A Recursive Pyramidal Algorithm for Solving the Image Registration Problem
This addresses image registration for scenarios with limited training data, time, or code complexity, though it appears incremental.
The paper tackles image registration by introducing a simple, end-to-end trainable algorithm that requires minimal training data and code, achieving accurate results in some settings, such as stereo vision trained on 74 images.
The problem of image registration is finding a transformation that aligns two images, such that the corresponding points are in the same location. This paper introduces a simple, end-to-end trainable algorithm that is implementable in a few lines of Python code. The approach is shown to work with very little training data and training time, while achieving accurate results in some settings. An example application to stereo vision was trained from 74 images on a 19x15 input window. With just a dozen lines of Python code this algorithm excels in brevity and may serve as a good start in related scenarios with limitations to training data, training time or code complexity.