Convert box coordinates to center and scale. adapted from https://github.com/Microsoft/human-pose-estimation.pytorch

Flip 3d joints. Parameters ---------- joints_3d : numpy.ndarray Joints in shape (num_joints, 3, 2) width : int Image

Transform ndarray image to torch tensor. Parameters ---------- img: numpy.ndarray An ndarray with shape: `(H, W, 3)`. Return

Scatter function. Args: inputs (Tensor): Input Tensor. kwargs (dict): Args for ``mmcv.parallel.scatte

Performs Linear Blend Skinning with the given shape and pose parameters Parameters ---------- betas : torch.tensor BxNB

Criterion of wingloss. Note: - batch_size: N - num_keypoints: K - dimension of keypoints: D (D=2 or D=3)

Update a new data sample.

保持检测框中心点位置不变调整检测框比例为cfg.input_img_size ration: 扩大检测框大小

Update currently maintained parameters. Call this every time the parameters are updated, such as the result of the `optimizer.step()` ca

nn.Module: the normalization layer named "norm1"

Flip 3d xyz joints. Parameters ---------- joints_3d : numpy.ndarray Joints in shape (num_joints, 3) joint_pairs : list

Use `get_logger` method in mmcv to get the root logger. The logger will be initialized if it has not been initialized. By default a StreamHan

Load checkpoint from a file or URI. Args: model (Module): Module to load checkpoint. filename (str): Accept local filepath, URL, `

nn.Module: the normalization layer named "norm2"

Do Procrustes alignment and compute reconstruction error.

Return vector c that perpendicular to (a - b).

Get the expansion of a residual block. The block expansion will be obtained by the following order: 1. If ``expansion`` is given, just retur

Add dpg for data augmentation, including random crop and random sample.

Clip bounding box with format (xmin, ymin, xmax, ymax) to specified boundary. All bounding boxes will be clipped to the new region `(0, 0, width,

Convert bounding boxes from format (x, y, w, h) to (xmin, ymin, xmax, ymax) Parameters ---------- xywh : list, tuple or numpy.ndarray

Performs Linear Blend Skinning with the given shape and pose parameters Parameters ---------- betas : torch.tensor BxNB

Performs Linear Blend Skinning with the given shape and pose parameters Parameters ---------- betas : torch.tensor BxNB

Calculates the 3D joint locations from the vertices Parameters ---------- J_regressor : torch.tensor JxV The regressor array tha

SMPL model constructor Parameters ---------- model_path: str The path to the folder or to the fi

SMPL model constructor Parameters ---------- model_path: str The path to the folder or to the fi

SMPL model constructor Parameters ---------- model_path: str The path to the folder or to the fi

SMPL model constructor Parameters ---------- model_path: str The path to the folder or to the fi

Recursively convert Tensor in inputs from src_type to dst_type. Args: inputs: Inputs that to be casted. src_type (torch.dtype): S

Copy current parameters into given collection of parameters. Args: parameters: Iterable of `torch.nn.Parameter`; the parameters to be

Compute the faces, barycentric coordinates for the dynamic landmarks To do so, we first compute the rotation of the neck around the y-axis

Calculates landmarks by barycentric interpolation Parameters ---------- vertices: torch.tensor BxVx3, dtype = torch.float32

Build loss.

nn.Module: the normalization layer named "norm3"

Update currently maintained parameters. Call this every time the parameters are updated, such as the result of the `optimizer.step()` ca

Calculates the rotation matrices for a batch of rotation vectors Parameters ---------- rot_vecs: numpy.ndarray Nx3

Calculates the per vertex displacement due to the blend shapes Parameters ---------- betas : torch.tensor Bx(num_betas) Blend s