jponttuset / mcg

Multiscale Combinatorial Grouping - Object Proposals and Segmentation
http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/mcg/
Other
305 stars 139 forks source link

How to match candidate result and bounding box? #17

Closed lssily closed 5 years ago

lssily commented 5 years ago

Hello! Your work is very helpful for my research. When i implement this demo file, I find a question which bewilder me.

I get the candidate result and bbox from the following way, but sometimes the candidate result and bbox are mismatched.

mask = ismenber(candidates.mcg.superpixels, candidates_mcg.labels{id}); box = candidates_mcg.bboxes(id);

There is another way to get the matched candidate result and bbox?

Looking forward to your answer!

jponttuset commented 5 years ago

Dear Issily,

The box candidates and the segmented candidates go through a different NMS (Non-Maximum Supression) algorithm, so the N-th box does not correspond to the N-th segmentation. The two lists are independent.

If you need to work with corresponding lists, simply extract the boxes from the segmentations.

Best,