One-Dimensional Segmentation for Motion Estimation Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Motion estimation algorithms, such as block matching, often lack a holistic view, leading to artifacts like periodic pattern inconsistencies, object breakups, and motion field leakage, which are challenging to correct in real-time environments.
Innovation Solution
A computer-implemented method for segmenting images that groups adjacent blocks into pseudo-objects with uniform motion vectors, using similarity metrics and cumulative motion vector calculations to identify and correct discontinuities, thereby reducing artifacts in the motion field.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If block matching algorithms are used for motion estimation, then fast/real-time processing is achieved, but motion field artifacts such as periodic pattern inconsistencies, object breakups, and motion field leakage occur
Solution Approach 1:
The patent applies segmentation by dividing the image into blocks and further grouping adjacent similar blocks into superblocks. This hierarchical segmentation allows motion estimation to be performed at multiple levels, maintaining real-time processing while reducing artifacts through coordinated motion vector calculation across block boundaries.
Solution Approach 2:
The patent merges adjacent similar blocks into superblocks, treating them as unified entities for motion estimation. This combining approach ensures that blocks belonging to the same object share consistent motion vectors, preventing object breakup artifacts while maintaining processing efficiency.
2Productivity
If each block is matched individually in block search, then processing speed is maintained, but a holistic view of object motion is lost leading to motion inconsistencies
Solution Approach 1:
The patent introduces a new dimension of organization by grouping blocks into superblocks based on spatial adjacency and similarity. This additional organizational layer enables the system to maintain individual block processing speed while simultaneously achieving holistic object motion understanding through superblock-level coordination.
Solution Approach 2:
The patent uses motion vectors from reference blocks to initialize and constrain motion estimation in adjacent blocks within the same superblock. This copying approach propagates coherent motion information across the superblock while allowing local adjustments, maintaining both speed and motion consistency.
3Reliability
If post processing algorithms such as FFT with motion collision detector are used to fix periodic pattern artifacts, then artifact correction is achieved, but implementation complexity increases making real-time execution challenging
Solution Approach 1:
The patent performs preliminary action by detecting and correcting motion field inconsistencies during the motion estimation process itself, before final motion compensation. By identifying and resolving artifacts at the superblock level during estimation, the need for complex post-processing algorithms is eliminated, maintaining real-time performance.
Solution Approach 2:
The patent extracts and addresses artifact problems directly within the motion estimation framework by identifying inconsistent motion vectors at superblock boundaries and correcting them through optimized motion vector selection, rather than applying separate post-processing correction algorithms.
Data Source
AI summary
A method for segmenting images for estimating motion includes scanning block rows of an image to determine whether image blocks that are adjacent within a row are similar, and grouping similar image blocks into a row segment. If a block in a row is not similar to a preceding adjacent block, the block is used to initialize a new row segment. The method further includes calculating data for each row segment in an image, and using this data to determine whether a motion vector field for a row segment is consistent, whether discontinuities exist in a row segment that can be corrected, or whether discontinuities exist in a row segment that cannot be corrected, and correcting those row segment discontinuities that can be fixed to remove artifacts from a set of motion vectors associated with said row segment.