ACTION RECOGNITION BASED ON DEPTH MAPS

摘要

A plurality of depth maps corresponding to respective depth measurements determined over a respective plurality of time frames may be obtained. A plurality of skeleton representations respectively corresponding to the respective time frames may be obtained. Each skeleton representation may include joints associated with an observed entity. Local feature descriptors corresponding to the respective time frames may be determined, based on the depth maps and the joints associated with the skeleton representations. An activity recognition associated with the observed entity may be determined, based on the obtained skeleton representations and the determined local feature descriptors.

主权项

1. A system comprising:
a device that includes at least one processor, the device including an activity recognition engine comprising instructions tangibly embodied on a computer readable storage medium for execution by the at least one processor, the activity recognition engine including:
a depth map acquisition component configured to obtain a plurality of depth maps corresponding to respective depth measurements determined over a respective plurality of time frames;a skeleton acquisition component configured to obtain a plurality of skeleton representations respectively corresponding to the respective time frames, each skeleton representation including at least one joint associated with an observed entity;a local feature descriptor determination component configured to determine local feature descriptors corresponding to the respective time frames, based on the depth maps and the joints associated with the skeleton representations; andan activity determination component configured to determine at least one activity recognition associated with the observed entity, based on the obtained skeleton representations and the determined local feature descriptors.