Low Frame Rate Video Streaming for Face-to-Face Conferencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video conferencing systems are inefficient at very low bit rates, as they degrade image quality due to bandwidth constraints, and struggle to maintain natural-looking results, especially in multi-party conferences where multiple video streams strain network bandwidth.
Innovation Solution
A low frame rate video transmission strategy that selectively transmits only important regions of carefully selected frames, using real-time face tracking and template matching to identify 'good' frames with open eyes, and employs image morphing to maintain visual quality, allowing for efficient compression and transmission at very low bit rates like 8 kb/s.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional compression techniques are used to transmit every video frame at very low bit rates, then bandwidth requirements are reduced, but image quality degrades significantly
Solution Approach 1:
The patent segments the video stream into individual frames and selectively transmits only important frames (e.g., when eyes are open) rather than every frame. This segmentation allows the system to reduce bandwidth by skipping unimportant frames while maintaining acceptable image quality through strategic transmission of critical frames.
Solution Approach 2:
The patent implements periodic action by transmitting video frames at reduced frame rates rather than continuously. The system periodically selects and transmits frames based on eye state detection, creating a pulsed transmission pattern that reduces bandwidth consumption while maintaining visual quality through selective updates.
2Loss of energy
If video frame rate is reduced to lower bandwidth requirements, then bandwidth consumption is reduced, but natural appearance of video deteriorates
Solution Approach 1:
The patent changes the parameter of frame rate dynamically based on eye state detection. When eyes are open, frames are transmitted at normal frame rate to maintain natural appearance. When eyes are closed, frame rate is reduced to lower bandwidth consumption. This parameter adaptation allows the system to maintain natural appearance during critical moments while reducing bandwidth during less critical periods.
Solution Approach 2:
The patent employs feedback through eye state detection that continuously monitors the video stream and adjusts transmission parameters accordingly. The eye detection system provides feedback about the current frame's importance, allowing the compression system to adaptively adjust frame rate and transmission quality in real-time, maintaining natural appearance when needed while reducing bandwidth when appropriate.
3Adaptability or versatility
If multiple video streams are transmitted simultaneously for multi-party conferences, then all participants can communicate, but network bandwidth requirements increase significantly
Solution Approach 1:
The patent applies segmentation to multi-party conferences by processing each participant's video stream independently and selectively transmitting only important frames for each participant. This allows the system to maintain multi-party communication capability while significantly reducing total bandwidth consumption by not transmitting every frame for every participant.
Solution Approach 2:
The patent implements partial action by transmitting only the necessary portions of video data (important frames with open eyes) rather than complete continuous streams for all participants. This partial transmission approach enables multi-party conference functionality while reducing overall bandwidth requirements compared to transmitting full video streams for all participants simultaneously.
Data Source
AI summary
A real-time low frame-rate video compression system and method that allows the user to perform face-to-face communication through an extremely low bandwidth network. At the encoder side, the system is able to automatically select only a few good faces from the original sequence with high visual quality and compress and transmit them. At the decoder side, the system use image-morphing based rendering method to generate a normal frame-rate video. Experimental results show that the system is superior to more traditional video codecs for low bit-rate face-to-face communication.