A LayerNorm variant, popularized by Transformers, that performs point-wise mean and variance normalization over the channel dimension for inp

Computes and stores the average and current value

A container for any type of objects. Typically tensors will be stacked in the collate function and sliced along some dimension in the scatter

A transform that does nothing.

A wrapper around :class:`torch.nn.Conv2d` to support empty inputs and more features.

Loading image from file. Args: color_type (str): Flags specifying the color type of a loaded image, candidates are 'color', 'gr

Affine transform the image to make input. Required keys:'img', 'joints_3d', 'joints_3d_visible', 'ann_info','scale', 'rotation' and 'center'.

Collect data from the loader relevant to the specific task. This keeps the items in `keys` as it is, and collect items in `meta_keys` into a

Transform image to Tensor. Required key: 'img'. Modifies key: 'img'. Args: results (dict): contain all information about training.

This class represents a list of instances in an image. It stores the attributes of instances (e.g., boxes, masks, labels, scores) as "fields"

A simple structure that contains basic shape specification about a tensor. It is often used as the auxiliary inputs/outputs of models, to

Maintain a list of transform operations which will be applied in sequence. Attributes: transforms (list[Transform])

This class computes the loss for DETR. The process happens in two steps: 1) we compute hungarian assignment between ground truth boxes and

Generate the target heatmap. Required keys: 'joints_3d', 'joints_3d_visible', 'ann_info'. Modified keys: 'target', and 'target_weight'. Ar

Data augmentation with random scaling & rotating. Required key: 'scale'. Modifies key: 'scale' and 'rotation'. Args: rot_factor (int

Data augmentation with random image flip. Required keys: 'img', 'joints_3d', 'joints_3d_visible', 'center' and 'ann_info'. Modifies key:

This class stores the segmentation masks for all objects in one image, in the form of bitmaps. Attributes: tensor: bool Tensor o

This structure stores a list of boxes as a Nx4 torch.Tensor. It supports some common methods about boxes (`area`, `clip`, `nonempty`, etc

Evaluate semantic segmentation metrics.

Aggregate the context features according to the initial predicted probability distribution. Employ the soft-weighted method to aggregate the

Data augmentation with half-body transform. Keep only the upper body or the lower body at random. Required keys: 'joints_3d', 'joints_3d_visi

:class: bounding box for detection groundtruth, inherited from BoxBase :ivar int ign: if the bounding box should be ignored or not. 0 for NOT

Resize the image to a target size.

Resize the image to a target size.

Implementation of the OCR module: We aggregate the global object representation to update the representation for each pixel. use_gt=True

:class: bounding box for detection result, inherited from BoxBase :ivar float score: detection score (for one class) of bounding box :iva

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

Perform horizontal flip.

Evaluate human parsing metrics, specifically, for Human3.6M

r""" LayerNorm that supports two data formats: channels_last (default) or channels_first. The ordering of the dimensions in the inputs. channels_

Very simple multi-layer perceptron (also called FFN)

Evaluate Pedestrain Detection metrics

Evaluate semantic segmentation metrics.

This method returns a copy of this image, rotated the given number of degrees counter clockwise around its center.

A SemanticSegmentor with test-time augmentation enabled. Its :meth:`__call__` method has the same interface as :meth:`SemanticSegmentor.forwa

Perform vertical flip.

Vision Transformer with support for patch or hybrid CNN input stage

Vision Transformer with support for patch or hybrid CNN input stage

Vision Transformer with support for patch or hybrid CNN input stage

Input that can be used with :meth:`Augmentation.__call__`. This is a standard implementation for the majority of use cases. This class pr

Input that can be used with :meth:`Augmentation.__call__`. This is a standard implementation for the majority of use cases. This class pr

Apply a sequence of augmentations. It has ``__call__`` method to apply the augmentations. Note that :meth:`get_transform` method is imp

Apply a sequence of augmentations. It has ``__call__`` method to apply the augmentations. Note that :meth:`get_transform` method is impos

A wrapper around :class:`torch.nn.Conv2d` to support empty inputs and more features.

Tensorflow like 'SAME' convolution wrapper for 2D convolutions

This class computes the loss for DETR. The process happens in two steps: 1) we compute hungarian assignment between ground truth boxes an

This class computes an assignment between the targets and the predictions of the network For efficiency reasons, the targets don't include the no

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

Extracts a subregion from the source image and scales it to the output size. The fill color is used to map pixels from the source rect that f