Volumetric descriptors

A computer-based multi-modal sensitive recognition system, comprising:

• a non-transitory, computer readable memory storing software instructions of an implementation of a multi-modal recognition algorithm;
• and
• at least one processor coupled with the memory operating as a multi-modal recognition engine and, upon execution of the multi-modal recognition algorithm, configured as a special-purpose processor to:
• obtain a first digital data set according to a first modality, wherein the first digital data set includes digital representations of an object at different values of a first dimension of relevance of the first modality;
• select at least one reference location associated with the object;
• derive at least one first modal descriptor for the first modality by:
• deriving, according to an implementation of a first feature detection algorithm, a first set of feature descriptors for the at least one reference location and at the different values of the corresponding first dimension of relevance;
• calculating a first set of differences between the feature descriptors in the first set of feature descriptors;
• and
• aggregating the first set of differences into the at least one first modal descriptor representing differences across the different values of the first dimension of relevance;
• and
• configure a device to initiate an action as a function of the at least one first modal descriptor.