Refining Face Tracking with Integral Projections

Abstract

Abstract. Integral projections can be used, by themselves, to accurately track human faces in video sequences. Using projections, the tracking problem is effectively separated into the vertical, horizontal and rotational dimensions. Each of these parts is solved, basically, through the alignment of a projection signal -a one-dimensional pattern-with a projection model. The effect of this separation is an important improvement in feature location accuracy and computational efficiency. A comparison has been done with respect to the CamShift algorithm. Our experiments have also shown a high robustness of the method to 3D pose, facial expression, lighting conditions, partial occlusion, and facial features.

Discriminative Spatiotemporal Local Binary Pattern with Revisited Integral Projection for Spontaneous Facial Micro-Expression Recognition

“…Essentially, the integral projection can capture the common underlying structure of face images of the same face subject. In other words, it is very relative to face identity [41], [42]. However, it is double that whether integral projection works for micro-expression.…”

“…Image projection techniques are classical methods for pattern analysis, widely used, e.g., in motion estimation [40] and face tracking [41] [42], as they enhance shape properties and increase discrimination of images. Integral projection provides simple and efficient computation in computer vision amongst image projection techniques.…”

Mouth area analysis by the use of selected spectral energies

“…The technique has been pioneered by Kanade for face recognition, [10], used afterwards by Brunelli and Poggio as a feature for the same purpose, [11] and by Mateos, [12] and Saxena et al [13] for localizing particular face features like mouth and eyes. For an image I(i, j) with M × N pixels, the horizontal projection is:…”

Facial Micro-Expression Recognition Using Spatiotemporal Local Binary Pattern with Integral Projection

“…But most of previous methods used in micro-expression recognition just consider texture information to represent face images, but ignore the shape information of face images. Image projection techniques are classical methods for pattern analysis, widely used, e.g., in motion estimation [18] and face tracking [12,13], as they enhance shape properties and increase discrimination of images. One image projection technique, integral projection, provides simple and efficient computation as well as a very interesting set of properties.…”