Generate anchor boxes

Methods to generate anchor boxes of different aspect ratios.

To generate anchor boxes, we need three basic information:

Input image size, image_sz: To position our anchor boxes within the maximum coordinates (width, height) of the image.
Feature map size, feature_sz: Feature map is the size (width, height) of the output of a convolutional operation. A \(10\times10\) feature map would mean \(10\times10\) local receptive field locations can be traced back into the input image. These 100 receptive field locations (\(10\times10=100\)) in the input image would act as our initial anchor box candidates.

Aspect ratio of anchor boxes, asp_ratio: To generate anchor boxes with different width to height ratio (default asp_ratio=1).

bx

 bx (image_sz:(<class'int'>,<class'tuple'>),
     feature_sz:(<class'int'>,<class'tuple'>), asp_ratio:float=None,
     clip:bool=True, named:bool=True, anchor_sfx:str='a',
     min_visibility:float=0.25)

Calculate anchor box coords given an image size and feature size for a single aspect ratio.

	Type	Default	Details
image_sz	(<class ‘int’>, <class ‘tuple’>)		image size (width, height)
feature_sz	(<class ‘int’>, <class ‘tuple’>)		feature map size (width, height)
asp_ratio	float	None	aspect ratio (width:height), by default None
clip	bool	True	whether to apply np.clip, by default True
named	bool	True	whether to return (coords, labels), by default True
anchor_sfx	str	a	suffix anchor label with anchor_sfx, by default “a”
min_visibility	float	0.25	minimum visibility dictates the condition for a box to be considered valid. The value corresponds to the ratio of expected area of an anchor box to the calculated area after clipping to image dimensions., by default 0.25
Returns	typing.Union[typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[typing.Any]]]]], numpy.typing._array_like._SupportsArray[numpy.dtype], typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]], typing.Sequence[typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]]], typing.Sequence[typing.Sequence[typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]]]], typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]]]]], bool, int, float, complex, str, bytes, typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]], typing.Sequence[typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]]], typing.Sequence[typing.Sequence[typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]]]], typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]]]]]]		anchor box coordinates in `pascal_voc` format if named=True, a list of anchor box labels are also returned.

coords_1, labels_1 = bx(100, 10, 0.5)

Usually multiple anchor boxes with different feature_sz and asp_ratio are needed. This requirement arises in the case of multiscale object detection.

For multiscale object detection, feature maps from different convolution operations of the network are used to trace back into the input image, to generate anchor boxes. The bxs method of pybx provides this possibility.

source

bxs

 bxs (image_sz:(<class'int'>,<class'tuple'>), feature_szs:list=None,
      asp_ratios:list=None, named:bool=True, **kwargs)

Calculate anchor box coords given an image size and multiple feature sizes for mutiple aspect ratios.

	Type	Default	Details
image_sz	(<class ‘int’>, <class ‘tuple’>)		image size (width, height)
feature_szs	list	None	list of feature map sizes, each feature map size being an int or tuple, by default [(8, 8), (2, 2)]
asp_ratios	list	None	list of aspect ratios for anchor boxes, each aspect ratio being a float calculated by (width:height), by default [1 / 2.0, 1.0, 2.0]
named	bool	True	whether to return (coords, labels), by default True
kwargs
Returns	typing.Union[typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[typing.Any]]]]], numpy.typing._array_like._SupportsArray[numpy.dtype], typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]], typing.Sequence[typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]]], typing.Sequence[typing.Sequence[typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]]]], typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[numpy.typing._array_like._SupportsArray[numpy.dtype]]]]], bool, int, float, complex, str, bytes, typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]], typing.Sequence[typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]]], typing.Sequence[typing.Sequence[typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]]]], typing.Sequence[typing.Sequence[typing.Sequence[typing.Sequence[typing.Union[bool, int, float, complex, str, bytes]]]]]]		anchor box coordinates in pascal_voc format if named=True, a list of anchor box labels are also returned.

coords, labels = bxs(100, [10, 8, 5, 2], [1, 0.5, 0.3])

coords.shape, len(labels)

((579, 4), 579)

All methods work with asymetric image_sz (and or feature_szs as well):

coords, labels = bxs((100, 200), [10, 8, 5, 2], [1, 0.5, 0.3])

coords.shape, len(labels)

((654, 4), 654)