Create rotation matrix.

Calculate the distance between two points. Args: point1 (torch.Tensor): shape(n, 2). point2 (torch.Tensor): shape(n, 2). Ret

enn Max Pooling.

Estimate the log-likelihood probability that samples belong to the k-th Gaussian. Args: x (torch.Tensor): (T, n, d) or (T

Fits Gaussian mixture model to the data. Args: x (torch.Tensor): input tensor. delta (float, optional): threshold.

Format the results to submission text (standard format for DOTA evaluation). Args: results (list): Testing results of the

Get anchors according to feature map sizes. Args: featmap_sizes (list[tuple]): Multi-level feature map sizes. img_met

Get the expansion of a residual block. The block expansion will be obtained by the following order: 1. If ``expansion`` is given, just retur

Get number of every level anchors inside. Args: num_level_anchors (List[int]): List of number of every level's anchors. inside_fl

Get points according to feature map sizes. Args: featmap_sizes (list[tuple]): Multi-level feature map sizes. img_meta

Get points according to feature map sizes. Args: featmap_sizes (list[tuple]): Multi-level feature map sizes. img_meta

Get points according to feature map sizes. Args: featmap_sizes (list[tuple]): Multi-level feature map sizes. img_meta

Get root logger. Args: log_file (str, optional): File path of log. Defaults to None. log_level (int, optional): The level of logg

Compute corresponding GT box and classification targets for proposals. Args: proposals_list (list[list]): Multi level poi

Compute corresponding GT box and classification targets for proposals. Args: proposals_list (list[list]): Multi level poi

Compute corresponding GT box and classification targets for proposals in initial stage. Args: proposals_list (list[list])

Convert polygons to Gaussian distributions. Args: target (torch.Tensor): Polygons with shape (N, 8). Returns: dict[str, torc

Draw bboxes and class labels (with scores) on an image. Args: img (str | ndarray): The image to be displayed. bboxes (ndarray): B

Initialize weights of the head.

Change from point offset to point coordinate.

Change from point offset to point coordinate.

Change from point offset to point coordinate.

Random select some elements from the gallery. If `gallery` is a Tensor, the returned indices will be a Tensor; If `gallery` is a ndar

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

Train function of ReResNet.

Updates mean to the provided value. Args: mu (torch.Tensor):

Updates variance to the provided value. Args: var (torch.Tensor): (T, k, d) or (T, k, d, d)

Convert oriented bounding box to 2-D Gaussian distribution. Args: xywhr (torch.Tensor): rbboxes with shape (N, 5). Returns:

Compute the aspect ratio of all gts. Args: gt_rbboxes (torch.Tensor): Groundtruth polygons, shape (k, 8). Returns: ratios (t

Compute the aspect ratio of all gts. Args: gt_rbboxes (torch.Tensor): Groundtruth polygons, shape (k, 8). Returns: ratios (t

compute the aspect ratio of all gts. Args: gt_rbboxes (torch.Tensor): Groundtruth polygons, shape (k, 8). Returns:

From the log-probabilities, computes new parameters pi, mu, var (that maximize the log-likelihood). This is the maximization step of t

Add shared or separable branch. convs -> avg pool (optional) -> fcs

Run forward function and calculate loss for box head in training. Args: x (list[Tensor]): list of multi-level img features.

Run forward function and calculate loss for box head in training. Args: x (list[Tensor]): list of multi-level img features.

Run forward function and calculate loss for box head in training. Args: x (list[Tensor]): list of multi-level img features.

Create dummy results.

Create dummy results.

Filter out small bboxes and outside bboxes after Mosaic.

Transform outputs for a single batch item into bbox predictions. Args: cls_scores (list[Tensor]): Box scores for a single scale l

Transform outputs of a single image into bbox predictions. Args: cls_score_list (list[Tensor]): Box scores from all scale

Transform outputs of a single image into bbox predictions. Args: cls_score_list (list[Tensor]): Box scores from all scale

Transform outputs of a single image into bbox predictions. Args: cls_score_list (list[Tensor]): Box scores from all scale

Find the predefined detector config directory.

