Map rois to corresponding feature levels by scales. - scale < finest_scale: level 0 - finest_scale <= scale < finest_scale * 2: le

Merge augmented bbox scores.

int: Input feature map levels.

int: Input feature map levels.

int: Input feature map levels.

Plot IoU-Recalls curve. Args: recalls(ndarray or list): shape (k,) iou_thrs(ndarray or list): same shape as `recalls`

Plot Proposal_num-Recalls curve. Args: recalls(ndarray or list): shape (k,) proposal_nums(ndarray or list): same shape as `recall

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

Test without augmentation.

Test without augmentation.

Split the combined 1-D polys into masks. A mask is represented as a list of polys, and a poly is represented as a 1-D array. In dataset, all

Check if detected bboxes are true positive or false positive. Args: det_bbox (ndarray): the detected bbox gt_bboxes (ndarray): gr

Check if detected bboxes are true positive or false positive. Args: det_bbox (ndarray): the detected bbox gt_bboxes (ndarray): gr

Unmap a subset of item (data) back to the original set of items (of size count)