Hugging Face's logo

Transformers documentation

ZoeDepth

Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

ZoeDepth

Overview

The ZoeDepth model was proposed in ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth by Shariq Farooq Bhat, Reiner Birkl, Diana Wofk, Peter Wonka, Matthias Müller. ZoeDepth extends the DPT framework for metric (also called absolute) depth estimation. ZoeDepth is pre-trained on 12 datasets using relative depth and fine-tuned on two domains (NYU and KITTI) using metric depth. A lightweight head is used with a novel bin adjustment design called metric bins module for each domain. During inference, each input image is automatically routed to the appropriate head using a latent classifier.

The abstract from the paper is the following:

This paper tackles the problem of depth estimation from a single image. Existing work either focuses on generalization performance disregarding metric scale, i.e. relative depth estimation, or state-of-the-art results on specific datasets, i.e. metric depth estimation. We propose the first approach that combines both worlds, leading to a model with excellent generalization performance while maintaining metric scale. Our flagship model, ZoeD-M12-NK, is pre-trained on 12 datasets using relative depth and fine-tuned on two datasets using metric depth. We use a lightweight head with a novel bin adjustment design called metric bins module for each domain. During inference, each input image is automatically routed to the appropriate head using a latent classifier. Our framework admits multiple configurations depending on the datasets used for relative depth pre-training and metric fine-tuning. Without pre-training, we can already significantly improve the state of the art (SOTA) on the NYU Depth v2 indoor dataset. Pre-training on twelve datasets and fine-tuning on the NYU Depth v2 indoor dataset, we can further improve SOTA for a total of 21% in terms of relative absolute error (REL). Finally, ZoeD-M12-NK is the first model that can jointly train on multiple datasets (NYU Depth v2 and KITTI) without a significant drop in performance and achieve unprecedented zero-shot generalization performance to eight unseen datasets from both indoor and outdoor domains.

drawing ZoeDepth architecture. Taken from the original paper.

This model was contributed by nielsr. The original code can be found here.

Usage tips

  • ZoeDepth is an absolute (also called metric) depth estimation model, unlike DPT which is a relative depth estimation model. This means that ZoeDepth is able to estimate depth in metric units like meters.

The easiest to perform inference with ZoeDepth is by leveraging the pipeline API:

>>> from transformers import pipeline
>>> from PIL import Image
>>> import requests

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> pipe = pipeline(task="depth-estimation", model="Intel/zoedepth-nyu-kitti")
>>> result = pipe(image)
>>> depth = result["depth"]

Alternatively, one can also perform inference using the classes:

>>> from transformers import AutoImageProcessor, ZoeDepthForDepthEstimation
>>> import torch
>>> import numpy as np
>>> from PIL import Image
>>> import requests

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> image_processor = AutoImageProcessor.from_pretrained("Intel/zoedepth-nyu-kitti")
>>> model = ZoeDepthForDepthEstimation.from_pretrained("Intel/zoedepth-nyu-kitti")

>>> # prepare image for the model
>>> inputs = image_processor(images=image, return_tensors="pt")

>>> with torch.no_grad():   
...     outputs = model(pixel_values)

>>> # interpolate to original size and visualize the prediction
>>> ## ZoeDepth dynamically pads the input image. Thus we pass the original image size as argument
>>> ## to `post_process_depth_estimation` to remove the padding and resize to original dimensions.
>>> post_processed_output = image_processor.post_process_depth_estimation(
...     outputs,
...     source_sizes=[(image.height, image.width)],
... )

>>> predicted_depth = post_processed_output[0]["predicted_depth"]
>>> depth = (predicted_depth - predicted_depth.min()) / (predicted_depth.max() - predicted_depth.min())
>>> depth = depth.detach().cpu().numpy() * 255
>>> depth = Image.fromarray(depth.astype("uint8"))

In the original implementation ZoeDepth model performs inference on both the original and flipped images and averages out the results. The post_process_depth_estimation function can handle this for us by passing the flipped outputs to the optional outputs_flipped argument:

>>> with torch.no_grad():   
...     outputs = model(pixel_values)
...     outputs_flipped = model(pixel_values=torch.flip(inputs.pixel_values, dims=[3]))
>>> post_processed_output = image_processor.post_process_depth_estimation(
...     outputs,
...     source_sizes=[(image.height, image.width)],
...     outputs_flipped=outputs_flipped,
... )

Resources

A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with ZoeDepth.

