Do you have results of k-means Mask Transformer on ADE20K dataset?

Question

MengzhangLI opened this issue 2 years ago · 3 comments

Hi, first congrats to ECCV'2022 acceptance of k-means Mask Transformer.

I think it is an excellent work but I did not find results of ADE20K dataset which is widely used as benchmark in semantic segmentation task.

Could you share the performance of k-means Mask Transformer on ADE20K dataset if you implemented it?

Best,

Answer 1 · 2022-08-06T17:21:04.000Z

Hi, thanks for the interest in our work.

We have not tried kMaX-DeepLab on ADE20K yet, perhaps we will try it in the near future, please stay tuned!

Answer 2 · 2022-10-18T20:51:57.000Z

Hi,

We have just updated kMaX-DeepLab's model zoo with ADE20K results, welcome to check it!

Answer 3 · 2022-10-19T08:02:12.000Z

Hi,

We have just updated kMaX-DeepLab's model zoo with ADE20K results, welcome to check it!

Copy that. Thank you so much.