The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.

The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.

Date of Patent:
Jun. 09, 2026

Filed:

Jun. 28, 2022
Applicant:

Amazon Technologies, Inc., Seattle, WA (US);

Inventors:

Mohamed Kamal Omar, Seattle, WA (US);

Ashutosh Sanan, Seattle, WA (US);

Xiaohang Sun, Bellevue, WA (US);

Han-Kai Hsu, Seattle, WA (US);

Wentao Zhu, Redmond, WA (US);

Xiang Hao, Redmond, WA (US);

Ahmed Aly Saad Ahmed, Bothell, WA (US);

Assignee:

Amazon Technologies, Inc., Seattle, WA (US);

Attorneys:
Primary Examiner:
Int. Cl.
CPC ...
G06V 20/40 (2022.01); G06F 16/738 (2019.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 10/82 (2022.01);
U.S. Cl.
CPC ...
G06V 20/41 (2022.01); G06F 16/738 (2019.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 10/82 (2022.01);
Abstract

Systems and techniques for training a machine learning model to identify content labels within a video catalog using multimodal inputs are described. The techniques include receiving a multimodal input including the content. The techniques include determining a first selection of video data in data clusters based on the input and determining metadata indicating a correlation between the first selection and the attribute. Subsequently a second selection is selected based on the metadata and the first selection that may be used as a training dataset. A machine learning model is trained using the training data to determine instances of the attribute and build a content repository that summarizes the video data using the attribute labels.


Find Patent Forward Citations

Loading…