_rand_bbox

Random crop text box height top or bottom, we don't need image information in this step, only change box pnts :param text_box_pnts: b

Apply perspective transform on image :param img: origin numpy image :param text_box_pnts: four corner points of text

模糊图像，模拟小图片放大的效果

Create AugMix transform :param config_str: String defining configuration of random augmentation. Consists of multiple sections separated by

Create a AutoAugment transform :param config_str: String defining configuration of auto augmentation. Consists of multiple sections separate

Binarises a given low frequency image such that it has mean lambda. :param mask: Low frequency image, usually the result of `make_low_freq_image

box_score_slow: use polyon mean score as the mean score

_bitmap: single map with shape (1, H, W), whose values are binarized as {0, 1}

build index from scratch

computing IoU :param rec1: (y0, x0, y1, x1), which reflects (top, left, bottom, right) :param rec2: (y0, x0, y1, x1) :ret

Create an instance of given module class. Args: cls_name(str): Class of which to create instnce. Return: instance of type `cls_

refer to https://github.com/ilovepose/DarkPose/lib/core/inference.py

Draw word in the center of background :param word: word to draw :param font: font to draw word :param bg: background

Get bin values for discrete fourier transform of size (h, w, z) :param h: Required, first dimension size :param w: Optional, second dimensio

Resize background, let bg_width>=width, bg_height >=height, and random crop from resized background

Args: im (np.ndarray): image (np.ndarray) Returns: im_scale_x: the resize ratio of X im_scale_y:

Args: num_classes (int): number of class Returns: color_map (list): RGB color list

Get config path from CONFIGS_HOME, if not exists, download it from url.

Get demo path from DEMO_HOME, if not exists, download it from url.

Get config path from CONFIGS_HOME, if not exists, download it from url.

the highest heatvalue location with a quarter offset in the direction from the highest response to the second highest response. Args:

get predictions from score maps Args: heatmaps: numpy.ndarray([batch_size, num_joints, height, width]) Returns: