In this work, we present a novel non-rigid shape matching framework based on multi-resolution functional maps with spectral attention. Indeed, existing functional map learning methods all rely on a choice of the critical spectral resolution hyper-parameter, which can severely affect the overall accuracy or lead to overfitting, if not chosen carefully. In this paper, we show that spectral resolution tuning can be alleviated by introducing spectral attention. Our framework is applicable in both su…