We present an experimental comparison of the performance of representative saliency detectors from three guiding principles for the detection of salient image locations: locations of maximum stability with respect to image transformations, locations of greatest image complexity, and most discriminant locations. It is shown that discriminant saliency performs better in terms of 1) capturing relevant information for classification, 2) being more robust to image clutter, and 3) exhibiting greater stability to image transformations associated with variations of 3D object pose. We then investigate the dependence of discriminant saliency on the underlying set of candidate discriminant features, by comparing the performance achieved with three popular feature sets: the discrete cosine transform, a Gabor, and a Haar wavelet decomposition. It is show that, even though different feature sets produce equivalent results, there may be advantages in considering features explicitly learned from examples of the image classes of interest.
展开▼
