Monte Carlo human identification refinement using joints uncertainty

In this work, we propose a new method to re-identify the same individual among different people using RGB-D data. Each human signature is combination of soft biometric traits. particular, extract color-based descriptor and local feature through Monte Carlo-based algorithm taking into account uncertainty joints and, applied each descriptor, refines similarity match against spatiotemporal database that updates over time. We analyzed effects Carlo refinement in terms final maximum matching score obtained for two descriptors. addition, tested performance proposed on widely used public dataset one best re-identification methods literature. Our achieves an average recognition rate 99.1 % rank-1 without identification error. Its robustness also makes it suitable industrial applications.