This example demonstrates how to use the MTCNN detector and MTCNN feature set and to estimate a face quality on an image and to detect attributes on a face.

As said in the introduction page, this repository doesn't provide SDK headers, libraries and tools; you have to obtain them from VisionLabs.

This example assumes that you have read the FaceEngine Handbook already (or at least have it somewhere nearby for reference) and are familiar with some core concepts, like memory management, object ownership and life-time control. This sample will not explain these aspects in detail.

To get familiar with FSDK usage and common practices, please go through Example 1 first.

./Example3 <some_image.ppm>

Warped images with faces.

Detection 1
Rect: x=277 y=426 w=73 h=94
Quality estimated
Quality: 0.955808
Complex attributes estimated
Gender: 0.999705 (1 - man, 0 - woman)
Wear glasses: 0.000118364 (1 - person wears glasses, 0 - person doesn't wear glasses)
Age: 17.7197 (in years)

Detection 2
Rect: x=203 y=159 w=63 h=89
Quality estimated
Quality: 0.955808
Complex attributes estimated
Gender: 0.0053403 (1 - man, 0 - woman)
Wear glasses: 0.000911222 (1 - person wears glasses, 0 - person doesn't wear glasses)
Age: 16.1504 (in years)