Human-centric
AI Computer Vision
3DiVi Inc., founded in 2011, is one of the leading developers of AI and machine learning (ML) technologies for computer vision.
Your request has been successfully sent. We'll get in touch shortly.
THANK YOU!
3DiVi News

When Humans Make Mistakes

Ever mistaken someone you barely know for someone else? It’s more common than you think—even trained security staff can make mistakes, especially under stressful conditions, poor lighting, or awkward angles. While humans are great at recognizing familiar faces, we all have our limits.
Here’s why we get face recognition wrong:
  • Familiar Faces: We’re excellent at recognizing people we know, with accuracy over 90%. But when it comes to barely known ones, our accuracy drops to 60-70%.

  • Age: As we age, our ability to recognize faces can decline. Kids and older adults are also harder to identify due to changing facial features.

  • Cross-Racial Recognition: We tend to recognize faces from our own ethnic group better. This “cross-racial” effect happens because we interact more with people from our own background.

  • Emotions and Expressions: Faces with emotional expressions are easier to recognize, thanks to their unique features.

  • Image Quality: Low light or obscured faces (masks, glasses) can make recognition tricky. Better lighting and clearer faces mean better accuracy.
Humans are great at recognizing familiar faces, but when it comes to little-known ones, accuracy takes a dive. In contrast, computer vision algorithms can achieve near 100% accuracy, often surpassing human recognition abilities—especially in challenging situations. See for yourself in our live demo — it's just a click away.
News