Comparison of SIFT Features With Other Local Features
There has been an extensive study done on the performance evaluation of different local descriptors, including SIFT, using a range of detectors. The main results are summarized below:
- SIFT and SIFT-like GLOH features exhibit the highest matching accuracies (recall rates) for an affine transformation of 50 degrees. After this transformation limit, results start to become unreliable.
- Distinctiveness of descriptors is measured by summing the eigenvalues of the descriptors, obtained by the Principal components analysis of the descriptors normalized by their variance. This corresponds to the amount of variance captured by different descriptors, therefore, to their distinctiveness. PCA-SIFT (Principal Components Analysis applied to SIFT descriptors), GLOH and SIFT features give the highest values.
- SIFT-based descriptors outperform other local descriptors on both textured and structured scenes, with the difference in performance larger on the textured scene.
- For scale changes in the range 2-2.5 and image rotations in the range 30 to 45 degrees, SIFT and SIFT-based descriptors again outperform other local descriptors with both textured and structured scene content.
- Introduction of blur affects all local descriptors, especially those based on edges, like shape context, because edges disappear in the case of a strong blur. But GLOH, PCA-SIFT and SIFT still performed better than the others. This is also true for evaluation in the case of illumination changes.
The evaluations carried out suggests strongly that SIFT-based descriptors, which are region-based, are the most robust and distinctive, and are therefore best suited for feature matching. However, most recent feature descriptors such as SURF have not been evaluated in this study.
SURF has later been shown to have similar performance to SIFT, while at the same time being much faster.
Recently, a slight variation of the descriptor employing an irregular histogram grid has been proposed that significantly improves its performance. Instead of using a 4x4 grid of histogram bins, all bins extend to the center of the feature. This improves the descriptor's robustness to scale changes.
The Rank SIFT technique was shown to improve the performance of the standard SIFT descriptor for affine feature matching. A SIFT-Rank descriptor is generated from a standard SIFT descriptor, by setting each histogram bin to its rank in a sorted array of bins. The Euclidean distance between SIFT-Rank descriptors is invariant to arbitrary monotonic changes in histogram bin values, and is related to Spearman's rank correlation coefficient.
Read more about this topic: Scale-invariant Feature Transform
Famous quotes containing the words comparison, features and/or local:
“What is man in nature? A nothing in comparison with the infinite, an all in comparison with the nothinga mean between nothing and everything.”
—Blaise Pascal (16231662)
“It looks as if
Some pallid thing had squashed its features flat
And its eyes shut with overeagerness
To see what people found so interesting
In one another, and had gone to sleep
Of its own stupid lack of understanding,
Or broken its white neck of mushroom stuff
Short off, and died against the windowpane.”
—Robert Frost (18741963)
“Civility, which is a disposition to accommodate and oblige others, is essentially the same in every country; but good breeding, as it is called, which is the manner of exerting that disposition, is different in almost every country, and merely local; and every man of sense imitates and conforms to that local good breeding of the place which he is at.”
—Philip Dormer Stanhope, 4th Earl Chesterfield (16941773)