Combined Video Stream Generation Without Transcoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating a combined video stream from multiple images require transcoding, which leads to decreased image quality and delayed display time due to the fixed location of the region of interest (ROI) and the need to decode and re-encode images, especially in video conference and multi-channel broadcasting scenarios.
Innovation Solution
An apparatus and method that acquire and combine individual video streams from multiple channels by extracting image codes and header information, inserting them into a predetermined layout, and generating a combined video stream without transcoding, thereby synchronizing and displaying multiple images on a single screen.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If transcoding is performed to generate combined video stream from multiple images, then multiple images can be combined and transmitted over single channel, but image quality decreases and display time is delayed
Solution Approach 1:
The patent extracts only the necessary components (image codes and header information) from each individual video stream without performing full decoding. This allows the essential data to be retrieved and recombined while avoiding the quality degradation associated with complete transcoding operations.
Solution Approach 2:
The patent segments the video stream processing into distinct components: extracting image codes from each channel, extracting header information, and recombining them in a predetermined layout. This segmentation enables efficient combination without requiring full transcoding of each image.
2Adaptability or versatility
If transcoding is performed to generate combined video stream from multiple images, then multiple images can be combined and transmitted over single channel, but display time is delayed
Solution Approach 1:
The patent performs preliminary extraction of image codes and header information from each video stream before combination. By preparing these components in advance and organizing them according to a predetermined layout, the system eliminates time-consuming transcoding operations during the combination phase, thereby reducing display time delays.
Solution Approach 2:
Instead of decoding and re-encoding (transcoding) the images, the patent copies the existing image codes from the individual video streams and reassembles them with modified header information. This copying approach preserves the original encoded data and avoids the time penalty of full transcoding.
3Ease of manufacture
If fixed ROI location is used for transcoding, then encoding can be performed, but entire image must be transcoded even when ROI is relocated
Solution Approach 1:
The patent makes the ROI location dynamic by extracting and processing header information that contains ROI position data for each image. This allows the system to adapt to relocated ROIs without requiring complete re-transcoding, as the dynamic header information guides the combination process to focus only on the relevant regions of interest.
Data Source
AI summary
An apparatus of generating a combined video stream of multiple images, includes: an image acquiring unit to acquire a plurality of individual video streams transmitted over multiple channels; an image information extracting unit to extract individual image codes and individual header information from each of the plurality of the individual video streams; and a bit stream generating unit to generate combined image codes by inserting individual image codes into corresponding respective positions corresponding to each of the plurality of individual video streams in a predetermined layout of a display picture, generate combined header information corresponding to the combined image codes based on respective individual header information, and generate a combined video stream including the combined image codes and the combined header information.