r"""ResNet-50 model from `"Deep Residual Learning for Image Recognition" <https://arxiv.org/pdf/1512.03385.pdf>`_ Args: pretrained (b

Runs a list of trackers on a dataset. args: dataset: List of Sequence instances, forming a dataset. trackers: List of Tracker inst

Runs a tracker on a sequence.

Run tracker on sequence or dataset. args: tracker_name: Name of tracking method. tracker_param: Name of parameter file. ru

Extracts a square crop centered at target_bb box, of area search_area_factor^2 times target_bb area args: im - cv image target_b

This function disables printing when not in master process

Visualize tracking results. Args: img (str | ndarray): Filename of loaded image. result (dict): Tracking result.

deserialize tensor to picklable python object

Generate list of trackers. args: name: Name of tracking method. parameter_name: Name of parameter file. run_ids: A single

Must be deterministic

Transform the box co-ordinates from the original image co-ordinates to the co-ordinates of the cropped image args: box_in - the box for w

Transform the box co-ordinates from the original image co-ordinates to the co-ordinates of the cropped image args: box_in - the box for w

Upsample pred [N, 1, H/8, W/8] -> [N, 1, H, W] using convex combination

ap = voc_ap(rec, prec, [use_07_metric]) Compute VOC AP given precision and recall. If use_07_metric is true, uses the VOC 07 11 point meth

Args: windows: (num_windows*B, window_size, window_size, C) window_size (int): Window size H (int): Height of image

Can be used to sample random parameter values.

args: data - The input data, should contain the following fields: 'template_images', search_images', 'template_an

joint: whether to jointly flip the reference and the current frame

joint: whether to jointly flip the reference and the current frame

joint: whether to jointly flip the reference and the current frame

Arguments: target (annotation) : the target annotation to be made usable will be an ET.Element Returns:

images: (b, F, 3, H, W), targets: (b, F, M, 6), task_ids: (b, 1) M is max number of instances targets: [cls, bbox, trackid]

mask_feats: (1, 8, H//8, W//8), mask_feat_stride=8, mask_head_params: (N, 169), instance_locations: (N, 2), instance_fpn_levels: (N,

Not to be used! Check get_frames() instead.