bob.ip.VLSIFT¶

class bob.ip.VLSIFT((object)self, (int)height, (int)width, (int)n_intervals, (int)n_octaves, (int)octave_min[, (float)peak_thres=0.03[, (float)edge_thres=10.0[, (float)magnif=3.0]]]) → None :¶

Bases: Boost.Python.instance

Computes SIFT features using the VLFeat library

Creates an object to compute SIFT features using the VLFeat library

__init__( (object)self, (VLSIFT)other) -> None

__init__((object)self, (int)height, (int)width, (int)n_intervals, (int)n_octaves, (int)octave_min[, (float)peak_thres=0.03[, (float)edge_thres=10.0[, (float)magnif=3.0]]]) → None :¶

Creates an object to compute SIFT features using the VLFeat library

__init__( (object)self, (VLSIFT)other) -> None

Methods

__init__((object)self, (int)height, …) Creates an object to compute SIFT features using the VLFeat library

Attributes

`edge_thres`	The edge rejection threshold
`height`	The height of the image to process
`magnif`	The magnification factor (descriptor size is determined by multiplying the keypoint scale by this factor)
`n_intervals`	The number of intervals in each octave
`n_octaves`	The number of intervals in each octave
`octave_min`	The index of the minimum octave
`peak_thres`	The peak threshold (minimum amount of contrast to accept a keypoint)
`width`	The width of the image to process

__call__((VLSIFT)self, (object)src) → object :¶

Computes the SIFT features from an input image (by first detecting keypoints). It returns a list of descriptors, one for each keypoint and orientation. The first four values are the x, y, sigma and orientation of the values. The 128 remaining values define the descriptor.

__call__( (VLSIFT)self, (object)src, (object)keypoints) -> object :: Computes the SIFT features from an input image and a set of keypoints. A keypoint is specified by a 3- or 4-tuple (y, x, sigma, [orientation]). The orientation is estimated if not specified. It returns a list of descriptors, one for each keypoint and orientation. The first four values are the x, y, sigma and orientation of the values. The 128 remaining values define the descriptor.

edge_thres¶: The edge rejection threshold

height¶: The height of the image to process

magnif¶: The magnification factor (descriptor size is determined by multiplying the keypoint scale by this factor)

n_intervals¶: The number of intervals in each octave

n_octaves¶: The number of intervals in each octave

octave_min¶: The index of the minimum octave

peak_thres¶: The peak threshold (minimum amount of contrast to accept a keypoint)

width¶: The width of the image to process