Adaptive Frame Interpolation for Video Jitter and Bad Edit Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional frame interpolation methods suffer from image jitter and loss of image information due to unstable input frames, and they either discard necessary frames or incur significant frame delays to address the 'bad edit' problem.
Innovation Solution
A frame interpolation method that determines whether an input frame follows a long-term cadence or a bad edit cadence, applying phase coefficients from a regular phase table or a bad edit phase table respectively to generate interpolated frames, thereby reducing image jitter and frame delay without losing image information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods discard wrong image frames to handle bad edit problem, then image jitter is reduced, but image information is lost
Solution Approach 1:
The patent applies dynamics by making the frame selection process adaptive rather than fixed. The system dynamically determines whether to use long-term cadence or bad edit cadence based on real-time detection of bad edits, allowing the frame interpolation to adapt to changing input conditions and preserve relevant image information while maintaining stability.
Solution Approach 2:
The patent changes the cadence parameter from a fixed long-term cadence to a variable cadence that can switch between long-term cadence and bad edit cadence. This parameter change allows the system to adjust its frame selection strategy based on the detected input conditions, resolving the contradiction between stability and information preservation.
2Loss of information
If conventional methods use long-term cadence for frame interpolation, then image information is preserved, but image jitter occurs due to bad edits
Solution Approach 1:
The patent implements feedback by detecting bad edits in the input frames and using this detection to adjust the cadence selection. The system continuously monitors input frame stability and feedback-drivenly switches between long-term cadence and bad edit cadence, ensuring both information preservation and jitter reduction.
Solution Approach 2:
The system transitions from a static long-term cadence approach to a dynamic cadence selection mechanism that adapts to real-time input conditions. This dynamics allows the system to maintain information preservation when conditions are good while switching to jitter-reducing mode when bad edits are detected.
3Loss of information
If methods are designed to handle bad edit problem while preserving image information, then image information is retained, but frame delay increases
Solution Approach 1:
The patent segments the frame interpolation process into two distinct modes: long-term cadence mode for normal operation and bad edit cadence mode for handling problematic inputs. This segmentation allows the system to use the faster long-term cadence approach most of the time while only invoking the more careful bad edit handling when necessary, thus reducing overall frame delay while preserving information when needed.
Data Source
AI summary
A frame interpolation method for a video processor includes the steps of: determining whether a first input frame is received following a long-term cadence; generating interpolated frames by applying phase coefficients in a regular phase table corresponding to the long-term cadence when the first input frame is received following the long-term cadence; and generating interpolated frames by applying phase coefficients in a bad edit phase table corresponding to a bad edit cadence when the first input frame is received without following the long-term cadence.