update intrinsics for cropped images

Append new node information to the route information. Args: node (Node, optional): An instance of Node that provides basic

Project points from world coordinates to pixel coordinates.

Wrap the FP32 model to FP16. 1. Convert FP32 model to FP16. 2. Remain some necessary layers to be FP32, e.g., normalization layers. Args

Initializes Unit3D module.

Extract the first or third Euler angle from the two members of the matrix which are positive constant times its sine and cosine. Args:

Extract the first or third Euler angle from the two members of the matrix which are positive constant times its sine and cosine. Args:

Return a Body object with the specified joint names.

calculate the angle of vectors v1(constructed by points p2 and p1) and v2(constructed by points p4 and p3)

Calculate the aesthetic score for a single image. Args: image (torch.Tensor): The processed image tensor. Returns:

Calculate the CLIP score for a single image and prompt. Args: image (torch.Tensor): The processed image tensor. promp

Calculate the reward score for a single image and prompt. Args: image (torch.Tensor): The processed image tensor. pro

Calculate the HPS score for a single image and prompt. Args: image (torch.Tensor): The processed image tensor. prompt

Calculate the score for a single image and prompt. Args: prompt (str): The prompt text. image (torch.Tensor): The pro

Calculate the score for a single image and prompt. Args: image (torch.Tensor): The processed image tensor. prompt (st

Transform the joints from their camera coordinates to their pixel coordinates. Note: - N: number of joints Args:

Check if the origin_imgs are flipped correctly.

Check if the origin_imgs are flipped correctly.

Compute the Intersection over Union (IoU) between two boxes . Args: bboxA (list): The first bbox info (left, top, right, bottom, score).

Compute padding tuple.

Construct the in-plane rotation matrix. Args: rot (float): Rotation angle (degree). size (int): The size of the rotation matrix.

Convert the syncBNs into normal BN3ds.

Create a superset of inputs needed to run backbone. Args: input_shape (tuple): input batch dimensions. Default: (1, 3, 64, 64

Create a superset of inputs needed to run mesh head. Args: input_shape (tuple): input batch dimensions. Default: (1, 3, 64, 6

Create a superset of inputs needed to run backbone. Args: input_shape (tuple): input batch dimensions. Default: (1, 3, 64, 64

Create a superset of inputs needed to run test or train batches. Args: input_shape (tuple): input batch dimensions

Additional filtering is applied to filter out-of-box samples. Modifications made by Zexin He.

Freeze parameters.

Freeze parameters.

Freeze parameters.

Freeze parameters.

Modulate heatmap distribution with Gaussian. sigma = 0.3*((kernel_size-1)*0.5-1)+0.8 sigma~=3 if k=17 sigma=2 if k=11; sigma~=1.5

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 anno for a single image.

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

r"""Adapted from: https://github.com/Project-MONAI/MONAI/blob/master/monai/networks/layers/convutils.py

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

Positionally encode points that are normalized to [0,1].

Cauculate the Area Under the Curve (AUC) computed for a range of 3DPCK thresholds. Args: keypoint_results (list): Keypoin

Cauculate Percentage of Correct Keypoints (3DPCK) w. or w/o Procrustes alignment. Args: keypoint_results (list): Keypoint

Keypoint evaluation. Args: res_file (str): Json file stored prediction results. metrics (str | list[str]): Metric to