Apply mapping form the data space to the latent space and calculate the log determinant of the Jacobian matrix.

Applies a batch of inverse kinematics transfoirm to the joints Parameters ---------- pose_skeleton : torch.tensor BxNx3 Loca

Applies a batch of inverse kinematics transfoirm to the joints Parameters ---------- pose_skeleton : torch.tensor BxNx3 Loca

Applies a batch of inverse kinematics transfoirm to the joints Parameters ---------- pose_skeleton : torch.tensor BxNx3 Loca

Applies a batch of inverse kinematics transfoirm to the joints Parameters ---------- pose_skeleton : torch.tensor BxNx3 Locat

Applies a batch of rigid transformations to the joints Parameters ---------- rot_mats : torch.tensor BxNx3x3 Tensor of rotat

Applies a batch of rigid transformations to the joints Parameters ---------- rot_mats : torch.tensor BxNx3x3 Tensor of rotat

Applies a batch of rigid transformations to the joints Parameters ---------- rot_mats : torch.tensor BxNx3x3 Tensor of rotat

Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of (1-beta) over time from t = [

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

Build drop path layer.

Build posenet.

Builder for Transformer.

Project points from camera coordinates to pixel coordinates.

Cluster tokens with DPC-KNN algorithm. Note: B: batch size N: token number C: channel number Args: token_dic

Batched version of compute_similarity_transform.

Batched version of compute_similarity_transform.

Copy gradients from fp16 model to fp32 weight copy.

Copy updated params from fp32 weight copy to fp16 model.

Create layers.

Crop bbox from image by Affinetransform. Parameters ---------- img: torch.Tensor A tensor with shape: `(3, H, W)`. bbox: list

Crop bbox from image by Affinetransform. Parameters ---------- img: torch.Tensor A tensor with shape: `(3, H, W)`. bbox: list

Decode keypoints from output regression. Args: img_metas (list(dict)): Information about data augmentation By def

If a key contains 'keypoints', and f'{key}_mask' is in self.keys(), invalid zeros will be inserted to the right places and f'{key}_mask'

If a key contains 'keypoints', and f'{key}_mask' is in self.keys(), invalid zeros will be inserted to the right places and f'{key}_mask'

Flip human joints horizontally. Note: batch_size: N num_keypoint: K Args: regression (np.ndarray([..., K, C])): Coord

Forward function.

Forward function.

Forward function. Returns: out (list[Tensor]): a list of heatmaps from multiple stages.

Forward function. Returns: out (list[Tensor]): a list of heatmaps from multiple stages and units.

:param img: :param coord_target: [2, 17, 2] :param coord_target_weight: [2, 17, 2] :param hp_target: [2, 4, 17, 64, 4

Defines the computation performed at every call when testing.

Defines the computation performed at every call when training.

heatmaps: [N, j, h, w]

Generates per block width from RegNet parameters. Args: initial_width ([int]): Initial width of the backbone width_sl

Calculate accuracy for top-down keypoint loss. Note: batch_size: N num_keypoints: K Args: output

Calculate top-down keypoint loss. Note: batch_size: N num_keypoints: K Args: output (torch.Tenso

Delta-nDCG w.r.t. pairwise swapping of the currently predicted ltr_adhoc :param batch_ideal_rankings: the standard labels sorted in a descend

Calculate top-down keypoint loss. Note: batch_size: N num_keypoints: K Args: output (torch.Tensor[

Get the layer id to set the different learning rates in ``layer_wise`` decay_type. Args: var_name (str): The key of the model.