The Analysis of Projective Transformation Algorithms for Image Recognition on Mobile Devices
This work addresses the need for computationally efficient and artifact-free image transformation methods in real-time mobile recognition systems, but it is incremental as it applies existing methods to a modern context without introducing new algorithms.
The paper tackles the problem of selecting projective transformation algorithms for real-time mobile image recognition by comparing known interpolation and sampling methods from the 1990s-2000s, finding that bilinear interpolation with mip-map pre-filtering or FAST sampling offers the best balance of low computational complexity and high quality without artifacts in experiments on ARM mobile processors.
In this work we apply commonly known methods of non-adaptive interpolation (nearest pixel, bilinear, B-spline, bicubic, Hermite spline) and sampling (point sampling, supersampling, mip-map pre-filtering, rip-map pre-filtering and FAST) to the problem of projective image transformation. We compare their computational complexity, describe their artifacts and than experimentally measure their quality and working time on mobile processor with ARM architecture. Those methods were widely developed in the 90s and early 2000s, but were not in an area of active research in resent years due to a lower need in computationally efficient algorithms. However, real-time mobile recognition systems, which collect more and more attention, do not only require fast projective transform methods, but also demand high quality images without artifacts. As a result, in this work we choose methods appropriate for those systems, which allow to avoid artifacts, while preserving low computational complexity. Based on the experimental results for our setting they are bilinear interpolation combined with either mip-map pre-filtering or FAST sampling, but could be modified for specific use cases.