move all tensor attributes to device. Args: device (torch.device): target device. Returns: Mesh: self.

Returns a dictionary of info about the object.

load mesh from path. Args: path (str): path to mesh file, supports ply, obj, glb. clean (bool, optional): perform mes

Change the device of the data inplace. Example: >>> self = SamplingResult.random() >>> print(f'self = {self.to(None)}

Update a new data sample.

Build loss.

Initialize the weights in backbone. Args: pretrained (str, optional): Path to pre-trained weights. Defaults to Non

Convert the model into training mode.

Dump model and checkpointables to a file. Args: name (str): name of the file. kwargs (dict): extra arbitrary

write the mesh to a path. Args: path (str): path to write, supports ply, obj and glb.

Initialize a pose model from config file. Args: config (str or :obj:`mmcv.Config`): Config file path or the config object.

This encodes bbox(x,y,w,h) into (center, scale) Args: x, y, w, h (float): left, top, width and height padding (float)

nn.Module: the normalization layer named "norm1"

nn.Module: the normalization layer named "norm2"

Decode hand keypoint, relative root depth and hand type. Args: img_metas (list(dict)): Information about data augmentation

Get the frame image. Returns: array: The frame image.

Inference a single image with a list of person bounding boxes. Note: - num_people: P - num_keypoints: K - bbox height: H

sort kpts and remove the repeated ones.

Write results into a json file.

Build a dataset from config dict. Args: cfg (dict): Config dict. It should at least contain the key "type". default_args (dict, o

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

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

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

Inference function. Returns: output (list[np.ndarray]): list of output hand keypoint heatmaps, relative root depth an

Args: rng (None | int | numpy.random.RandomState): seed or state. kwargs (keyword arguments): - num_p

Convert rotations given as axis/angle to rotation matrices. Args: axis_angle: Rotations given as a vector in axis angle form,

Method representing the Node's activity. This method override the standard run() method of Thread. Users should not override this met

Calculate bottom-up masked mse loss. Note: - batch_size: N - num_channels: C - heatmaps height: H

Get track id for each person instance on the current frame. Args: results (list[dict]): The bbox & pose results of the current frame

Visualize the detection results on the image. Args: model (nn.Module): The loaded detector. img (str | np.ndarray): Image filenam

Get pose estimation results of the message. Args: tag (str, optional): If given, only the results with the tag wi

Initialize the weights in backbone. Args: pretrained (str, optional): Path to pre-trained weights. Defaults to No

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

Keypoint evaluation. Report mean per joint position error (MPJPE) and mean per joint position error after rigid alignment (MPJPE-PA)

Initialize model weights.

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

Build posenet.

Multiple with alpha and add beta with clip.

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

Calls either forward_train or forward_test depending on whether return_loss=True. Note this setting will change the expected inputs. W

Calls either forward_train or forward_test depending on whether return_loss=True. Note this setting will change the expected inputs. W

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

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

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

Load an image file, from disk or url. Args: filename (str): file name on the disk or url link. readFlag (int): readFlag for imdec

similar to ImageOps.contain(img_pil, (img_size, img_size)) # keep the same aspect ratio

SMPL model constructor Parameters ---------- model_path: str The path to the folder or to the file where the mode

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

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

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

Evaluate PCKh for MPII dataset. Adapted from https://github.com/leoxiaobin/deep-high-resolution-net.pytorch Copyright (c) Microsoft, u

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

Greedily select boxes with high confidence and overlap <= thr. Args: dets: [[x1, y1, x2, y2, score]]. thr: Retain overlap < thr.