Using RCNN to autocrop an image

Asked Mar 14 '16 at 20:52

Active Mar 16 '16 at 15:47

Viewed 339 times

I am new to machine learning. I've been messing around with NVIDIA Digits to train a new dataset. My dataset however is too inaccurate and I think it is because there is too much background in the image that it is getting confused as to what the actual object is. My question:

Is there a way (possibly using RCNN) to crop out the background and then proceed to train using the cropped image? The object is consistent (ex only one object like a singular person but there may be people in the background) and always by itself.

edited Mar 16 '16 at 15:47

asked Mar 14 '16 at 20:52

Kroden

How big is your dataset? How many categories and how many images do you have in each categories? – gyurisc Apr 27 '16 at 09:16
@gyurisc Roughly 10,000 images and 20 categories so max of 500 images per category. – Kroden Apr 27 '16 at 19:12

Using RCNN to autocrop an image

0 Answers0