SFINGE 3D: A novel benchmark for online detection and recognition of heterogeneous hand gestures from 3D fingers’ trajectories

•SFINGE 3D, novel benchmark for evaluating online gesture detection and recognition.•13 gestures dictionary, recognition task for detection and recognition with 72 gesture sequences.•Different approaches to test the benchmark: visual rendering and convolutional nerual networks.•Different approaches to test the benchmark: geometry-based and dissimilarity-based classifiers.