Transpose bounding box (flip or rotate in 90 degree steps) :param method: One of :py:attr:`PIL.Image.FLIP_LEFT_RIGHT`, :py:

Accesses dataset associated with the specified field in the box collection. Args: field: a string parameter used to speficy a related field

Convenience function for accesssing box coordinates. Returns: a numpy array of shape [N, 4] representing box corners

Add dataset to a specified field. Args: field: a string parameter used to speficy a related field to be accessed. field_data: a numpy

Return number of boxes held in collections.

Return a extended bounding box copy of this bounding box. All other fields should be keep unchanged. :param scale: By what ex

Convenience function for accessing masks. Returns: a numpy array of shape [N, height, width] representing masks

include a separator `/` at the end of `prefix`

Returns a resized copy of this bounding box :param size: The requested size in pixels, as a 2-tuple: (width, height).

Efficient version of torch.cat that avoids a copy if there is only a single element in a list

dump cfg to `json` file. Args: cfg (dict): The dict to dump. savepath (str): The filepath to save the dumped dict.

Gather boxes from BoxList according to indices and return new BoxList. By default, gather returns boxes corresponding to the input index list, as

Constructor. Args: categories: A list of dicts, each of which has the following keys - 'id': (required) an integer id uniquely iden

Generate bbox and apply lambda correction.

Load video content using Decord

Load video content using Decord

Load video content using Decord

Load video content using Decord

Perform a spatial short scale jittering on the given images and corresponding boxes. Args: images (tensor): images to perform sca

With 50% prob, negate the value

Return all non-box fields.

Given scale, ratio, height and width, return sampled coordinates of the videos.

Adds groundtruth for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. grou

Converts a BoxList containing 'masks' into a BoxMaskList. Args: boxlist: An np_box_list.BoxList object. Returns: An np_box_mask_list.Box

Get video randaug transform. Args: input_size: The size of the input video in tuple. auto_augment: Parameters for randaug. A

Cropss a rectangular region from this bounding box. The box is a 4-tuple defining the left, upper, right, and lower pixel coo

encode image / videos as features. Args: image (torch.Tensor): The input images. test (bool): Whether testing.

Gather boxes from np_box_mask_list.BoxMaskList according to indices. By default, gather returns boxes corresponding to the input index list, as w

include a separator (e.g., `/`, or "_") at the end of `prefix`

Sinusoid position encoding table

get text encoder, used for text and cross-modal encoding

[summary] Args: ann_file_list (List[List[str, str]] or List[str, str]): the latter will be automatically converted to the for

:param things: [n2, c, t, h, w] :param proposals: :param max_things: :return [n, max_other, c, t, h, w]

Setup config, logger, output_dir, etc. Shared for pretrain and all downstream tasks.