Motion Vector Estimation Using Image Hash Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for motion vector estimation in image sequences are computationally expensive, especially in rasterized image processing applications, due to the need to characterize the entire image on each frame, which hampers efficient communication across band-limited networks in remote computing environments.
Innovation Solution
A method and apparatus for motion vector estimation that preprocesses images to generate hash values for spatially oriented blocks of pixels, allowing for efficient computation of motion vectors by comparing hash values between reference and updated image frames, thereby reducing the computational load and enabling efficient image compression and transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature detection methods search a reference image for features located on a subsequent input image, then motion vectors can be estimated, but computation becomes expensive due to swapping areas of reference image in and out of memory
Solution Approach 1:
The image is divided into fixed-size blocks (e.g., 16x16 pixels) that are processed independently. Each block generates a hash value representing its content, allowing parallel processing and eliminating the need to swap large image areas in and out of memory. This segmentation reduces computation while maintaining motion estimation accuracy.
Solution Approach 2:
A hash value is introduced as an intermediary representation of image block content. Instead of directly comparing pixel data or swapping image areas in memory, the system compares compact hash values (e.g., 32-bit integers) that uniquely represent each block. This intermediary dramatically reduces memory access requirements and computation cost.
2Measurement precision
If the entire image is characterized on each frame in rasterized image processing applications, then motion vectors can be computed, but computational expense increases significantly
Solution Approach 1:
The system extracts only the essential information from each image block by generating a compact hash value that represents the block's content. This extraction eliminates redundant pixel data processing while retaining sufficient information for accurate motion vector estimation, thereby improving processing speed.
Solution Approach 2:
The representation of image data is changed from full pixel values to compressed hash values. This parameter transformation reduces the data size from thousands of pixels per block to a single integer, enabling rapid comparison and computation without sacrificing motion estimation accuracy.
3Measurement precision
If large search areas are used to calculate long motion vectors, then accuracy improves, but computation time increases due to the need to scan entire updated image regions
Solution Approach 1:
The system creates a copy of the reference frame's hash values and compares them directly with hash values from the current frame. This copying approach allows efficient comparison across large search areas without physically accessing or swapping large image regions in memory, reducing computation time while maintaining accuracy for long motion vectors.
Data Source
AI summary
A method and apparatus for motion vector estimation for a plurality of image frames including retrieving a reference hash value associated with a reference frame, wherein the reference hash value represents a spatially oriented block of pixels, generating a plurality of image hash values including a first hash value for a first area represented by a spatially oriented block of pixels of an image frame and computing a motion vector estimate using a plurality of first hash values and the reference hash value is disclosed.