Open yiminglin-ai opened 4 years ago
same question,+1
@ylin-ai 'we improve the original ResNet with the dilation strategy and the final feature map size is set to 1/8 of the input image'
@yangdonghan50 hi, do you know how the pooling sizes 20 and 12 are inferred? or the number 20 and 12 were just set to the given proportion of 480? Thanks !!!
Hi, Thank you for open-sourcing the amazing work. In the
SPHead
module, the pool_size parameter er is set to 20, 12.. Can I ask why these two values? If the input image is480x480
as inade20k
, the feature map produced by the ResNet backbone is15x15
(32 down) which is smaller than20x20
. It actually becomes an upsampling operation rather than pooling.