rpn::rpn Namespace Reference

Functions
def	assign_anchor
def	get_rpn_batch
def	get_rpn_testbatch

---

Detailed Description

RPN:
data =
{'data': [num_images, c, h, w],
 'im_info': [num_images, 4] (optional)}
label =
{'gt_boxes': [num_boxes, 5] (optional),
 'label': [batch_size, 1] <- [batch_size, num_anchors, feat_height, feat_width],
 'bbox_target': [batch_size, num_anchors, feat_height, feat_width],
 'bbox_weight': [batch_size, num_anchors, feat_height, feat_width]}

---

Function Documentation

def rpn.rpn.assign_anchor	(		feat_shape,
			gt_boxes,
			im_info,
			cfg,
			feat_stride = 16,
			scales = (8, 16,
			ratios = (0.5, 1,
			allowed_border = 0
	)

assign ground truth boxes to anchor positions
:param feat_shape: infer output shape
:param gt_boxes: assign ground truth
:param im_info: filter out anchors overlapped with edges
:param feat_stride: anchor position step
:param scales: used to generate anchors, affects num_anchors (per location)
:param ratios: aspect ratios of generated anchors
:param allowed_border: filter out anchors with edge overlap > allowed_border
:return: dict of label
'label': of shape (batch_size, 1) <- (batch_size, num_anchors, feat_height, feat_width)
'bbox_target': of shape (batch_size, num_anchors * 4, feat_height, feat_width)
'bbox_inside_weight': *todo* mark the assigned anchors
'bbox_outside_weight': used to normalize the bbox_loss, all weights sums to RPN_POSITIVE_WEIGHT

Definition at line 66 of file rpn.py.

def rpn.rpn.get_rpn_batch	(		roidb,
			cfg
	)

prototype for rpn batch: data, im_info, gt_boxes
:param roidb: ['image', 'flipped'] + ['gt_boxes', 'boxes', 'gt_classes']
:return: data, label

Definition at line 39 of file rpn.py.

def rpn.rpn.get_rpn_testbatch	(		roidb,
			cfg
	)

return a dict of testbatch
:param roidb: ['image', 'flipped']
:return: data, label, im_info

Definition at line 21 of file rpn.py.