Video Conferencing Facial Correction for Handheld Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video conferencing technologies using handheld devices suffer from poor video quality due to head motion, gaze misalignment, and distorted facial views, leading to a suboptimal user experience with lack of eye contact.
Innovation Solution
A video conferencing device equipped with a data collection module to gather data on face location and orientation, and a facial correction module that modifies facial appearance for transmission, using an accelerometer to measure camera orientation and movement, and a range finder to determine face-camera distance, in conjunction with a database of facial characteristics for enhancing image quality and eye contact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If handheld devices are used for video conferencing, then portability and accessibility are improved, but video quality deteriorates due to head motion and distorted facial views
Solution Approach 1:
The patent replaces mechanical stabilization methods with sensor-based detection (accelerometers, gyroscopes) and software-based correction. The system detects head motion through sensors and uses image processing algorithms to compensate for distortion, substituting physical stabilization mechanisms with electronic and computational approaches that maintain video quality while preserving handheld portability.
Solution Approach 2:
The system dynamically changes image parameters (orientation, scale, position) based on detected head motion and facial characteristics. By adjusting these parameters in real-time according to sensor data and facial recognition results, the system compensates for motion-induced distortion and maintains optimal video quality despite handheld usage conditions.
2Adaptability or versatility
If handheld devices are used for video conferencing, then accessibility is improved, but gaze alignment deteriorates leading to misaligned facial views
Solution Approach 1:
The system implements a feedback loop where facial characteristics are continuously detected and analyzed, and image parameters are adjusted based on this feedback. The facial recognition system provides real-time information about gaze direction and facial orientation, which feeds back to the image processing module to correct alignment, ensuring accurate gaze representation despite handheld device movement.
3Manufacturing precision
If facial correction processing is applied, then video quality is improved, but processing time and computational load increase
Solution Approach 1:
The system applies selective correction based on detected facial characteristics and motion levels. Rather than applying full correction processing to all frames, the system adjusts image parameters proportionally to the detected distortion level, applying only the necessary correction amount. This reduces unnecessary processing while maintaining video quality where needed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves video quality by ensuring a clear, aligned, and engaging facial view for participants, maintaining eye contact and reducing distortion, thereby enhancing the overall video conferencing experience.
Implementation Method 1
using an accelerometer to measure camera orientation and movement
Implementation Method 2
a range finder to determine face-camera distance
Data Source
AI summary
A video conferencing device for improving video quality when a communications device is used for a video conference. In one embodiment the device includes (1) a data collection module for collecting data regarding the location and orientation of a user's face relative to a camera on a communications device; and (2) a facial correction module associated with the data collection module, the facial correction module modifying a user's facial orientation and characteristics for transmission during a video conference based on data from the data collection module and from an associated database of relevant facial characteristics.