Detecting and tracking human faces and facial features (mouth, nose, eyes, eyebrows...) in images and videos enables numerous applications ranging from character animation to market research. Various technologies are available and the choice depends on various factors:

• Is it needed  to detect just the location of the face or also (some of) the facial features such as mouth, nose, eyes, eyebrows?
• Do we work with still images or video?
• What return information is required: 2D image-coordinates or 3D coordinates?
• Is fully automatic operation a must or is some manual setup allowed?

visage|SDK™ VISION-D, VISION-T and VISION-E packages offer solutions for some of the more common situations, and Visage Technologies custom development services let you leverage our expertise to provide a solution to your specific requirements.

Facial Feature Detection

Facial Feature Detection means detecting the face and all its parts - eyes, eyebrows, nose, mouth contour - in still images. Visage Technologies offers fully automatic high performance Facial Feature Detection in the visage|SDK™ VISION-D package. [READ MORE]

Face Detection means finding the location of a face in a still image, without detecting the locations of the facial features.

Facial Feature Tracking

Facial Feature Tracking means finding and tracking (following) the face and all its parts - eyes, eyebrows, nose, mouth contour - in video sequences. Results can be the 2D coordinates of the facial features in the video image and/or 3D coordinates of the features, including full 3D head pose. Visage Techniologies offers real-time Facial Feature Tracking in the visage|SDK™ VISION-T package with both 2D and 3D information available. Tracking itself is fully automatic, but requires a semi-automatic initialisation step. [READ MORE]

Head Tracking

Head Tracking means finding and tracking (following) the head in video sequences, returning the 3D head pose (position and rotation of the head). visage|SDK™ VISION-T package provides Head Tracking that requires a semi-automatic initialization step. Alternatively, a fully automatic head motion estimation is available in the visage|SDK™ VISION-E package.

The term Face Tracking is sometimes used to mean head tracking, or can mean tracking the 2D position of the face in the video image. Both VISION-T and VISION-E packages can be used for Face Tracking.

Eye Location Tracking

Eye Location Tracking (sometimes called Eye Position Tracking) means finding and tracking location of the eyes in video sequences. The location can be returned either in 2D coordinates in the video image, or in 3D space coordinates. visage|SDK™ VISION-T package provides 3D Eye LocationTracking that requires a semi-automatic initialization step. Alternatively, a fully automatic 2D Eye Location Tracking with 3D head motion estimation is available in the visage|SDK™ VISION-E package.

Eye Tracking

Eye Tracking is the process of measuring either the point of gaze ("where we are looking") or the motion of an eye relative to the head. Visage Technologies does not currently offer an Eye Tracking product.

Try visage|SDK now

Back to visage|SDK main page

CONTACT US FOR FURTHER INFORMATION