Improving Facial Recognition with Super-Fine Attributes and TensorFlow
Published in
1 min readNov 26, 2018
Hard biometrics — such as facial recognition, DNA detection, iris scanning, and fingerprint analysis — can uniquely identify a person, but they are very challenging to capture on video cameras or still images. This blog post looks into an alternative method of super-fine attributes that rely on soft biometrics, such as gender, age, height, weight, build, hair color and length, etc. to improve facial recognition. Learn how super-fine attributes can be used to classify and create labels when hard biometrics aren’t available.
Read the full article on our blog:
Stay in touch with the latest Altoros’ updates, subscribe to our social accounts: Twitter, Facebook, LinkedIn, Reddit.
