Frame Rate Interpolation via Forward-Backward Motion Trajectory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for improving frame rates in images suffer from inaccurate motion vector estimation due to limitations in motion estimating algorithms and lack of frame information, leading to deterioration in image quality, especially in high-speed motion videos displayed on LCD devices with limited response speed.
Innovation Solution
A frame rate improving apparatus and method that determines forward feature point trajectory information, performs block-based backward motion estimation, and generates intermediate frames using motion vectors, with additional feature point extraction in cases of scene changes or insufficient feature points, and detects occlusion areas using Sum of Absolute Differences (SAD) to prevent image quality deterioration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional motion vector estimation is used to generate intermediate frames, then frame rate is improved, but image quality deteriorates due to inaccurate motion vector estimation
Solution Approach 1:
The patent divides the frame into multiple blocks and performs motion estimation on each block separately. It also segments the motion estimation process into forward estimation (from reference frame to current frame) and backward estimation (from current frame to previous frame), allowing more precise local motion analysis and improving both frame rate and motion vector accuracy
Solution Approach 2:
The patent adds temporal dimension by utilizing both forward and backward motion estimation across multiple frames. By tracing feature points through time in both directions and combining this information, the system achieves more accurate motion vectors while maintaining high frame rate
2Reliability
If feature points are traced forward through multiple frames to determine trajectory information, then motion information confidence is enhanced, but computational complexity increases
Solution Approach 1:
The patent performs forward feature point tracing and trajectory determination as a preliminary step before backward motion estimation. By pre-computing the forward trajectories and storing this information, the subsequent backward estimation can proceed more efficiently, reducing overall computational complexity while maintaining high motion information confidence
Solution Approach 2:
The system uses the forward trajectory information to guide and constrain the backward motion estimation process. The pre-computed trajectories serve as self-generated reference data that simplifies the backward estimation, reducing the computational burden while enhancing reliability
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A frame rate improving apparatus and method use a motion trajectory. The frame rate improving apparatus determines forward feature point trajectory information through a feature point, and generates a new intermediate frame by performing backward motion estimation through the feature point trajectory information, thereby generating an interpolated image with high quality.