User input-based interference removal in media items

• Main Authors: LIBA, OLE, JACOBS DAVID, SARMA NAVEEN, CAI LONGQI, LI SIYANG, FELDMAN BRYAN, KERNAN YAEL PRICE, CHEN HUIZHONG, SCHIFAUER, ALEXANDER
• Format: Patent
• Language: Chinese; English
• Published: 31.10.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

A media application receives a user input indicating that one or more objects are to be removed from a media item. The media application translates the user input into a bounding box. The media application provides tailoring of the media item to the segmented machine learning model based on the bounding box. The segmented machine learning model outputs a segment mask of one or more segment objects in a cut of the media item, and a corresponding segment score indicating a quality of the segment mask. 媒体应用接收指示要从媒体项中移除一个或多个对象的用户输入。媒体应用将用户输入转换为边界框。媒体应用基于边界框来向分段机器学习模型提供媒体项的裁剪。分段机器学习模型输出媒体项的裁剪中的一个或多个分段对象的分段蒙版，以及指示分段蒙版的质量的对应分段分数。