Crop Multiple Images with Indices

vision
1

Hi all,
I want to crop images with given indices like follows:

images = given Tensor with shape N x C x H x W
lt_x = given Tensor with shape N
lt_y = given Tensor with shape N
rb_x = given Tensor with shape N
rb_y = given Tensor with shape N
for i in range(images.size(0)):
     croped_images.append(images[i, :, lt_x[i]: rb_x[i], lt_y[i]: rb_y[i]])

But the for loop is too slow. Thus I want to know if there are some batch operation that can do the similar operation, like

croped_images = images[:, :, lt_x: rb_x, lt_y: rb_y]

Thanks.

2

Hi,
I don’t know of a way to do this, but here’s a link to someone trying to do exactly the same thing.
Good luck

1 Like