FzuLJ

@lixiang-ucas Hi, I met same problem like yours. Did you solve the problem?

@842430478 Hi, I have the same problem as yours. Do you find any solution to sovle it?

Hi, @ZhouYanzhao. Thank you for your code. I ran the SPN code which is your another work related to object localization before. And recently I notice your new work, PRM....

@vtddggg Yes, I use this proposals method. https://github.com/jponttuset/mcg

@842430478 I think maybe you can select the top 200 ranked candidates from mcg as the "pre-computed object proposals". But my problem is how to convert these proposals to json...

@Yby-Bob Thank you for your advice. I succeeded getting the masks from mcg, and also converted them to json format. But my result is not as good as the paper...

@anoo6504 According to the project COB, I first saved these proposals in .mat format, then converted mat to json format.

@anoo6527 Yes, you can use COB for your dataset. I extracted my dataset proposals using COB successfully.

@llyfzu This is the link of COB: https://github.com/kmaninis/COB

> @FzuLJ Hi, I saw the COB and configured the needed MatCaffe, do I need to train my own COB caffemodel that fits my own datasets?('cause my datasets is about...