Title :
Localization accuracy of interest point detectors with different scale space representations
Author :
Cordes, Kai ; Rosenhahn, Bodo ; Ostermann, Jorn
Author_Institution :
Inst. fur Informationsverarbeitung, Leibniz Univ. Hannover, Hannover, Germany
Abstract :
The detection of scale invariant image features is a fundamental task for computer vision applications like object recognition or re-identification. Features are localized by computing extrema of the gradients in the Laplacian of Gaussian (LoG) scale space. The most popular detector for scale invariant features is the SIFT detector which uses the Difference of Gaussians (DoG) pyramid as an approximation of the LoG. Recently, the alternative interest point (ALP) detector demonstrated its strength in fast computation on highly parallel architectures like the GPU. It uses the LoG scale space representation for the localization of interest points. This paper evaluates the localization accuracy of ALP in comparison to SIFT. By using synthetic images, it is demonstrated that both localization approaches show a systematic error which is dependent on the subpixel position of the feature. The error increases with the scale of the detected feature. However, using the LoG instead of the DoG representation reduces the maximum systematic error by 77 %. For the evaluation with natural images, benchmark data sets are used. The repeatability criterion evaluates the accuracy of the detectors. The LoG based detector results in up to 16 % higher repeatability. The comparisons are completed with a reference feature localization which uses a signal based approach for the gradient approximation. Based on this approach, a new feature selection criterion is proposed.
Keywords :
approximation theory; computer vision; gradient methods; image representation; DoG representation; GPU; Laplacian of Gaussian scale space; LoG based detector; LoG scale space representation; SIFT detector; alternative interest point; computer vision; difference of Gaussians; feature selection criterion; gradient approximation; interest point detectors; localization accuracy; localization approaches; object recognition; object reidentification; reference feature localization; repeatability criterion; scale invariant features; scale invariant image features detection; scale space representations; signal based approach; synthetic images; Accuracy; Approximation methods; Benchmark testing; Detectors; Feature extraction; Shape; Systematics;
Conference_Titel :
Advanced Video and Signal Based Surveillance (AVSS), 2014 11th IEEE International Conference on
Conference_Location :
Seoul
DOI :
10.1109/AVSS.2014.6918676