utils::image_processing Namespace Reference

Functions
def	resize
def	tensor_vstack
def	transform
def	transform_inverse

---

Function Documentation

def utils.image_processing.resize	(		im,
			target_size,
			max_size
	)

only resize input image to target size and return scale
:param im: BGR image input by opencv
:param target_size: one dimensional size (the short side)
:param max_size: one dimensional max size (the long side)
:return:

Definition at line 5 of file image_processing.py.

def utils.image_processing.tensor_vstack	(		tensor_list,
			pad = 0
	)

vertically stack tensors
:param tensor_list: list of tensor to be stacked vertically
:param pad: label to pad with
:return: tensor with max shape

Definition at line 65 of file image_processing.py.

def utils.image_processing.transform	(		im,
			pixel_means,
			need_mean = False
	)

transform into mxnet tensor
subtract pixel size and transform to correct format
:param im: [height, width, channel] in BGR
:param pixel_means: [[[R, G, B pixel means]]]
:return: [batch, channel, height, width]

Definition at line 24 of file image_processing.py.

def utils.image_processing.transform_inverse	(		im_tensor,
			pixel_means
	)

transform from mxnet im_tensor to ordinary RGB image
im_tensor is limited to one image
:param im_tensor: [batch, channel, height, width]
:param pixel_means: [[[R, G, B pixel means]]]
:return: im [height, width, channel(RGB)]

Definition at line 45 of file image_processing.py.