Evaluate depth estimation model on a dataset using standard depth metrics. This function performs comprehensive evaluation of a depth es

Evaluate depth estimation model on DA2K dataset using relative depth accuracy. DA2K (Depth Anything 2K) evaluation focuses on relative d

Visualize feature map by averaging all feature maps into one image and resize to 518*518 Args: feature_map: feature map tensor with s

Creates a dataset with the specified parameters. Args: dataset_str: A dataset string description (e.g. ImageNet:split=TRAIN).

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

Creates a data loader with the specified parameters. Args: dataset: A dataset (third party, LaViDa or WebDataset). batch_siz

Setup logging. Args: output: A file name or a directory to save log files. If None, log files will not be saved. If outp

Recursively find files with specified extensions Args: directory (str): Root directory to search extensions (list): List

Perform depth estimation inference on a single image

Returns: True if distributed training is enabled

Postprocess depth map

Visualize geometric prior matrix and overlay the result on the original image Args: img: Original image tensor [B,C,H,W] geo_

Extract intermediate features from the model Args: x: Input tensor of shape (B, C, H, W) Re

Perform depth estimation inference on a video file Args: model: Trained depth estimation model video_path (str): Path to

Save depth map visualization as a colored image. Args: depth_map (torch.Tensor): Depth map tensor with shape [H, W] or [B, H, W]

Create configs and perform basic setups.

Create corruptions: 'Dark'. Parameters: image: Input image severity_lower: Lower bound of severity (inclusive) severi

Create corruptions: 'Motion Blur'.

Create corruptions: 'Zoom Blur'.

generate 1d decay mask, the result is l*l

generate 1d depth decay mask, the result is l*l

Generate geometric prior Args: HW_tuple: (H,W) represents the height and width of the feature map depth_map: depth map tensor

this will perform the index select, cat the tensors, and provide the attn_bias from cache

this will perform the index select, cat the tensors, and provide the attn_bias from cache

Returns: The rank of the current process within the global process group.

Returns: The number of processes in the process group

Check if the given file is a video file based on its extension Args: filepath (str): Path to the file Returns:

Convert disparity to depth using normalization method. Args: disparity_tensor (torch.Tensor): Predicted disparity values

Preprocess input image

Save image tensor as a BGR image file. Args: img_tensor (torch.Tensor): Image tensor with shape [C, H, W] or [B, C, H, W]