Synthesized Video Flicker Distortion Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 2D quality metrics, such as MSE and PSNR, fail to accurately reflect the perceptual quality of 3D and virtual reality videos due to their pixel-wise differences, which do not consider human perception and are inadequate for assessing flicker distortion in synthesized virtual view videos.
Innovation Solution
A computer-implemented method for determining the quality of synthesized video files by processing reference and synthesized video files to compare and quantify flicker distortion, using techniques like Depth Image-based Rendering (DIBR), sparse representation, and weighted pooling to assess flicker and spatial-temporal activity distortions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional 2D quality metrics (MSE, PSNR) are used to assess video quality, then the measurement process is simple and computationally efficient, but the metrics cannot properly reflect the real perceptual quality of visual signals and fail to capture flicker distortion in synthesized videos
Solution Approach 1:
The patent transitions from conventional 2D pixel-wise comparison to 3D spatio-temporal analysis by incorporating temporal dimension. It processes video sequences across multiple frames and depths, creating a three-dimensional representation that captures flicker distortion and temporal inconsistencies that 2D metrics cannot detect.
Solution Approach 2:
The patent introduces an intermediary processing stage that transforms raw video data into spatio-temporal gradient representations. This intermediary representation serves as a bridge between the input video and quality assessment, enabling detection of subtle temporal variations and flicker artifacts through gradient analysis.
2Reliability
If pixel-wise difference metrics are used for quality assessment, then the computation is straightforward, but the metrics do not consider human perception and cannot detect flicker distortion in synthesized virtual view videos
Solution Approach 1:
The patent applies local quality assessment by analyzing spatio-temporal gradients in specific regions of video frames. Instead of uniform pixel-wise comparison, it focuses on local variations in gradients across space and time, which correspond to human visual sensitivity to flicker and temporal artifacts in different video regions.
Solution Approach 2:
The patent changes the assessment parameters from static pixel intensity values to dynamic spatio-temporal gradients. By computing gradients across multiple frames and depths, it transforms the measurement parameters to capture temporal variations and flicker characteristics that align with human perception of video quality.
3Productivity
If conventional quality metrics are used for 3D and VR video assessment, then the evaluation process is simple, but the metrics are inadequate for optimizing compression techniques and improving quality of experience in 3D/VR systems
Solution Approach 1:
The patent performs preliminary analysis of spatio-temporal gradients and flicker distortion characteristics before compression optimization. By pre-characterizing the temporal and spatial variations in the video content, it enables targeted compression strategies that preserve perceptually important information while removing redundant data.
Solution Approach 2:
The patent establishes a feedback mechanism where quality assessment results based on spatio-temporal analysis inform compression parameter optimization. The measured flicker distortion and temporal activity metrics provide feedback for adjusting compression settings to maintain perceptual quality while improving compression efficiency.
Data Source
AI summary
A computer-implemented method and related system for determining a quality of a synthesized video file. The method includes processing a reference video file and a synthesized video file associated with the reference video file to compare the original video file and the synthesized video file. The method also includes determining an extent of flicker distortion of the synthesized video file based on the processing.