Distinctive Image Features from Scale-Invariant Keypoints
Citations
200 citations
Cites background from "Distinctive Image Features from Sca..."
...The invariance to rotation is often obtained by normalizing the region with respect to rotation and then computing the description [27], [30], [31]....
[...]
...2, as suggested in [27], are set to the threshold....
[...]
...A wide variety of detectors and descriptors have already been proposed in the literature [27], [29]–[31]....
[...]
...Finding correspondences between two images of the same scene or object is part of many computer vision applications, such as object recognition [27] and wide baseline matching [28]....
[...]
...Lowe [27] proposed a SIFT, which combines a scale-invariant region detector and a descriptor based on gradient distribution in the detected regions....
[...]
199 citations
Cites background or methods from "Distinctive Image Features from Sca..."
...The classical gradient based corner detection leads to gradient based features such as SIFT [39], SURF [40]....
[...]
...Contour based DoG-curve [35], ANDD [36],Hyperbola Fitting [37], ACJ [38] Blob detection Interest point PDE based LoG, DoG, DoH, Hessian-Laplacian [7], SIFT [39], SURF [40], Cer-SURF [41], DART [42], Rank-SIFT [43], RLOG [44], MO-GP [45], KAZE [46], A-KAZE [47], WADE [48] Template based ORB [49], BRISK [50], FREAK [51]...
[...]
...According to the experiments, the computational efficiency is lifted up meanwhile the repeatability of SURF [40], CerSURF [41], DART [42] is comparable of classical SIFT [39] in viewpoint, scale, illumination changes....
[...]
...SIFT (Scale Invariant Feature Transform) [39] locates interest points with DoG pyramid and Hessian matrix....
[...]
...NMX [20] Pb edge responses, SIFT [39] F-measure AnyBoost...
[...]
199 citations
Cites methods from "Distinctive Image Features from Sca..."
...For SIFT and HOG features, we apply dense sampling strategy with 4 4 step size and then generate the BoWs features....
[...]
...We extract four types of features for each patch, including color, SIFT [24], HOG [8] and location features....
[...]
...LabelMe subset (also known as SIFT Flow dataset) [19] includes 2688 images....
[...]
...The dictionary sizes for SIFT and HOG are both set as 300 [5]....
[...]
198 citations
Cites methods from "Distinctive Image Features from Sca..."
...We extracted 5 datasets of 17000, 50000, 118000 and 204000 (128-dimensional) SIFT descriptors2 [Lowe, 2004]....
[...]
198 citations
Cites methods from "Distinctive Image Features from Sca..."
...To estimate the parameters of a single moving camera we apply a feature-based approach, where corresponding feature points are determined in consecutive frames with the KLT-Tracker [21] or SIFT matching [10]....
[...]
References
46,906 citations
16,989 citations
13,993 citations
7,057 citations
3,422 citations