Versatile User Identification in Extended Reality using Pretrained Similarity-Learning Model

A pretrained similarity-learning model can effectively identify users in extended reality (XR) environments based on their motion patterns, outperforming a classification-learning baseline, especially when limited enrollment data is available.