SAFR trains its algorithms on large datasets that are demographically representative of a global population. This training ensures our algorithms are robust to variations in ethnicity, race, gender, age, and facial hair.
The SAFR algorithm's resilience to these variations has been documented by NIST indirectly in Faces in the Wild evaluations, and directly as part of their bias and race analysis. SAFR performs well with Faces in the Wild to recognize faces with varying degrees of pitch, roll, yaw, and occlusion. More details on NIST testing can be found at https://safr.com/general/nist-test-confirms-safr-delivers-the-highest-effective-accuracy-for-live-video/. Further details can be found at https://safr.com/general/5-ways-safr-excels-at-facial-recognition-for-live-video-and-why-you-should-care/.
The algorithm exhibits high performance uniformity across gender and skin tones. SAFR ranks second (out of 103 algorithms) among the least variable algorithms for gender and skin tone according to an evaluation by National Institute of Standards and Technology (NIST) using a sample of the worldwide population, with an accuracy variance of <0.25% (@FMR 1:1K). SAFR ranks in the top five facial recognition systems for lowest bias across tested racial groups. See https://safr.com/general/recognizing-bias for more information.
You don't need to specially configure SAFR to take advantage of the low bias recognition; this feature is inherent to the SAFR facial recognition algorithm.