Systems and Methods for Customizing Text in Media Content
AssigneeCyberlink Corp.
InventorHsieh-Wei Chen
Life of this patent
Estimating the term…
Abstract
Various embodiments are disclosed for facilitating automatic media editing. Media content is obtained and semantic analysis is performed on text in at least a portion of the media content to obtain at least one semantic textual segment each corresponding to a text section of the media content, wherein the text section comprises at least one word in the text in the at least a portion of the media content. At least one context token corresponding to the at least one semantic textual segment is generated. The text section is visually accentuated according to the context token.
Claims32
A method implemented in a media processing device, comprising:
- obtaining, by the media processing device, media content;
- performing, by the media processing device, semantic analysis on text in at least a portion of the media content to obtain at least one semantic textual segment each corresponding to a text section of the media content, wherein the text section comprises at least one word in the text in the at least a portion of the media content;
- generating, by the media processing device, at least one context token corresponding to the at least one semantic textual segment;
- and
- visually accentuating, by the media processing device, the text section according to the at least one context token.
A system for editing media content, comprising:
- a processor;
- and
- at least one application executable in the processor, the at least one application comprising:
- a media interface for obtaining media content
- a content analyzer for performing semantic analysis on text in at least a portion of the media content to obtain at least one semantic textual segment each corresponding to a text section of the media content, wherein the text section comprises at least one word in the text in the at least a portion of the media content;
- a tokenizer for generating at least one context token corresponding to the at least one semantic textual segment;
- and
- a visualizer for visually accentuating the text section according to the at least one context token.
A non-transitory computer-readable medium embodying a program executable in a computing device, comprising:
- code that obtains media content;
- code that performs semantic analysis on text in at least a portion of the media content to obtain at least one semantic textual segment each corresponding to a text section of the media content, wherein the text section comprises at least one word in the text in the at least a portion of the media content;
- code that generates at least one context token corresponding to the at least one semantic textual segment;
- and
- code that visually accentuates the text section according to the at least one context token.
A method implemented in a media processing, comprising:
- obtaining, by the media processing device, media content and performing semantic analysis on a textual portion of the media content;
- generating, by the media processing device, textual context tokens based on the semantic analysis;
- performing semantic analysis on an audio portion and on a visual portion of the media content corresponding to the textual portion;
- generating context tokens relating to the audio and visual portions;
- combining, by the media processing device, the textual context tokens and the context tokens relating to the audio and visual portions;
- and
- visually accentuating, by the media processing device, at least one context portrayed in at least a portion of media content according to the combined context tokens.
A method implemented in a media processing device, comprising:
- obtaining, by the media processing device, a photo collection comprising digital images and textual content;
- performing, by the media processing device, semantic analysis on the textual content to obtain at least one semantic textual segment each corresponding to a text section of the photo collection, wherein the text section comprises at least one word in the textual content in the at least a portion of the photo collection;
- generating, by the media processing device, at least one context token corresponding to the at least one semantic textual segment;
- and
- visually accentuating, by the media processing device, the text section according to the at least one context token.
Dependencies are inferred from the claim text.