Partial Motion Vector for Frame Rate Conversion Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current frame rate converters (FRCs) introduce artifacts like halo and break-up due to hardware limitations and power constraints, especially in mobile devices, when interpolating frames for low-frame-rate video sources.
Innovation Solution
A partial motion vector (MV) based solution is implemented, where partial MVs are generated and packed with compressed video to provide hints or replace calculated MVs in FRC ICs, improving frame rate conversion quality by reducing halo and break-up issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If FRC IC uses hardware algorithms for frame interpolation, then frame rate conversion is achieved, but artifacts such as halo and break up are introduced
Solution Approach 1:
The patent applies preliminary action by pre-calculating motion vectors during video encoding and embedding them in the bitstream. This allows the FRC IC to receive pre-processed motion information before frame interpolation, reducing the computational burden during real-time conversion and improving both speed and accuracy of artifact reduction.
Solution Approach 2:
The patent introduces motion vectors as an intermediary element that mediates between the original video frames and the interpolated frames. These motion vectors serve as a bridge, providing guidance information that helps the FRC IC perform more accurate pixel assignment during frame rate conversion, thereby reducing halo and break-up artifacts.
2Manufacturing precision
If FRC IC performs motion estimation for each interpolated phase, then interpolation accuracy is improved, but computational complexity and power consumption increase
Solution Approach 1:
The patent applies partial action by performing motion estimation only for a subset of reference frames rather than all interpolated phases. Motion vectors from selected reference frames are reused for multiple intermediate frames, reducing the total number of motion estimation operations while maintaining sufficient interpolation accuracy for visual quality.
Solution Approach 2:
The patent uses preliminary motion estimation results from encoding phase to guide the interpolation process. By pre-calculating and embedding motion vectors, the system avoids performing full motion estimation during real-time FRC operation, significantly reducing computational complexity while maintaining accuracy.
3Manufacturing precision
If FRC IC performs motion estimation for each interpolated phase, then interpolation accuracy is improved, but power consumption increases
Solution Approach 1:
The patent reduces power consumption by performing motion estimation only partially - specifically for selected reference frames rather than all interpolated phases. The embedded motion vectors from these partial estimations are reused across multiple frames, maintaining interpolation quality while dramatically reducing the energy required for motion computation in mobile devices.
4Manufacturing precision
If partial MVs are packed with compressed video, then FRC performance is improved, but data transmission requirements increase
Solution Approach 1:
The patent extracts only the essential motion vector information needed for FRC from the full video data, packing selectively only these partial MVs with the compressed video stream. This extraction approach includes only the critical motion parameters required for interpolation, avoiding transmission of redundant data while still improving FRC performance.
Data Source
AI summary
A system can include a memory and a motion vector calculator (MVC) configured to receive a block from the memory, the MVC including an MVC core configured to generate a partial motion vector (MV) for the block, a halo reduction block configured to perform halo reduction on the partial MV, and an MV post-processing block configured to smooth the partial MV and output the smoothed partial MV.