Visual Data Prioritization for Remote Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtual meetings and videoconferencing sessions, participants often struggle to identify and keep track of multiple participants, their roles, and changes in attendance, leading to difficulties in managing participant information effectively.
Innovation Solution
A system that processes video streams to identify and track object areas, such as facial features, generates visual data sets, and prioritizes them based on confidence values and network communication factors, selecting only necessary data for transmission to a remote system for image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all visual data from video streams is transmitted to the remote system for image processing, then the remote system can accurately identify and track participants, but the network bandwidth consumption and data transmission volume increase significantly
Solution Approach 1:
The patent extracts only the essential visual data (object areas representing participants) from the complete video stream for transmission to the remote system. By identifying and extracting specific regions of interest rather than transmitting the entire video feed, the system achieves accurate participant identification while significantly reducing the volume of data that needs to be transmitted over the network.
Solution Approach 2:
The patent applies different processing qualities to different parts of the video stream. Instead of uniformly processing or transmitting all visual data at the same quality level, the system identifies specific object areas (participants) and transmits only those regions with sufficient quality for identification purposes, while other areas are either not transmitted or transmitted at lower quality, thereby optimizing bandwidth usage.
2Speed
If visual data is transmitted frequently to ensure real-time participant tracking, then the responsiveness and accuracy of participant identification improve, but the network bandwidth consumption and data transmission frequency increase
Solution Approach 1:
The patent performs preliminary processing of the video stream locally to identify object areas (participants) before transmission. By pre-identifying which regions contain participants and preparing the relevant visual data in advance, the system can transmit only the necessary data at the required frequency, avoiding unnecessary transmissions and optimizing network bandwidth consumption while maintaining real-time tracking responsiveness.
3Loss of information
If complete video streams are transmitted for archiving and future reference, then all participant information is preserved, but the storage requirements and data processing load increase
Solution Approach 1:
The patent extracts and transmits only the object areas representing participants for archiving purposes, rather than storing complete video streams. This extraction approach preserves all necessary participant information for identification and tracking while significantly reducing the storage capacity required, as only the essential visual data is retained for future reference.
Data Source
AI summary
In a system including a processor and memory, the memory includes instructions that, when executed by the processor, cause the processor to control the system to perform receiving a video stream capturing objects; identifying, based on the received video stream, object areas corresponding to the objects, respectively; tracking the object areas in the received video stream; generating, based on the tracking of the object areas, visual data sets at a plurality of times, wherein each visual data set is generated at a different time and includes visual data representing each object area; determining a priority of each visual data in each visual data set; selecting, based on the determined priority of each visual data, a group of the visual data to be transmitted to a remote system; and transmitting, to the remote system, the selected group of the visual data.