EP-4742686-A1 - DATA PROCESSING METHOD AND RELATED DEVICE

EP4742686A1EP 4742686 A1EP4742686 A1EP 4742686A1EP-4742686-A1

Abstract

This application relates to the field of media applications and provides a data processing method. The method includes: determining whether one or more trust indicators of a media asset correspond to a metric requirement, to obtain trustworthiness information of the media asset. The media asset may include media content. The media asset further includes at least one of metadata of the media content and a trust record of the media content. The trust indicator is an indicator related to a trustworthiness level of the media asset. A user may determine a trustworthiness level of the media asset based on the trustworthiness information.

Inventors

KANG, Xin
HU, Ziyuan
ZHOU, HAIBO
LI, TIEYAN

Assignees

Huawei Technologies Co., Ltd.

Dates

Publication Date: 20260513
Application Date: 20240819

Claims (20)

A data processing method, wherein the method comprises: obtaining one or more trust indicators of a media asset, wherein the media asset comprises media content, and the media asset further comprises at least one of metadata of the media content and a trust record of the media content; and determining whether the one or more trust indicators satisfy a metric.
The method according to claim 1, wherein the method further comprises: obtaining first trust configuration information, wherein the first trust configuration information comprises a metric that the media asset needs to satisfy.
The method according to claim 1 or 2, wherein the one or more trust indicators are determined based on at least one of the media content, the metadata corresponding to the media content, and the trust record corresponding to the media content.
The method according to any one of claims 1 to 3, wherein the one or more trust indicators are encapsulated in a trust credential.
The method according to any one of claims 1 to 4, further comprising: generating trustworthiness information of the media asset, wherein the trustworthiness information indicates whether the one or more trust indicators satisfy the metric, and the trust indicator comprises a parameter indicating a trustworthiness level of the media asset.
The method according to any one of claims 1 to 4, wherein the metric comprises data indicating a trust indicator that the media asset needs to satisfy.
The method according to any one of claims 1 to 6, wherein determining whether the one or more trust indicators satisfy the metric comprises: determining whether a corresponding trust indicator in the metric is comprised in the one or more trust indicators, or determining whether the one or more trust indicators comprise a corresponding trust indicator in the metric.
The method according to any one of claims 1 to 7, wherein the method further comprises: adding the trustworthiness information to the trust record.
The method according to claim 8, wherein the trust record comprises a trust manifest; and adding the trustworthiness information to the trust record comprises: adding the trustworthiness information to the trust manifest.
The method according to claim 8 or 9, wherein the method further comprises: obtaining indication information of the media asset; and adding to the trust record comprises: adding, based on the indication information, the trustworthiness information to the trust record corresponding to the media asset indicated by the indication information.
The method according to any one of claims 1 to 10, wherein the trustworthiness information is encapsulated in a trust report.
The method according to any one of claims 1 to 10, wherein the method further comprises: generating, based on the trustworthiness information, a file associated with the media asset.
The method according to any one of claims 1 to 12, wherein the first trust configuration information comprises one of a plurality of pieces of trust configuration information, and different pieces of trust configuration information indicate metric requirements, of different regions or users, that the media asset needs to satisfy.
The method according to any one of claims 1 to 13, wherein the action of obtaining the one or more trust indicators of the media asset is triggered by capturing the media content through a hardware sensor or generating the media content through generation software.
The method according to any one of claims 1 to 13, wherein before obtaining the one or more trust indicators of the media asset, the method further comprises: receiving a trustworthiness evaluation request for the media content.
The method according to any one of claims 1 to 15, wherein the media content is at least one of an image, a video, or an audio.
A data processing apparatus, wherein the apparatus comprises: an obtaining module, configured to obtain one or more trust indicators of a media asset, wherein the media asset comprises media content, and the media asset further comprises at least one of metadata of the media content and a trust record of the media content; and a processing module, configured to determine whether the one or more trust indicators satisfy a metric.
A data processing apparatus, wherein the apparatus comprises a memory and a processor, the memory stores code, and the processor is configured to obtain the code and perform the method according to any one of claims 1 to 16.
A computer-readable storage medium, comprising computer-readable instructions, wherein when the computer-readable instructions are run on a computer device, the computer device is enabled to perform the method according to any one of claims 1 to 16.
A computer program product, comprising computer-readable instructions, wherein when the computer-readable instructions are run on a computer device, the computer device is enabled to perform the method according to any one of claims 1 to 16.