  • A demo notebook regarding inference with ZoeDepth models can be found here. 🌎

ZoeDepthConfig

class transformers.ZoeDepthConfig

< >

( backbone_config = None backbone = None use_pretrained_backbone = False backbone_kwargs = None hidden_act = 'gelu' initializer_range = 0.02 batch_norm_eps = 1e-05 readout_type = 'project' reassemble_factors = [4, 2, 1, 0.5] neck_hidden_sizes = [96, 192, 384, 768] fusion_hidden_size = 256 head_in_index = -1 use_batch_norm_in_fusion_residual = False use_bias_in_fusion_residual = None num_relative_features = 32 add_projection = False bottleneck_features = 256 num_attractors = [16, 8, 4, 1] bin_embedding_dim = 128 attractor_alpha = 1000 attractor_gamma = 2 attractor_kind = 'mean' min_temp = 0.0212 max_temp = 50.0 bin_centers_type = 'softplus' bin_configurations = [{'n_bins': 64, 'min_depth': 0.001, 'max_depth': 10.0}] num_patch_transformer_layers = None patch_transformer_hidden_size = None patch_transformer_intermediate_size = None patch_transformer_num_attention_heads = None **kwargs )

Parameters

  • backbone_config (Union[Dict[str, Any], PretrainedConfig], optional, defaults to BeitConfig()) — The configuration of the backbone model.
  • backbone (str, optional) — Name of backbone to use when backbone_config is None. If use_pretrained_backbone is True, this will load the corresponding pretrained weights from the timm or transformers library. If use_pretrained_backbone is False, this loads the backbone’s config and uses that to initialize the backbone with random weights.
  • use_pretrained_backbone (bool, optional, defaults to False) — Whether to use pretrained weights for the backbone.
  • backbone_kwargs (dict, optional) — Keyword arguments to be passed to AutoBackbone when loading from a checkpoint e.g. {'out_indices': (0, 1, 2, 3)}. Cannot be specified if backbone_config is set.
  • hidden_act (str or function, optional, defaults to "gelu") — The non-linear activation function (function or string) in the encoder and pooler. If string, "gelu", "relu", "selu" and "gelu_new" are supported.
  • initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • batch_norm_eps (float, optional, defaults to 1e-05) — The epsilon used by the batch normalization layers.
  • readout_type (str, optional, defaults to "project") — The readout type to use when processing the readout token (CLS token) of the intermediate hidden states of the ViT backbone. Can be one of ["ignore", "add", "project"].

