is it something wrong that the code returns only one image' label and mask in mask_head?

[zimenglan-sysu-512]

hi @zjhuang22
i have a question that in this line, it seems that it returns one image's mask and label info. if one gpu holds more than one image, how to get the corresponding mask and label?

[zjhuang22]

hi @zjhuang22
i have a question that in this line, it seems that it returns one image's mask and label info. if one gpu holds more than one image, how to get the corresponding mask and label?

Thanks for your attention. Our code only uses one image for one GPU during the test, without considering more than one image for one GPU during the test, which we will add later.

[zimenglan-sysu-512]

another question, why it returns mask instead of selected_mask in that line?

[zjhuang22]

another question, why it returns mask instead of selected_mask? in that line?

In the post-processing of mask_head, the mask has been selected according to the prediction category.

[zimenglan-sysu-512]

u are right, i forget it. thanks