Distinctive Image Features from Scale-Invariant Keypoints
Citations
191 citations
Cites methods from "Distinctive Image Features from Sca..."
...The method in [23] uses as quality measure the displacement of each vertex from its original position after the application of the Difference-of-Gaussians (DoG) filter [24]....
[...]
...In MeshDoG additional filtering steps are introduced after detection: a maximum number of keypoints is detected, corresponding to a percentage value of the number of vertices of the mesh; as done in [24], noncorner responses are eliminated....
[...]
191 citations
Additional excerpts
...For example, SIFT and HOG features are used for object recognition, LBP and Gabor features are used for texture and face classification....
[...]
...Actually, the histogram offers some degree of translation invariance in the extracted features, just like in hand-crafted features such as Scale-invariant Feature Transform (SIFT) [33] or Histogram of Oriented Gradients (HOG) [34], and average or max pooling process in ConvNet [20,19,35,16,36]....
[...]
...We use PHOW features (dense multi-scale SIFT descriptors), Elkan k-means for fast visual word dictionary construction, spatial histograms as image descriptors, a homogeneous kernel map to transform a Chi2 support vector machine (SVM) into a linear one, and SVM classifiers....
[...]
...LDAþSVM 80.14 Raw-pixel SVM 82.92 Raw-pixel Softmax 87.56 Raw-pixel Nearest Neighbor 89.79 VLFeat Dense-SIFT 93.58 DeepFish-SVM (our) 98:23 DeepFish-SVM-aug (our) 98:59 enlarge the training set for the species whose image number is less than 300....
[...]
191 citations
Cites methods from "Distinctive Image Features from Sca..."
...We evaluate four representative lowlevel descriptors: LBP [18], SIFT [16], Gabor [31], and Learning-based (LE) descriptor [3]....
[...]
...The descriptor-based methods [3, 5, 9, 16, 18, 24, 31, 32] and subspace-based methods [1, 14, 15, 17, 19, 26, 28, 29, 30, 35] are two representative appearance-based approaches....
[...]
...We use the default parameter described in [31] for the Gabor descriptor and 32-orientation quantization for the SIFT descriptor....
[...]
191 citations
Cites methods from "Distinctive Image Features from Sca..."
...As MMSC, five types of features were used, LBP [21], GIST [22], CENTRIST [31], Dog-SIFT [18], and HOG [9]....
[...]
...We denote SCL, SCG, SCC , SCD and SCH as the single-affinity spectral clustering methods with five different affinity matrices derived from the above five features (LBP, GIST, CENTRIST, Dog-SIFT, and HOG), respectively....
[...]
191 citations
Cites methods from "Distinctive Image Features from Sca..."
...In particular the combination of features reflecting different information, e.g., LBP and SIFT, leads to improvements....
[...]
...E.g., for ScaleInvariant Feature Transform (SIFT) [106] the number of extracted keypoints has been shown to be suitable for the task ofmorph detection [38], [78]....
[...]
...Invariant Feature Transform (SIFT) [106] the number of extracted keypoints has been shown to be suitable for the task ofmorph detection [38], [78]....
[...]
...Therefore, LBP, BSIF, SIFT, Speeded Up Robust Features (SURF) [107], Histogram ofOrientedGradients (HOG) [108] and the deep features of Openface [109] were fused and evaluated in [79]....
[...]
References
46,906 citations
16,989 citations
13,993 citations
7,057 citations
3,422 citations