    • “ignore” simply ignores the CLS token.
    • “add” passes the information from the CLS token to all other tokens by adding the representations.
    • “project” passes information to the other tokens by concatenating the readout to all other tokens before projecting the representation to the original feature dimension D using a linear layer followed by a GELU non-linearity.
  • reassemble_factors (List[int], optional, defaults to [4, 2, 1, 0.5]) — The up/downsampling factors of the reassemble layers.
  • neck_hidden_sizes (List[str], optional, defaults to [96, 192, 384, 768]) — The hidden sizes to project to for the feature maps of the backbone.
  • fusion_hidden_size (int, optional, defaults to 256) — The number of channels before fusion.
  • head_in_index (int, optional, defaults to -1) — The index of the features to use in the heads.
  • use_batch_norm_in_fusion_residual (bool, optional, defaults to False) — Whether to use batch normalization in the pre-activate residual units of the fusion blocks.
  • use_bias_in_fusion_residual (bool, optional, defaults to True) — Whether to use bias in the pre-activate residual units of the fusion blocks.
  • num_relative_features (int, optional, defaults to 32) — The number of features to use in the relative depth estimation head.
  • add_projection (bool, optional, defaults to False) — Whether to add a projection layer before the depth estimation head.
  • bottleneck_features (int, optional, defaults to 256) — The number of features in the bottleneck layer.
  • num_attractors (List[int], *optional*, defaults to [16, 8, 4, 1]`) — The number of attractors to use in each stage.
  • bin_embedding_dim (int, optional, defaults to 128) — The dimension of the bin embeddings.
  • attractor_alpha (int, optional, defaults to 1000) — The alpha value to use in the attractor.
  • attractor_gamma (int, optional, defaults to 2) — The gamma value to use in the attractor.
  • attractor_kind (str, optional, defaults to "mean") — The kind of attractor to use. Can be one of ["mean", "sum"].
  • min_temp (float, optional, defaults to 0.0212) — The minimum temperature value to consider.
  • max_temp (float, optional, defaults to 50.0) — The maximum temperature value to consider.
  • bin_centers_type (str, optional, defaults to "softplus") — Activation type used for bin centers. Can be “normed” or “softplus”. For “normed” bin centers, linear normalization trick is applied. This results in bounded bin centers. For “softplus”, softplus activation is used and thus are unbounded.
  • bin_configurations (List[dict], optional, defaults to [{'n_bins' -- 64, 'min_depth': 0.001, 'max_depth': 10.0}]): Configuration for each of the bin heads. Each configuration should consist of the following keys:

    • name (str): The name of the bin head - only required in case of multiple bin configurations.
    • n_bins (int): The number of bins to use.
    • min_depth (float): The minimum depth value to consider.
    • max_depth (float): The maximum depth value to consider. In case only a single configuration is passed, the model will use a single head with the specified configuration. In case multiple configurations are passed, the model will use multiple heads with the specified configurations.
  • num_patch_transformer_layers (int, optional) — The number of transformer layers to use in the patch transformer. Only used in case of multiple bin configurations.
  • patch_transformer_hidden_size (int, optional) — The hidden size to use in the patch transformer. Only used in case of multiple bin configurations.
  • patch_transformer_intermediate_size (int, optional) — The intermediate size to use in the patch transformer. Only used in case of multiple bin configurations.
  • patch_transformer_num_attention_heads (int, optional) — The number of attention heads to use in the patch transformer. Only used in case of multiple bin configurations.

This is the configuration class to store the configuration of a ZoeDepthForDepthEstimation. It is used to instantiate an ZoeDepth model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the ZoeDepth Intel/zoedepth-nyu architecture.

Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.

Example:

>>> from transformers import ZoeDepthConfig, ZoeDepthForDepthEstimation

>>> # Initializing a ZoeDepth zoedepth-large style configuration
>>> configuration = ZoeDepthConfig()

>>> # Initializing a model from the zoedepth-large style configuration
>>> model = ZoeDepthForDepthEstimation(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

ZoeDepthImageProcessor

class transformers.ZoeDepthImageProcessor

< >

( do_pad: bool = True do_rescale: bool = True rescale_factor: typing.Union[int, float] = 0.00392156862745098 do_normalize: bool = True image_mean: typing.Union[float, typing.List[float], NoneType] = None image_std: typing.Union[float, typing.List[float], NoneType] = None do_resize: bool = True size: typing.Dict[str, int] = None resample: Resampling = <Resampling.BILINEAR: 2> keep_aspect_ratio: bool = True ensure_multiple_of: int = 32 **kwargs )

