关于AxialAttention中kernel_size的问题

[pengweiweiwei]

您好！想请问一下，如果我不是用AxialAttention作为backbone，而是仅仅用它作为一个attention机制的话，效果怎么样？此外，AxialAttention中的kernel size是不是取决于输入的feature map的尺寸，那么如果我在train和infer的时候输入尺寸不同，feature map尺寸也不同，是不是会出现问题呢？希望能够得到解答，非常感谢！

[csrhddlam]

1. In the paper we only explored using axial-self-attention as a backbone, but one can certainly explore extending it to non-self-attention or cross-attention, e.g. attending from a 2D map to another 2D map.
2. The current code only supports global attention and the same train and eval resolution. But in general, axial-attention is not limited to different input resolutions: one should use local axial-attention with a fixed span (e.g. 65) for that. In the paper, we used span = (65x65) for the main panoptic segmentation results, and this allows us to do multi-scale inference with different input resolutions.