Calculate the pose accuracy of PCK for each individual keypoint and the averaged accuracy across all keypoints for coordinates. Note:

Register the given function as a handler that this transform will use for a specific data type. Args: data_type

In-place transform all attributes of this class. By "in-place", it means after calling this method, accessing an attribute such

Return an activation function given a string

Returns accuracy score evaluation result. - overall accuracy - mean accuracy - mean IU - fwavacc

Uniform float random number between low and high.

Recursively apply to all objects in different kinds of container types that matches a type function.

Multiple with alpha and add beat with clip.

:meth: draw bounding box on the given image :param img: the image to draw :param bold: if bold the bounding box or not (defau

Evaluate coco keypoint results. The pose prediction results will be saved in `${res_folder}/result_keypoints.json`. Note:

:meth: read the object from a dict

A wrapper around :function:`torch.nn.functional.grid_sample` to support 3D point_coords tensors. Unlike :function:`torch.nn.functional.grid_s

Split a tuple into a list of tensors and the rest with information for later reconstruction. When called with a tensor X, will return: (x

Zero fp32 and fp16 parameter grads.

Args: image (ndarray): (H,W) or (H,W,C) ndarray of type uint8 in range [0, 255], or floating point in range [0, 1

Creates the matcher Params: cost_class: This is the relative weight of the classification error in the matching cost

Create the criterion. Parameters: num_classes: number of object categories, omitting the special no-object category ma

Keypoint evaluation using COCOAPI.

To calculate the affine matrix, three pairs of points are required. This function is used to get the 3rd point, given 2D points a & b. The 3r

Get keypoint predictions from score maps. Note: batch_size: N num_keypoints: K heatmap height: H heatmap width: W

A context which ignores CUDA OOM exception from pytorch.

Generate the target heatmap via "Megvii" approach. Args: cfg (dict): data config joints_3d: np.ndarray ([num_joints,

Generate the target heatmap via "Megvii" approach. Args: cfg (dict): data config joints_3d: np.ndarray ([num_joints, 3

Generate the target heatmap via "MSRA" approach. Args: cfg (dict): data config joints_3d: np.ndarray ([num_joints, 3]

Generate the target heatmap via "MSRA" approach. Args: cfg (dict): data config joints_3d: np.ndarray ([num_joints, 3])

Distribution aware coordinate decoding method. Note: heatmap height: H heatmap width: W Args: heatmap (np.ndarray[H,

Wrap Transform into Augmentation. Private, used internally to implement augmentations.

Generate the target heatmap via 'UDP' approach. Paper ref: Huang et al. The Devil is in the Details: Delving into Unbiased Data Processing

Generate the target heatmap via 'UDP' approach. Paper ref: Huang et al. The Devil is in the Details: Delving into Unbiased Data Processing

Write results into a json file.

This encodes bbox(x,y,w,w) into (center, scale) Args: x, y, w, h Returns: tuple: A tuple containing center a

This encodes bbox(x,y,w,w) into (center, scale) Args: x, y, w, h Returns: tuple: A tuple containing center a

This encodes bbox(x,y,w,w) into (center, scale) Args: x, y, w, h Returns: tuple: A tuple containing center a

:param img: :param src_points: :param dst_points: :param shape: :param borderMode: cv2.BORDER_CONSTANT

:param img: :param src_points: :param dst_points: :param shape: :param borderMode: cv2.BORDER_CONSTANT

Apply an affine transformation to the points. Args: pt (np.ndarray): a 2 dimensional point to be transformed trans_mat (np.ndarra

Apply the transform on an axis-aligned box. By default will transform the corner points and use their minimum/maximum to create a new

Apply the transform on coordinates. Args: coords (ndarray): floating point array of shape Nx2. Each row is (x, y).

Apply the transform on an image. Args: img (ndarray): of shape NxHxWxC, or HxWxC or HxW. The array can be

Compute the DICE loss, similar to generalized IOU for masks Args: inputs: A float tensor of arbitrary shape. The pred

Compute the DICE loss, similar to generalized IOU for masks Args: inputs: A float tensor of arbitrary shape. The