Parameters

  • do_pad (bool, optional, defaults to True) — Whether to apply pad the input.
  • do_rescale (bool, optional, defaults to True) — Whether to rescale the image by the specified scale rescale_factor. Can be overidden by do_rescale in preprocess.
  • rescale_factor (int or float, optional, defaults to 1/255) — Scale factor to use if rescaling the image. Can be overidden by rescale_factor in preprocess.
  • do_normalize (bool, optional, defaults to True) — Whether to normalize the image. Can be overridden by the do_normalize parameter in the preprocess method.
  • image_mean (float or List[float], optional, defaults to IMAGENET_STANDARD_MEAN) — Mean to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the image_mean parameter in the preprocess method.
  • image_std (float or List[float], optional, defaults to IMAGENET_STANDARD_STD) — Standard deviation to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the image_std parameter in the preprocess method.
  • do_resize (bool, optional, defaults to True) — Whether to resize the image’s (height, width) dimensions. Can be overidden by do_resize in preprocess.
  • size (Dict[str, int] optional, defaults to {"height" -- 384, "width": 512}): Size of the image after resizing. Size of the image after resizing. If keep_aspect_ratio is True, the image is resized by choosing the smaller of the height and width scaling factors and using it for both dimensions. If ensure_multiple_of is also set, the image is further resized to a size that is a multiple of this value. Can be overidden by size in preprocess.
  • resample (PILImageResampling, optional, defaults to Resampling.BILINEAR) — Defines the resampling filter to use if resizing the image. Can be overidden by resample in preprocess.
  • keep_aspect_ratio (bool, optional, defaults to True) — If True, the image is resized by choosing the smaller of the height and width scaling factors and using it for both dimensions. This ensures that the image is scaled down as little as possible while still fitting within the desired output size. In case ensure_multiple_of is also set, the image is further resized to a size that is a multiple of this value by flooring the height and width to the nearest multiple of this value. Can be overidden by keep_aspect_ratio in preprocess.
  • ensure_multiple_of (int, optional, defaults to 32) — If do_resize is True, the image is resized to a size that is a multiple of this value. Works by flooring the height and width to the nearest multiple of this value.

    Works both with and without keep_aspect_ratio being set to True. Can be overidden by ensure_multiple_of in preprocess.

Constructs a ZoeDepth image processor.

preprocess

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), typing.List[ForwardRef('PIL.Image.Image')], typing.List[numpy.ndarray], typing.List[ForwardRef('torch.Tensor')]] do_pad: bool = None do_rescale: bool = None rescale_factor: float = None do_normalize: bool = None image_mean: typing.Union[float, typing.List[float], NoneType] = None image_std: typing.Union[float, typing.List[float], NoneType] = None do_resize: bool = None size: int = None keep_aspect_ratio: bool = None ensure_multiple_of: int = None resample: Resampling = None return_tensors: typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None data_format: ChannelDimension = <ChannelDimension.FIRST: 'channels_first'> input_data_format: typing.Union[str, transformers.image_utils.ChannelDimension, NoneType] = None )

Parameters

  • images (ImageInput) — Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set do_rescale=False.
  • do_pad (bool, optional, defaults to self.do_pad) — Whether to pad the input image.
  • do_rescale (bool, optional, defaults to self.do_rescale) — Whether to rescale the image values between [0 - 1].
  • rescale_factor (float, optional, defaults to self.rescale_factor) — Rescale factor to rescale the image by if do_rescale is set to True.
  • do_normalize (bool, optional, defaults to self.do_normalize) — Whether to normalize the image.
  • image_mean (float or List[float], optional, defaults to self.image_mean) — Image mean.
  • image_std (float or List[float], optional, defaults to self.image_std) — Image standard deviation.
  • do_resize (bool, optional, defaults to self.do_resize) — Whether to resize the image.
  • size (Dict[str, int], optional, defaults to self.size) — Size of the image after resizing. If keep_aspect_ratio is True, he image is resized by choosing the smaller of the height and width scaling factors and using it for both dimensions. If ensure_multiple_of is also set, the image is further resized to a size that is a multiple of this value.
  • keep_aspect_ratio (bool, optional, defaults to self.keep_aspect_ratio) — If True and do_resize=True, the image is resized by choosing the smaller of the height and width scaling factors and using it for both dimensions. This ensures that the image is scaled down as little as possible while still fitting within the desired output size. In case ensure_multiple_of is also set, the image is further resized to a size that is a multiple of this value by flooring the height and width to the nearest multiple of this value.
  • ensure_multiple_of (int, optional, defaults to self.ensure_multiple_of) — If do_resize is True, the image is resized to a size that is a multiple of this value. Works by flooring the height and width to the nearest multiple of this value.

