Abstract.
Recent studies on three-dimensional face recognition pro- posed to model facial expressions as isometries of the facial surface. Based on this model, expression-invariant signatures of the face were con- structed by means of approximate isometric embedding into flat spaces. Here, we apply a new method for measuring isometry-invariant similarity between faces by embedding one facial surface into another. We demon- strate that our approach has several significant advantages, one of which is the ability to handle partially missing data. Promising face recogni- tion results are obtained in numerical experiments even when the facial surfaces are severely occluded.