Iterative Motion Estimation Using Binary Mask Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion estimation methods in image sequences, particularly when dealing with mobile objects or low-contrast scenes, often result in unreliable global motion estimation, which affects the accuracy of subsequent dense motion estimation.
Innovation Solution
An iterative method combining global and dense motion estimation, where a binary image mask is applied to refine the estimation process, updating the mask based on the second estimated motion to improve reliability and convergence, even when initial global motion estimation is unreliable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a global motion estimation is performed first, then a dense motion estimation can be carried out on compensated images, but the estimation becomes unreliable when the scene contains mobile objects or low contrast
Solution Approach 1:
The patent divides the image into multiple regions using a binary mask, separating mobile objects from the static background. This segmentation allows different motion estimation strategies to be applied to different regions, improving overall accuracy while maintaining reliability in challenging scenes with mobile objects or low contrast
Solution Approach 2:
The patent implements an iterative process where the binary mask is dynamically updated based on the second estimated motion from dense motion estimation. This dynamic adaptation allows the system to refine its understanding of mobile objects and background regions across multiple iterations, improving reliability in difficult scenarios
2Reliability
If a binary image mask is applied to determine global motion estimation, then reliability improves, but device complexity increases due to iterative processing
Solution Approach 1:
The patent applies a binary mask that selectively processes only certain regions of the image (mobile objects vs. background) rather than the entire image. This partial action reduces the effective complexity of the iterative process while maintaining reliability in critical regions
Solution Approach 2:
The patent performs a preliminary global motion estimation before the dense motion estimation, using this initial result to guide the creation of the binary mask. This preliminary action simplifies the subsequent iterative process by providing a starting point that reduces the search space and computational complexity
Data Source
AI summary
The invention relates to a method which uses a series of images sensed by an image sensor including at least one preceding image and one following image to estimate movement. A first estimated movement is initially obtained by estimating the total movement from the preceding image to the following image. Next, an image compensated according to the first estimated movement is obtained from either one of the preceding and following images. Then, a second estimated movement is obtained by estimating dense movement between the compensated image and the other from the preceding and following images. Next, a residual value of global movement is determined. Finally, if the residual value is lower than a threshold value the second estimated movement is provided; otherwise, the preceding steps are repeated. The first estimated movement is determined by applying a binary image mask, and if during step /e/ the steps /a/ to /e/ are repeated, said steps are performed by applying a binary image mask updated according to the second estimated movement.