Factors Affecting Face Detection Performance
To get the best performance from face detection, consider the following points:
- Face direction. Face detection performs best when faces are looking towards the camera, so that both eyes are visible, but faces can be detected when viewed in profile (side-on).
- Face size. Media Server is less likely to detect smaller faces than larger ones, particularly in low resolution images or images affected by compression or other artifacts.
-
Media Quality. The quality of the image or video that you send to Media Server can have a significant effect on the performance of face detection:
- Faces should be in focus, not blurry.
- Ideally the image or video should have even illumination, because dark shadows or bright highlights on the face reduce accuracy.
- The image or video should not be affected by significant compression artifacts that can affect some formats (such as highly compressed JPEG images).
- Facial features. Although the face detector can process a relatively wide range of facial expressions, neutral faces are always detected with the greatest reliability. Particularly unusual facial expressions might cause face detection to fail. In addition, spectacles or large amounts of facial hair increase the difficulty in detecting faces. Glasses with dark lenses are likely to cause detection to fail.