Load a configuration as a python module.

Initialize layers of the head.

Initialize layers of the head.

Initialize layers of the head.

Initialize layers of the head.

Initialize layers of the head.

Initialize layers of the head.

Initialize layers of feature refine module.

Load DOTA's txt annotation. Args: txtfile (str): Filename of single txt annotation. Returns: dict: Annotation of single imag

Load DOTA's txt annotation. Args: txtfile (str): Filename of single txt annotation. Returns: dict: Annotation of single imag

Build stem layer.

Generate the submission of full images. Args: id_list (list): Id of images. dets_list (list): Detection results of pe

Generate the submission of full images. Args: id_list (list): Id of images. dets_list (list): Detection results of pe

Generate the submission of full images. Args: id_list (list): Id of images. dets_list (list): Detection results of pe

Randomly sample some negative samples.

Randomly sample some positive samples.

Add extra files that are required to support MIM into the package. These files will be added by creating a symlink to the originals if the pa

Add arguments.

Add arguments.

Add plot parser.

Add time parser.

Analyze detection results on each image. Args: confusion_matrix (ndarray): The confusion matrix, has shape (num_classes + 1,

coords should be a N * 2 array-like, containing N couples of (x, y) points

img should be a numpy array, formatted as Height * Width * Nchannels

Assign w.r.t. the overlaps of bboxes with gts. Args: overlaps (torch.Tensor): Overlaps between k gt_bboxes and n bboxes,

NMS for aug multi-class bboxes. Args: multi_bboxes (torch.Tensor): shape (n, #class*5) or (n, 5) multi_scores (torch.Tensor): sha

We usually compute the deltas of x, y, w, h, a of proposals w.r.t ground truth bboxes to get regression target. This is the inverse function of

We usually compute the deltas of x, y, w, h, a of proposals w.r.t ground truth bboxes to get regression target. This is the inverse function of

Compute deltas of proposals w.r.t. gt. We usually compute the deltas of x, y, w, h, a, b of proposals w.r.t ground truth bboxes to get regres

Flip bboxes horizontally or vertically. Args: bboxes (Tensor): Shape (..., 5*k) img_shape (tuple): Image shape. direction

Map bboxes from testing scale to original image scale.

Compute bbox overlaps (iof). Args: bboxes1 (np.array): Horizontal bboxes1. bboxes2 (np.array): Horizontal bboxes2. eps (f

Compute bbox overlaps (iof). Args: bboxes1 (np.array): Horizontal bboxes1. bboxes2 (np.array): Horizontal bboxes2. eps (f

Builder of IoU calculator.

Build shared head.

Calculate the confusion matrix. Args: dataset (Dataset): Test or val dataset. results (list[ndarray]): A list of detection result

Compute overlaps between polygons and points. Args: gt_rbboxes (torch.Tensor): Groundtruth polygons, shape (k, 8). points (torch.

Compute overlaps between polygons and points. Args: gt_rbboxes (torch.Tensor): Groundtruth polygons, shape (k, 8). points (torch.

Args: info (dict): Image's information. windows (np.array): information of sliding windows. window_anns (list[dict]): Li

Args: info (dict): Image's information. windows (np.array): information of sliding windows. window_anns (list[dict]): Li

Apply deltas to shift/scale base boxes. Typically the rois are anchor or proposed bounding boxes and the deltas are network outputs used to sh

Apply deltas to shift/scale base boxes. Typically the rois are anchor or proposed bounding boxes and the deltas are network outputs used to sh

Apply deltas to shift/scale base boxes. Typically the rois are anchor or proposed bounding boxes and the deltas are network outputs used to s

src_path: dota txt format result path. e.g. ../../../worl_dirs/results/orcnn_res50 tar_path: the file path to save the xml result ima