Process mmdet results, and return a list of bboxes. Args: mmdet_results (list|tuple): mmdet results. cat_id (int): category id (d

Project 3D joint coordinates to 2D image plane using camera intrinsic parameters. Args: x (torch.Tensor[N, K, 3]): 3D joi

Perform orthographic projection of 3D points using the camera parameters, return projected 2D points in image plane. Note:

Convert pytorch model to onnx model. Args: model (:obj:`nn.Module`): The pytorch model to be exported. input_shape (tuple[int]):

Replace fp16 linear with quantized linear

using voxel-based skinning method vs: [B n c]

Return a fully connected network.

Rotate a point by an angle. Args: pt (list[float]): 2 dimensional point to be rotated angle_rad (float): rotation angle by radian

normalize an array (along the last dim). Args: x (Union[Tensor, ndarray]): x, [..., C] eps (float, optional): eps. Defaults to 1e

Return depths of approximately uniformly spaced samples along rays.

scale bbox with scale factor

Set requies_grad for all the networks. Args: nets (nn.Module | list[nn.Module]): A list of networks or a single network.

Get human readable file size. Args: size (int): File size. suffix (str): Suffix. Default: 'B'. Return: str: Formated

Soft OKS NMS implementations. Args: kpts_db thr: retain oks overlap < thr. max_dets: max number of detections to keep.

n_views: number of surrounding views radius: camera dist to center height: height of the camera return: (M, 3, 4)

It will first crop input images to tiles, and then process each tile. Finally, all the processed tiles are merged into one images. Mo

Convert the model into training mode.

Convert the model into training mode.

Convert the model into training mode while keep layers freezed.

Train step function. In this function, the detector will finish the train step following the pipeline: 1. get fake and r

_summary_ Args: smplx_data (_type_): e.g., body_pose:[B*Nv, 21, 3], betas:[B*Nv, 100]

Transform the mean 3D vertices to posed vertices from the neutral pose. mean_3d (torch.Tensor): Mean 3D vertices with shape [B*N

Forward function for evaluation. Args: data_batch (dict): Contain data for forward. Returns: dict: Contain t

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

Copy a model state_dict to cpu. Args: state_dict (OrderedDict): Model weights on GPU. Returns: OrderedDict: Model weights on

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)