence score satisfies one or more coherence criteria, a determination is made that the set of media items corresponds to a media trend of a platform.