Generate 'GT' for new images.

[iPsych]

I just finished to run the demo codes and trying to run some images using demo code in your "Recurrent-Pixel-Embedding-for-Instance-Grouping". When open the main001_visualization.m in demo2_boundary_detection, the code reads .mat instead of .jpg as input. It seems the .mat contains GT variable, but I couldn't find the step to generate GT for new images. If you don't mind, would you let me know how I should generate GT for images?

[aimerykong]

Hello, thanks for your interest in trying the demo.

The mat files containing GT are a format I used for training and evaluation at my own side. I didn't put that on github as I expect folks may generate their own data format for their own use. That's why I only put a few images for people to visualize the results.

The GT's are annotated at pixel level, with different number indicating the instance ID.

Hope this helps!

Regards Shu

On Sat, Jul 27, 2019 at 8:16 AM iPsych [email protected] wrote:

I just finished to run the demo codes and trying to run some images using demo code in your "Recurrent-Pixel-Embedding-for-Instance-Grouping". When open the main001_visualization.m in demo2_boundary_detection, the code reads .mat instead of .jpg as input. It seems the .mat contains GT variable, but I couldn't find the step to generate GT for new images. If you don't mind, would you let me know how I should generate GT for images?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/aimerykong/Recurrent-Pixel-Embedding-for-Instance-Grouping/issues/11?email_source=notifications&email_token=ABRJSJGXDLUXTF2EMUL2H63QBRRDZA5CNFSM4IHJU6XKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HB3JMGA, or mute the thread https://github.com/notifications/unsubscribe-auth/ABRJSJHZRIW4IISLTMGACELQBRRDZANCNFSM4IHJU6XA .