Loads a depth map and validity map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file data_format

Loads a depth map and validity map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file data_format

Loads a depth map and validity map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file multiplier

Loads an RGB image Arg(s): path : str path to RGB image normalize : bool if set, then normalize imag

Loads an RGB image Arg(s): path : str path to RGB image normalize : bool if set, then normalize imag

Loads an RGB image Arg(s): path : str path to RGB image normalize : bool if set, then normalize imag

Loads a uncertainty map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file multiplier : float

Loads a validity map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file data_format : str

Loads a validity map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file data_format : str

Load validity map Arg(s): validity_map_path : str path to validity map Returns: numpy[float32] : validity ma

Loads a validity map from a 16-bit PNG file Arg(s): path : str path to 16-bit PNG file data_format : str

Mean absolute error Arg(s): src : numpy[float32] source array tgt : numpy[float32] target array

Mean absolute error Arg(s): src : numpy[float32] source array tgt : numpy[float32] target array

Mean absolute error Args: src : numpy source array tgt : numpy target array Returns: flo

Mean absolute relative error (normalize absolute error) Arg(s): src : numpy[float32] source array tgt : numpy[fl

Mean absolute relative error (normalize absolute error) Args: src : numpy source array tgt : numpy t

Creates N x 2 x H x W meshgrid in x, y directions Arg(s): n_batch : int batch size n_height : int he

Returns the list of parameters in the model Returns: list[torch.Tensor[float32]] : list of parameters

Returns the list of parameters in the model Returns: list[torch.Tensor[float32]] : list of parameters

Returns the list of parameters in the model Returns: list[torch.Tensor[float32]] : list of parameters

Returns the list of parameters in the model Returns: list[torch.Tensor[float32]] : list of parameters

Returns the list of parameters in the model Returns: list[torch.tensor[float32]] : list of parameters

Returns the list of parameters in the model Returns: list[torch.Tensor[float32]] : list of parameters

Returns the list of parameters in the model Returns: list[torch.Tensor[float32]] : list of parameters

Convert 6 DoF parameters to transformation matrix Arg(s): v : torch.Tensor[float32] N x 6 vector in the order of tx, ty,

Convert 6 DoF parameters to transformation matrix Arg(s): v : torch.Tensor[float32] N x 6 vector in the order of tx, ty,

Convert 6 DoF parameters to transformation matrix Arg(s): v : torch.Tensor[float32] N x 6 vector in the order of tx, ty,

Computes the prior depth consistency loss Arg(s): src : torch.Tensor[float32] N x 1 x H x W source depth tgt : t

Process image and depth map paths Arg(s): pathss : tuple list[str] : image paths list[str] : depth maps path

Processes a single frame Arg(s): inputs : tuple[str] image path at time t=0, image path at time t=-1,

Processes a single frame Arg(s): inputs : tuple image path at time t=0, image path at time t=1,

Processes a single frame Arg(s): inputs : tuple image path at time t=0, image path at time t=1,

Processes a single frame Arg(s): inputs : tuple[str] image path at time t=0, image path at time t=1,

Processes a single frame Arg(s): inputs : tuple image path at time t=0, image path at time t=1,

Processes a single frame Arg(s): inputs : tuple image path at time t=0, image path at time t=1,

Processes a single frame Args: inputs : tuple image path, ground truth path Returns: str : output sparse dept

Processes a single depth frame Arg(s): inputs : tuple image path at time t=0, image path at time t=1,

Projects points in camera coordinates to 2D pixel coordinates Arg(s): points : torch.Tensor[float32] N x 4 x (H x W) dep

Projects points in camera coordinates to 2D pixel coordinates Arg(s): points : torch.Tensor[float32] N x 4 x (H x W) dep

Projects points in camera coordinates to 2D pixel coordinates Arg(s): points : torch.Tensor[float32] N x 4 x (H x W) dep

Reads a newline delimited file containing paths Arg(s): filepath : str path to file to be read Return: list[

Reads a newline delimited file containing paths Arg(s): filepath : str path to file to be read Return: list[

Reads a newline delimited file containing paths Arg(s): filepath : str path to file to be read Return: list

Loads weights from checkpoint and loads and returns optimizer Arg(s): restore_path : str path to model w

Loads weights from checkpoint Arg(s): restore_path : str path to model weights optimizer : t

Loads weights from checkpoint and loads and returns optimizer Arg(s): restore_path : str path to model w

Loads weights from checkpoint and loads and returns optimizer Arg(s): restore_path : str path to model w

Loads weights from checkpoint Arg(s): restore_path : str path to model weights optimizer : t

Loads weights from checkpoint Arg(s): restore_path : str path to model weights optimizer : t

Rigid warping of image1 to image0 using pose01 Arg(s): image1 : torch.Tensor[float32] N x C x H x W image0 depth

Root mean squared error Arg(s): src : numpy[float32] source array tgt : numpy[float32] target array

Root mean squared error Arg(s): src : numpy[float32] source array tgt : numpy[float32] target array

Root mean squared error Args: src : numpy source array tgt : numpy target array Returns:

Saves a depth map to a 16-bit PNG file Arg(s): z : numpy[float32] depth map path : str path to store

Saves a depth map to a 16-bit PNG file Arg(s): z : numpy[float32] depth map path : str path to store

Saves a depth map to a 16-bit PNG file Arg(s): z : numpy[float32] depth map path : str path to store

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Saves weights to checkpoint Arg(s): checkpoint_path : str path to save checkpoint to step : i

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Save weights of the model to checkpoint path Arg(s): checkpoint_path : str path to save checkpoint

Saves a uncertainty map to a 16-bit PNG file Arg(s): uncertainty : numpy[float32] H x W uncertainty map path : s

Saves a validity map to a 16-bit PNG file Arg(s): v : numpy[float32] validity map path : str path to

Saves a validity map to a 16-bit PNG file Arg(s): v : numpy[float32] validity map path : str path to

Saves a validity map to a 16-bit PNG file Arg(s): v : numpy[float32] validity map path : str path to

Matches scale of input depth to sparse depth Arg(s): input_depth : torch.Tensor[float32] N x 1 x H x W input dense depth

Computes smooth_l1 loss Arg(s): src : torch.Tensor[float32] N x 3 x H x W source image tgt : torch.Tensor[float3

Computes the local smoothness loss Arg(s): predict : torch.Tensor[float32] N x 1 x H x W predictions image : tor

Computes the local smoothness loss Arg(s): predict : torch.Tensor[float32] N x 1 x H x W predictions image : tor

Computes the local smoothness loss Arg(s): predict : torch.Tensor[float32] N x 1 x H x W predictions image : tor

Computes the sparse depth consistency loss Arg(s): src : torch.Tensor[float32] N x 1 x H x W source depth tgt :

Computes the sparse depth consistency loss Arg(s): src : torch.Tensor[float32] N x 1 x H x W source depth tgt :

Computes the sparse depth consistency loss Arg(s): src : torch.Tensor[float32] N x 1 x H x W source depth tgt :

Given sparse depth, return tensor or integer evaluating teacher predictions against sparse depth Arg(s): sparse_depth : torch.Tensor

Squared relative error Arg(s): src : numpy[float32] source array tgt : numpy[float32] target array