Electronic Apparatus Object Recognition Reducing Server Usage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for content recognition in TV systems require additional operations and apparatus for collecting and processing program identification information, leading to significant costs and increased server usage.
Innovation Solution
An electronic apparatus equipped with a processor and communicator that identifies objects in multimedia data frames and uses content guide information from a server to recognize content, minimizing server usage by optionally acquiring digital video fingerprints for content identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional content recognition methods are used, then content identification can be achieved, but server usage and costs increase significantly
Solution Approach 1:
The patent extracts the content recognition function from the server and relocates it to the electronic apparatus. The processor in the electronic apparatus directly identifies objects in video frames and matches them with content guide information locally, eliminating the need to transmit video data or fingerprints to the server for recognition. This extraction principle reduces server usage while maintaining recognition accuracy.
Solution Approach 2:
The electronic apparatus performs self-service content recognition by using its own processor to identify objects in video frames and match them with locally stored content guide information. The system serves itself by autonomously completing the recognition process without requiring external server assistance, thereby reducing server dependency and associated costs.
2Reliability
If additional apparatus for collecting and processing program identification information is added, then content recognition capability is improved, but device complexity increases
Solution Approach 1:
The patent makes the processor multi-functional by enabling it to perform both general processing tasks and specific content recognition functions. The same processor that handles video playback and control also identifies objects in frames and matches them with content guide information. This universality eliminates the need for separate dedicated recognition apparatus, reducing device complexity while maintaining recognition capability.
Solution Approach 2:
The patent merges the content recognition function with the existing electronic apparatus architecture. The processor, communicator, and storage components that already exist in the apparatus are combined to perform content recognition by matching identified objects with content guide information stored in the apparatus. This merging approach avoids adding separate dedicated recognition hardware, thereby reducing overall device complexity.
3Productivity
If object recognition AI algorithms are implemented, then content identification efficiency is improved, but processing requirements increase
Solution Approach 1:
The patent applies partial action by using AI algorithms selectively only for identifying key objects in video frames that are relevant to content identification, rather than processing entire video sequences or all visual elements. The processor focuses on detecting and matching specific objects with content guide information, reducing overall processing requirements while maintaining identification efficiency.
Solution Approach 2:
The patent performs preliminary action by pre-storing content guide information including object descriptions and content metadata in the apparatus before recognition is needed. When video content is being played, the processor can quickly compare identified objects against this pre-prepared guide information, reducing real-time processing requirements while maintaining efficient content identification.
Data Source
AI summary
An electronic apparatus includes a communicator circuitry, and a processor for obtaining multimedia data from an external apparatus via the communicator, identifying an object in at least one frame from among a plurality of frames included in the multimedia data, and identify a content corresponding to the identified object based on content guide information provided from a first server.