What to do when the ROIs are smaller than $227 \times 227$ in R-CNN?

Question

As English is not my native language, I have some hard time understanding the following sentence:

Regardless of the size or aspect ratio of the candidate region, we warp all pixels in a tight bounding box around it to the required size. Prior to warping, we dilate the tight bounding box so that at the warped size there are exactly p pixels of warped image context around the original box (we use p = 16).

This is from the R-CNN paper. I already extracted the ROI, but now, they say that the input of the CNN should be 227 x 227, but a lot of my ROIs are much smaller. How can I deal with it?

Hello. I am still not sure if your question was theoretical/conceptual or practical (programming issue), but I would like to note that programming issues are generally off-topic here. See https://ai.stackexchange.com/help/on-topic for more details. — nbro, Jun 21 '21 at 23:26

score 2 · Accepted Answer · answered Jun 17 '21 at 14:16

2

It is actually very easy, I didn't know the solution would be that simple. With opencv, you can do :

resized_roi= cv2.resize(roi, (224,224))

and that's it :]

answered Jun 17 '21 at 14:16

Valentin

31
5

What to do when the ROIs are smaller than $227 \times 227$ in R-CNN?

1 Answers1