A test was performed in Aug 2012 by McGladrey, an independent research company in the USA to evaluate the accuracy of Quividi’s classification algorithms.
Viewers saw a short video clip and at the end of that clip the system ‘read’ them and showed a male video clip to males and a female video clip to females.
The test was very successful and overall we saw a 93% “instant” accuracy rate (i.e. the ability to tell for each single person his or her precise gender) after scanning 856 people.
Here are the highlights:
- Total People Tested: 856
- Percentage of Test Sample that is Male: 72.55%
- Percentage of Test Sample that is Female: 27.45%
- Total Male Accuracy: 95.97%
- Total Males Tested: 621
- Total Males Accurately Read: 596
- Total Males Misread: 25
- Total Females Accuracy: 86.8%
- Total Females Tested: 235*
- Total Females Accurately Read: 204
- Total Females Misread: 31
*52 of the females were wearing baseball caps and our accuracy of this group was only 42%. If we look at the accuracy of the software without any women wearing hats, the accuracy rate increases to over 96%
All in all, the combined “Instant” Accuracy (Males + Females) reached 93.4%
Furthermore, the system altogether output 596 + 31 = 627 Males and 204 + 25 = 229 Females, ie acompounded accuracy of over 97%.