Video Frame Confidence Level Fusion for Consistent Image Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In video frame processing, non-uniform confidence levels for detected objects across frames can lead to inconsistencies and affect subsequent operations, such as image quality adjustments.
Innovation Solution
A method and apparatus that acquire a sequence of video frames, ascertain previous and subsequent frames, detect object regions, fuse confidence levels, and update confidence levels based on overlap and consistency, ensuring uniform image quality by interpolating between frames when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If confidence levels from different video frames are used directly without processing, then the detection results reflect frame-specific variations, but the image quality becomes non-uniform and inconsistent
Solution Approach 1:
The patent applies parameter changes by transforming the confidence level values through mathematical operations (multiplication by weight coefficients, addition of bias terms) to convert non-uniform confidence levels into uniform adjusted confidence levels, thereby maintaining detection consistency while achieving image quality uniformity
Solution Approach 2:
The patent implements feedback by using the adjusted confidence levels from subsequent frames to modify the confidence levels of previous frames, creating a feedback loop that propagates uniformity across all frames while preserving accurate object detection
2Ease of manufacture
If confidence levels are adjusted using only current frame information, then the processing is simple, but the uniformity across frames cannot be achieved
Solution Approach 1:
The patent applies preliminary action by pre-calculating weight coefficients and bias terms based on historical frame information before processing the current frame, allowing the system to achieve uniform confidence levels across all frames while maintaining relatively simple processing for each individual frame
Solution Approach 2:
The patent implements dynamics by making the adjustment parameters (weight coefficients and bias terms) adaptive and frame-dependent, allowing the system to dynamically adjust confidence levels based on the specific characteristics of each frame while maintaining overall uniformity
Data Source
AI summary
Embodiments of the present disclosure relate to a method and apparatus for processing a video frame. The method may include: acquiring a sequence of video frames of a video; ascertaining, in the sequence of the video frames, a previous frame, and ascertaining, in the sequence of the video frames, a subsequent frame corresponding to the ascertained previous frame based on acquired number of frames from the previous frame to the subsequent frame. An update step is performed as follows: acquiring object regions detected respectively in the ascertained previous frame and the ascertained subsequent frame, and confidence levels of the object regions; fusing a confidence level of a first object region and a confidence level of a second object region, and updating the confidence level of the second object region based on the fusion result; and updating the ascertained previous frame and the ascertained subsequent frame.