prismformore/Multi-Task-Transformer

Questions regarding code for DAVIS demo

j2k0618 opened this issue · 4 comments

j2k0618 commented

Hi,

Thanks for your nice work.

I am very interested about your work, but I seems that your repository do not contains the things with DAVIS demo, could you upload the related code?

Best regards

prismformore commented

@j2k0618 Thank you for your interest in our work! I am afraid I don't have time to release the code of DAVIS recently. Actually what I did was just slightly change the dataloader to load images from DAVIS. I didn't use ground-truth labels from DAVIS. Hope this will be helpful for you : )

j2k0618 commented

@prismformore Thank you for your kind response.
I understand that you succeed to get good results on the unseen DAVIS dataset but I still have one question.
Are there any quantitative comparisons between results and ground-truth labels?

Best regards

prismformore commented

@j2k0618 No because DAVIS doesn't provide multi-task dense labels. We only did qualitative experiments on DAVIS.

Thanks.

  • 👍1
j2k0618 commented

Thanks again @prismformore for your timely and helpful reply!

  • 🎉1