    Works both with and without keep_aspect_ratio being set to True. Can be overidden by ensure_multiple_of in preprocess.

  • resample (int, optional, defaults to self.resample) — Resampling filter to use if resizing the image. This can be one of the enum PILImageResampling, Only has an effect if do_resize is set to True.
  • return_tensors (str or TensorType, optional) — The type of tensors to return. Can be one of:

    • Unset: Return a list of np.ndarray.
    • TensorType.TENSORFLOW or 'tf': Return a batch of type tf.Tensor.
    • TensorType.PYTORCH or 'pt': Return a batch of type torch.Tensor.
    • TensorType.NUMPY or 'np': Return a batch of type np.ndarray.
    • TensorType.JAX or 'jax': Return a batch of type jax.numpy.ndarray.
  • data_format (ChannelDimension or str, optional, defaults to ChannelDimension.FIRST) — The channel dimension format for the output image. Can be one of:

    • ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • ChannelDimension.LAST: image in (height, width, num_channels) format.
  • input_data_format (ChannelDimension or str, optional) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:

    • "channels_first" or ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: image in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: image in (height, width) format.

Preprocess an image or batch of images.

ZoeDepthForDepthEstimation

class transformers.ZoeDepthForDepthEstimation

< >

( config )

Parameters

  • config (ViTConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

ZoeDepth model with one or multiple metric depth estimation head(s) on top.

This model is a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( pixel_values: FloatTensor labels: typing.Optional[torch.LongTensor] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) → transformers.modeling_outputs.DepthEstimatorOutput or tuple(torch.FloatTensor)

Parameters

  • pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Pixel values can be obtained using AutoImageProcessor. See DPTImageProcessor.call() for details.
  • output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
  • labels (torch.LongTensor of shape (batch_size, height, width), optional) — Ground truth depth estimation maps for computing the loss.

Returns

transformers.modeling_outputs.DepthEstimatorOutput or tuple(torch.FloatTensor)

A transformers.modeling_outputs.DepthEstimatorOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (ZoeDepthConfig) and inputs.

  • loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Classification (or regression if config.num_labels==1) loss.

  • predicted_depth (torch.FloatTensor of shape (batch_size, height, width)) — Predicted depth for each pixel.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, num_channels, height, width).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, patch_size, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The ZoeDepthForDepthEstimation forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Examples:

>>> from transformers import AutoImageProcessor, ZoeDepthForDepthEstimation
>>> import torch
>>> import numpy as np
>>> from PIL import Image
>>> import requests

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> image_processor = AutoImageProcessor.from_pretrained("Intel/zoedepth-nyu-kitti")
>>> model = ZoeDepthForDepthEstimation.from_pretrained("Intel/zoedepth-nyu-kitti")

>>> # prepare image for the model
>>> inputs = image_processor(images=image, return_tensors="pt")

>>> with torch.no_grad():
...     outputs = model(**inputs)

>>> # interpolate to original size
>>> post_processed_output = image_processor.post_process_depth_estimation(
...     outputs,
...     source_sizes=[(image.height, image.width)],
... )

>>> # visualize the prediction
>>> predicted_depth = post_processed_output[0]["predicted_depth"]
>>> depth = predicted_depth * 255 / predicted_depth.max()
>>> depth = depth.detach().cpu().numpy()
>>> depth = Image.fromarray(depth.astype("uint8"))
< > Update on GitHub

←YOLOS Audio Spectrogram Transformer→