def get_all_boxes_info(node_container):
areas = np.array([((n.box[3] - n.box[1] + 1) *
(n.box[2] - n.box[0] + 1)).cpu().data.numpy()[0]
for n in node_container])
boxes = np.vstack([n.box.cpu().data.numpy() for n in node_container])
bbox_intersection = np.transpose(bbox_intersections_np(
boxes, boxes)) # row-wise, box of each row is the denominator
return areas, bbox_intersection
def bbox_intersections(box_a, box_b):
""" We resize both tensors to [A,B,2] without new malloc:
[A,2] -> [A,1,2] -> [A,B,2]
[B,2] -> [1,B,2] -> [A,B,2]
Then we compute the area of intersect between box_a and box_b.
Args:
box_a: (tensor) bounding boxes, Shape: [A,4].
box_b: (tensor) bounding boxes, Shape: [B,4].
Return:
(tensor) intersection area, Shape: [A,B].
"""
if isinstance(box_a, np.ndarray):
assert isinstance(box_b, np.ndarray)
return bbox_intersections_np(box_a, box_b)
A = box_a.size(0)
B = box_b.size(0)
max_xy = torch.min(box_a[:, 2:].unsqueeze(1).expand(A, B, 2),
box_b[:, 2:].unsqueeze(0).expand(A, B, 2))
min_xy = torch.max(box_a[:, :2].unsqueeze(1).expand(A, B, 2),
box_b[:, :2].unsqueeze(0).expand(A, B, 2))
inter = torch.clamp((max_xy - min_xy + 1.0), min=0)
return inter[:, :, 0] * inter[:, :, 1]