Footwear examiners are tasked with comparing an outsole impression (Q) left at a crime scene with an impression (K) from a database or from the suspect’s shoe. We propose semi-automated algorithm, MC-COMP-SURF, for comparing two shoe outsole impressions, that relies on robust features (SURF, Bay et al., 2006) on each impression and aligns them using a maximum clique (MC) approach. After alignment, the algorithm is used to extract additional features that are then combined into a univariate similarity score using a random forest (RF). We use a database of shoe outsole impressions that includes images from two models of athletic shoes worn by study participants for about six months. The shoes share class characteristics, and thus the comparison is challenging. We find that RF-SURF outperforms other methods recently proposed in the literature. In good quality images, the algorithm exhibits accuracy in the 96%-98% range. In more realistic scenarios when Q is degraded and partially observed, MC-COMP-SURF still reaches accuracy of about 88%-90%. The algorithm can be implemented with the R package shoeprintr.