Description

This application claims priority to Chinese Patent Application No. 202311063709.1, filed with the China National Intellectual Property Administration on August 22, 2023 and entitled "DATA PROCESSING METHOD AND RELATED DEVICE", which is incorporated herein by reference in its entirety. TECHNICAL FIELD This application relates to the field of terminals and media applications, and in particular, to a data processing method and a related device. BACKGROUND The popularization of more convenient image shooting devices such as mobile phones, coupled with the development of various types of simple, efficient, and powerful content editing software (for example, AIGC software), has lowered barriers to media creation and modification for people, enabling publication across various social media platforms. However, modification of media content exacerbates the spread of some misinformation and disinformation, undermining the trustworthiness of media content as a carrier of information. Particularly in light of AI advancements, distinguishing between fabricated fake content and authentic content has become increasingly challenging to naked eyes, making "seeing" no longer necessarily "believing". Therefore, technologies that can enhance our ability to assess a trustworthiness level of content are crucial. For example, media content is an image. In existing implementations, to detect whether the media content includes fake content, an AI detection model may be trained on a large quantity of real image sets and fake image sets. The detection model detects whether an input image has a feature similar to that of a fake image, to determine whether the image is trustworthy. However, this type of AI detection model can verify trustworthiness levels of only some specific fake content. Therefore, there is an urgent need for a method that can verify a trustworthiness level of media content. SUMMARY According to a first aspect, this application provides a data processing method, where the method includes: determining whether one or more trust indicators of a media asset correspond to a metric requirement, to obtain trustworthiness information of the media asset. A user may determine a trustworthiness level of the media asset based on the trustworthiness information. The media asset includes media content. The media asset further includes at least one of metadata of the media content and a trust record of the media content. The trust indicator is a parameter indicating the trustworthiness level of the media asset. In a possible implementation, the method further includes: generating the trustworthiness information of the media asset. In a possible implementation, the trustworthiness information may be encapsulated into a trust report. In a possible implementation, the one or more trust indicators are determined based on at least one of the media content, the metadata corresponding to the media content, and the trust record corresponding to the media content. For example, the trust indicator may be a feature extracted from the media content, the metadata corresponding to the media content, and the trust record corresponding to the media content, or may be data obtained by performing specific processing on the media content, the metadata corresponding to the media content, and the trust record corresponding to the media content. In a possible implementation, the trustworthiness information indicates whether the one or more trust indicators satisfy a metric, and the trust indicator includes a parameter indicating the trustworthiness level of the media asset. For example, the media content is an image, and features such as shooting time, a shooting location, a photographer, a camera lens parameter, and a shooting parameter in the metadata may be selected as trust indicators of the metadata. For example, the media content is an image, and an image generation manner, an image editing manner, and the like in the trust record may be selected as trust indicators of the trust record. The image generation manner may include but is not limited to whether the media content is generated through AIGC, whether the media content is generated through a camera, whether the media content is generated with software assistance, whether the media content is generated through synthetic media, and whether the media content can be used to train a model. The image editing manner may include but is not limited to rotation, resizing, cropping, content editing, and the like. For example, the media content is an image, and features such as a person, a scene, an object type, a position relationship between objects, or an image style (for example, color, lighting, or brightness) in the media content may be selected as trust indicators of the media content. In a possible implementation, a metric corresponding to the trust indicator is specified in a profile, and first trust configuration information may be obtained, where the first trust configuration information inc