Q2A: Querying Implicit Fully Continuous Feature Pyramid to Align Features for Medical Image Segmentation

arXiv:2404.09472v1 Announce Type: new
Abstract: Recent medical image segmentation methods apply implicit neural representation (INR) to the decoder for achieving a continuous coordinate decoding to tackle the drawback of conventional discrete grid-based data representations. However, the INR-based decoder cannot well handle the feature misalignment problem brought about by the naive latent code acquisition strategy in INR. Although there exist many feature alignment works, they all adopt a progressive multi-step aligning paradigm on a discrete feature pyramid, which is incompatible with …

abstract apply arxiv continuous cs.cv data decoder decoding feature features grid however image medical pyramid representation segmentation the decoder type