Real-Time Image Synthesis with Object Information Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audiovisual communication services in mobile terminals transmit real-time images, making it difficult to reflect changes in edited or synthesized images, as they do not account for real-time object changes, leading to potential exposure of unwanted portions of a user's image.
Innovation Solution
A method and device for synthesizing images in mobile terminals that generate a synthesis image by combining sensor input and user-designated images, storing object information, and updating it in real-time to reflect changes in the sensor image, ensuring that the synthesized image accurately represents the user's intended image during audiovisual communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If real-time image transmission is used in audiovisual communication, then communication speed and responsiveness are improved, but user privacy control deteriorates as unwanted portions of the user's image may be exposed
Solution Approach 1:
The image is segmented into multiple regions of interest (ROIs) based on object detection. Only the detected objects within these segmented regions are transmitted, rather than the entire image. This segmentation allows real-time transmission while protecting privacy by excluding unwanted portions from the transmitted data stream.
Solution Approach 2:
The patent extracts only the essential object information from the full image. By identifying and extracting specific objects of interest through image processing, the system transmits minimal necessary data (object boundaries, features) rather than complete image data, thereby maintaining communication speed while reducing privacy risks.
2Object-affected harmful factors
If image editing or synthesis is performed to protect privacy, then user privacy control is improved, but real-time responsiveness deteriorates as changes are not reflected in transmitted images
Solution Approach 1:
The system performs preliminary image processing by detecting objects and defining regions of interest before transmission. This preliminary action establishes the framework for privacy protection in advance, allowing the system to maintain real-time responsiveness by only processing and transmitting essential object information rather than performing extensive editing during communication.
Solution Approach 2:
The patent implements a feedback mechanism where object detection results from transmitted images are continuously monitored. This feedback loop allows the system to adaptively adjust the regions of interest and object selection in real-time, ensuring that privacy protection measures remain effective while maintaining responsive communication without requiring extensive re-editing.
3Manufacturing precision
If complete image data is transmitted for accurate representation, then image quality is improved, but data transmission volume increases leading to higher bandwidth consumption
Solution Approach 1:
The system extracts only the essential visual information related to detected objects rather than transmitting complete image data. By focusing on object boundaries, key features, and relevant regions, the patent maintains sufficient image quality for communication purposes while dramatically reducing the volume of transmitted data.
Solution Approach 2:
The patent applies different quality levels to different regions of the image based on their importance. High-quality detailed information is transmitted only for detected objects and their immediate surroundings, while peripheral or less important regions are transmitted at lower resolution or omitted entirely, optimizing the balance between image quality and data volume.
Data Source
AI summary
A method and device for synthesizing an image for audiovisual communication includes: generating, if a request for synthesizing a sensor image input through a camera sensor and an image designated by a user is input, a synthesis image of the sensor image and the designated image. The method also includes storing, when the synthesis image is generated, object information of an object of the sensor image and object information of an object of the designated image; and changing, if a sensor image is input in which object information of the sensor image object has changed, object information of the designated image object according to the changed object information of the sensor image object and synthesizing the sensor image and the designated image in which object information has changed.