Parse arguments.

Parse arguments.

Parse arguments.

Parse information from a line in a requirements text file.

Parse version information.

Draw confusion matrix with matplotlib. Args: confusion_matrix (ndarray): The confusion matrix. labels (list[str]): List of class

Convert points to oriented bboxes.

Convert points to oriented bboxes.

Compute center point of Pointsets. Args: RPoints (torch.Tensor): the lists of Pointsets, shape (k, 18). y_first (bool, optional)

Convert polygons to horizontal bboxes. Args: polys (np.array): Polygons with shape (N, 8) Returns: np.array: Horizontal bbox

Convert polygons to horizontal bboxes. Args: polys (np.array): Polygons with shape (N, 8) Returns: np.array: Horizontal bbox

Convert polygons to oriented bounding boxes. Args: polys (torch.Tensor): [x0,y0,x1,y1,x2,y2,x3,y3] Returns: obbs (torch.Tens

Convert polygons to oriented bounding boxes. Args: polys (torch.Tensor): [x0,y0,x1,y1,x2,y2,x3,y3] Returns: obbs (torch.Tens

Convert polygons to oriented bounding boxes. Args: polys (ndarray): [x0,y0,x1,y1,x2,y2,x3,y3] Returns: obbs (ndarray): [x_ct

Convert polygons to oriented bounding boxes. Args: polys (ndarray): [x0,y0,x1,y1,x2,y2,x3,y3] Returns: obbs (ndarray): [x_ct

Convert polygons to oriented bounding boxes. Args: polys (ndarray): [x0,y0,x1,y1,x2,y2,x3,y3] Returns: obbs (ndarray): [x_ct

Convert polygons to oriented bounding boxes. Args: polys (torch.Tensor): [x0,y0,x1,y1,x2,y2,x3,y3] Returns: obbs (torch.Tens

Print mAP and results of each class. A table will be printed to show the gts/dets/recall/AP of each class and the mAP. Args: mea

Only inference related parameters are retained. Args: in_file (str): Filename of input checkpoint. out_file (str): Filename of ou

Load README.md.

Refine bboxes during training. Args: rois (torch.Tensor): Shape (n*bs, 5), where n is image number per GPU, and b

Regress the bbox for the predicted class. Used in Cascade R-CNN. Args: rois (torch.Tensor): shape (n, 4) or (n, 5) la

Repeat to inference several times and take the average. Args: cfg (object): Test config object. checkpoint (str): Checkpoint file

Reset the parameters of ORConv2d.

Retrieve the dataset config file. Args: config_path (str): Path of the config file. skip_type (list[str]): List of the useless pi

Scale RoI coordinates by scale factor. Args: rois (torch.Tensor): RoI (Region of Interest), shape (n, 6) scale_factor

Build active rotating filter module.

Rotated IoU loss. Computing the IoU loss between a set of predicted rbboxes and target rbboxes. The loss is calculated as negative log o

Setup logger. Args: log_path (str): Path of log. Returns: object: Logger.

Setup logger. Args: log_path (str): Path of log. Returns: object: Logger.

Test only det bboxes without augmentation. Args: x (tuple[Tensor]): Feature maps of all scale level. img_metas (list[

Slide windows in images and get window position. Args: width (int): The width of the image. height (int): The height of the image

Smooth Focal Loss proposed in Circular Smooth Label (CSL). Args: pred (torch.Tensor): The prediction. target (torch.Tensor): The

The loss is used to penalize the learning points out of the assigned ground truth boxes (polygon by default). Args: pts (torch.Tensor

Map bboxes from window coordinate back to original coordinate. Args: bboxes (np.array): bboxes with window coordinate. x (float):

Map bboxes from window coordinate back to original coordinate. Args: bboxes (np.array): bboxes with window coordinate. x (float):

Translate bboxes according to its shape. If the bbox shape is (n, 5), the bboxes are regarded as horizontal bboxes and in (x, y, x, y, score)

Updates pi to the provided value. Args: pi (torch.Tensor): (T, k, 1)

Weghted spatial border loss. Args: pts (torch.Tensor): point sets with shape (N, 9*2). gt_bboxes (torch.Tensor): gt_bboxes with p

Calculate IoU between 2D bboxes. Args: bboxes1 (torch.Tensor): bboxes have shape (m, 5) in <cx, cy, w, h, a> form

Call function of PolyRandomRotate.

Call functions to add image meta information. Args: results (dict): Result dict with image in ``results['img']``. Return

Bottleneck block for ResNet. If style is "pytorch", the stride-two layer is the 3x3 conv layer, if it is "caffe", the stride-two laye