Subject-Level Unknown-Identity Identification from Leap Motion Controller 2 Hand Landmarks
2026-06-22 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMachine Learning
AI summaryⓘ
The authors studied how to recognize people using hand data from a Leap Motion Controller on a dataset of hand poses. They improved the hand data by adding measurements like fingertip-to-palm distances and angles between fingers. They tested methods to identify known people and also reject unknown individuals using a special way of splitting the data so unknown subjects were truly unseen. Among several models tested, a tree-based method called Extra Trees worked best at telling apart known and unknown people. Their work shows it is possible to use simple hand measurements to identify or reject people without contact, even in small groups.
Leap Motion Controllerhand landmarkssubject recognitionunknown-subject rejectionML2HP datasetLeave-One-Subject-Out (LOSO)Extra Treesopen-set recognitionembeddingcontactless biometrics
Authors
Bahar Moharrer, Susanna Cifani, Marco Raoul Marini, Luigi Cinque, Maria De Marsico
Abstract
This work studies subject recognition from Leap Motion Controller 2 (LMC2) hand landmark data under a subject-level unknown-identity identification protocol on the Multi View Leap2 Hand Pose (ML2HP) dataset. Using only the landmark modality, we retain the original geometric representation and enrich it with fingertip-to-palm distances and palm-normalized inter-finger angular descriptors. Evaluation is performed under a Leave-One-Subject-Out (LOSO) protocol in which, for each outer fold, one subject is excluded from the enrolled set and treated as unknown at test time. To avoid tuning on the true outer unknown subject, the unknown-rejection threshold is selected in an inner validation step by temporarily withholding one enrolled subject from the inner gallery and using it only for threshold estimation. We compare a tree ensemble baseline with two neural alternatives: a learned embedding baseline based on centroid matching and cosine-similarity-based rejection, and an MLP+OpenMax model, which represents a more established open-set recognition approach. Under this evaluation setup, Extra Trees remains the strongest overall method, indicating that the main challenge on this benchmark is not enrolled-subject discrimination alone, but robust score separation between known and unknown probes. The results support the feasibility of compact, interpretable landmark-based descriptors for contactless hand-based unknown-subject rejection and identification on a small-cohort dataset.