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

Parse a compact DSL like ``w-3,iw-1,none-5,ijd-5``. Each segment is ``<keys>-<duration>`` where the last ``-`` separates keys from the i

Preprocesses the input image before performing inference. Returns: image_data: Preprocessed image data ready for inferen

x: [B, L, N, C]. grid_sizes: [B, 3]. freqs: [M, C // 2].

calculate the number of frames for video used for model inputs. Args: ele (dict): a dict contains the configuration of video.

Transform pixel location to different reference.

Convert numpy arrays (representing WASD and IJKL key states, ranged 0-1) into a list of lists of pressed keys for each frame. Args:

Configures a model object. This includes setting evaluation modes, applying distributed parallel strategy, and handling device placem

Convert flow matching's prediction to x0 prediction. flow_pred: the prediction with shape [B, C, F, H, W] xt: the input noisy

Return the percentage below the distance threshold, while ignoring distances values with -1. Note: batch_size: N Args: di

Get keypoint predictions from 3D score maps. Note: batch size: N num keypoints: K heatmap depth size: D heatmap h

Initialize the step_index counter for the scheduler.

Initialize the step_index counter for the scheduler.

Apply the RMSNorm normalization to the input tensor. Args: x (torch.Tensor): The input tensor. Returns:

r""" Prepares and returns the required model for the current timestep. Args: t (torch.Tensor): current ti

Distribution aware coordinate decoding method. Note: - heatmap height: H - heatmap width: W Args: heatmap (np.ndarra

Autoencoder3d adapted from Stable Diffusion 1.x, 2.x and XL.

Convert ``action_string`` to the same layout as ``wasd_action.npy`` / ``ijkl_action.npy``. Returns: tuple: (wasd_action, ijkl_action

Perform QKV self attention. Args: q (torch.Tensor): Query tensor with shape [b, s, a, d], where a is the number of heads. k

Convert the model output to the corresponding type the DPMSolver/DPMSolver++ algorithm needs. DPM-Solver is designed to discretize an

r""" Convert the model output to the corresponding type the UniPC algorithm needs. Args: model_output (`torch.Tensor`):

One step for the first-order DPMSolver (equivalent to DDIM). Args: model_output (`torch.Tensor`): The dir

Draw keypoints and connections representing hand pose on a given canvas. Args: canvas (np.ndarray): A 3D numpy array representing th

Draw keypoints and connections representing hand pose on a given canvas. Args: canvas (np.ndarray): A 3D numpy array representing th

Draw a chevron (V-shape) arrow.

Draw IJKL rotation control UI (similar to WASD layout). Layout: I J K L I - Up, J - Left, K - Down, L - Right

Draw keypoints and connections representing hand pose on a given canvas. Args: canvas (np.ndarray): A 3D numpy array representing th

Draw the rotation UI with two modes: - 'arrow': Chevron arrows on screen edges (only when rotating) - 'keys': IJKL keys layout in bottom-

Draw the WASD UI on the frame with active keys highlighted.

r""" Args: x(Tensor): Shape [B, L1, C] e(Tensor): Shape [B, L1, C]

Performs inference using an ONNX model and returns the output image with drawn detections. Returns: output_img: The outp

r""" Generates video frames from input image and text prompt using diffusion process. Args: input_prompt (`str`):

Args: mask: [h, w] Returns: