Distinctive Image Features from Scale-Invariant Keypoints
Citations
102 citations
Cites methods from "Distinctive Image Features from Sca..."
...Here, we use the GPU-based implementation [39] of the Scale-invariant feature transform (SIFT) keypoint detection [40]....
[...]
101 citations
Cites background from "Distinctive Image Features from Sca..."
...It is based on a hierarchical model, which separates a salient window into several small patches [21], and then into local features [22]....
[...]
101 citations
101 citations
Cites methods from "Distinctive Image Features from Sca..."
...To this end, we use a variant of the recent optical flow technique of [1] with constraint normalisation and SIFT matches [15] as prior....
[...]
101 citations
Cites background or methods from "Distinctive Image Features from Sca..."
..., Gabor [42], local binary patterns (LBPs) [43], and scale-invariant feature transform (SIFT) [44]) and then build a holistic scene representation...
[...]
...In this paper, the last convolutional features can be pooled at multiple scales and encoded into a single FV just like SIFT....
[...]
...Traditionally, aerial scene classification methods rely on hand-crafted features for image description, such as Gabor [42], LBPs [43], and SIFT [44]....
[...]
...The low-level methods first extract handcrafted local features (e.g., Gabor [42], local binary patterns (LBPs) [43], and scale-invariant feature transform (SIFT) [44]) and then build a holistic scene representation by local descriptor encoding methods (e.g., BoVW [17], [18], SPM [2], VLAD [19], and FV [20])....
[...]
...These feature maps generated by deep convolutional layers are analogous to the local features (e.g., Gabor, LBP, and SIFT) in traditional scene classification methods [42]–[44]....
[...]
References
46,906 citations
16,989 citations
13,993 citations
7,057 citations
3,422 citations