Broadcast Object Detection and Keyword Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for searching information during digital TV broadcasts are inconvenient, requiring direct user manipulation and lacking in efficiency.
Innovation Solution
An apparatus and method that detects objects in a broadcast scene, generates keywords with category information to eliminate ambiguity, sets scene sections based on common keywords, and synchronizes related information with the scene for seamless display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional search methods are used in digital TV, then users can search for information, but the operation becomes inconvenient and complex
Solution Approach 1:
The system automatically detects objects in broadcast scenes, generates keywords, and searches for related information without requiring user manipulation. The terminal device performs self-service by autonomously capturing scene information, processing it through object detection and keyword generation, and retrieving relevant data from databases or networks.
Solution Approach 2:
The system performs preliminary actions by pre-processing broadcast content into structured data formats, pre-generating keywords from detected objects, and pre-organizing related information in databases before user requests are made. This preliminary preparation enables rapid information retrieval when needed.
2Reliability
If object names are used without category information, then searching is simpler, but ambiguity increases
Solution Approach 1:
The keyword generation process segments object identification into distinct components: object name detection, category classification, and keyword assembly. Each component handles a specific aspect of identification, with object detectors identifying what is present, category classifiers determining the type, and keyword generators combining them into structured search terms.
Solution Approach 2:
Category information acts as an intermediary between raw object detection and final information retrieval. The category classification component serves as a mediator that translates detected objects into categorized keywords, resolving ambiguity by adding contextual classification layers without requiring direct user intervention.
3Loss of information
If manual information searching is required, then information can be found, but user time and effort increase
Solution Approach 1:
The system maintains continuous operation by automatically and continuously detecting objects in broadcast scenes, generating keywords in real-time, and retrieving related information without interruption. This continuous automated process eliminates gaps between user actions and ensures information is always being prepared for retrieval.
Solution Approach 2:
The patent replaces manual user operations with automated electronic processing systems. Object detection algorithms, keyword generation software, and automated database search mechanisms substitute for manual information seeking, transforming a mechanically interactive process into an automated electronic information retrieval system.
Data Source
AI summary
A method and apparatus for providing information related to broadcast programs are provided. According to an example, objects are detected from broadcast scenes and keywords. Each of the objects include an object name and information regarding the meaning of the object, are generated. Then, scene sections are set based on the keywords and related information is searched for each scene section. The related information identified by the search is synchronized with the corresponding scene section and then provided to a user.