Uniqueness of ear morphology has been a widely debated issue in cephaloscopy, but past studies used only two-dimensional approaches. In the current investigation, the right and left ears of 10 healthy adults were imaged twice by stereophotogrammetry at the interval of a few seconds. The ear images obtained from the two acquisitions were superimposed both within subject (group of matches) and among subjects (group of mismatches). A point-to-point root mean square distance was calculated between the two three-dimensional models. Differences according to side and group were assessed by two-way analysis of variance. In total, 200 superimpositions were performed. On average, the point-to-point root mean square distance was 0.31 mm in cases of matches and 1.43 mm in cases of mismatches: differences were statistically significant (p < 0.01). Results provided quantitative data for the assessment of uniqueness of ear morphology, highlighting differences based on their three-dimensional morphology.