K. K. Sung and T. Poggio. Example-based learning for viewbased face detection. IEEE Trans. PAMI, 20:39--51, 1998.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and extremely rapid object detection. This framework is demonstrated on, and in part motivated by, the task of face detection. Toward this end we have constructed a frontal face detection system which achieves detection and false positive rates which are equivalent to the best published results [16, 12, 15, 11, 1]. This face detection system is most clearly distinguished from previous approaches in its ability to detect faces extremely rapidly. Operating on 384 by 288 pixel images, faces are detected at 15 frames per second on a conventional 700 MHz Intel Pentium III. In other face detection systems, ....

K. Sung and T. Poggio. Example-based learning for viewbased face detection. In IEEE Patt. Anal. Mach. Intell., volume 20, pages 39--51, 1998.

....a cascade which allows background regions of the image to be quickly discarded while spending more computation on promising object like regions. A set of experiments in the domain of face detection are presented. The system yields face detection performace comparable to the best previous systems [18, 13, 16, 12, 1]. Implemented on a conventional desktop, face detection proceeds at 15 frames per second. 1. Introduction This paper brings together new algorithms and insights to construct a framework for robust and extremely rapid object detection. This framework is demonstrated on, and in part motivated by, ....

....and extremely rapid object detection. This framework is demonstrated on, and in part motivated by, the task of face detection. Toward this end we have constructed a frontal face detection system which achieves detection and false positive rates which are equivalent to the best published results [18, 13, 16, 12, 1]. This face detection system is most clearly distinguished from previous approaches in its ability to detect faces extremely rapidly. Operating on 384 by 288 pixel images, faces are detected at 15 frames per second on a conventional 700 1 MHz Intel Pentium III. In other face detection systems, ....

[Article contains additional citation context not shown here]

K. Sung and T. Poggio. Example-based learning for view-based face detection. In IEEE Patt. Anal. Mach. Intell., volume 20, pages 39--51, 1998.

....directly address the issue of eciency in computation and resources required for detection of multiple objects. It should be noted however that this approach has led to successful computer vision algorithms for speci c objects. For example in the context of face detection see Rowley et al. 1998) Sung Poggio (1998). In view of these issues and in order to accommodate the constraints described in the previous section the selection system described below is guided by the following assumptions. The classi cation carried out at each image location Object versus Background must be a very simple ....

Sung, K. K. & Poggio, T. (1998), `Example-based learning for view-based face detection', IEEE Trans. PAMI 20, 39-51.

....information due to color, depth or motion. The generality of our approach is discussed in the concluding section; any potential limitations should then be apparent. A variety of methods have been proposed for face detection, including artificial neural networks (Rowley, Baluja Kanade 1998) (Sung Poggio 1998), support vector machines (Osuna, Freund Girosi 1997) graph matching (Leung, Burl Perona 1995) Maurer von der Malsburg 1996) Bayesian inference (Cootes Taylor 1996) deformable templates (Miao, Yin, Wang, Shen Chen 1999) Yuille, Cohen Hallinan 1992) and those based on color ....

.... of lighting; see for example the discussion in (Ullman 1996) In order to diminish the variation, methods such as those based on neural networks usually require preprocessing (Rowley 1999) for instance subtracting a linear component from the grey level map followed by histogram equalization (Sung Poggio 1998), which can be costly. Instead, the information we extract from the greylevels are comparisons of intensity differences, which are invariant to linear transformations of the greyscale. In Figure 7 we show three versions of a training face together with the detected edges. There is an one ....

[Article contains additional citation context not shown here]

Sung, K. K. & Poggio, T. (1998), `Example-based learning for view-based face detection ', IEEE Trans. PAMI 20, 39--51.

....information due to color, depth or motion. The generality of our approach is discussed in the concluding section; any potential limitations should then be apparent. A variety of methods have been proposed for face detection, including artificial neural networks (Rowley, Baluja Kanade 1998) (Sung Poggio 1998), support vector machines (Osuna, Freund Girosi 1997) graph matching (Leung, Burl Perona 1995) Maurer von der Malsburg 1996) Bayesian inference (Cootes Taylor 1996) deformable templates (Miao, Yin, Wang, Shen Chen 1999) Yuille, Cohen Halliman 1992) and those based on color ....

.... of lighting; see for example the discussion in (Ullman 1996) In order to diminish the variation, methods such as those based on neural networks usually require preprocessing (Rowley 1999) for instance subtracting a linear component from the grey level map followed by histogram equalization (Sung Poggio 1998), which can be costly. Instead, the information we extract from the greylevels are comparisons of intensity differences, which are invariant to linear transformations of the greyscale. In Figure 5 we show three versions of a training face together with the detected edges. There is an one ....

[Article contains additional citation context not shown here]

Sung, K. K. & Poggio, T. (1998), `Example-based learning for view-based face detection ', IEEE Trans. PAMI 20, 39--51.

....which render them rare in the background population. The statistical framework in (Rojer Schwartz 1992) is similar, although they do not suggest a systematic exploration of local features. Finally, there are shared properties with artificial neural networks (Rowley, Baluja Takeo 1998, Sung Poggio 1998), for example the emphasis on learning and the absence of formal models. However, our algorithm is not purely bottom up and our treatment of invariance is explicit; we do no expect the system to learn about it, or about weak dependence or coarse to fine processing. These properties are ....

Sung, K. K. & Poggio, T. (1998), `Example-based learning for view-based face detection ', IEEE Trans. PAMI 20, 39--51.

....from images of real people, quite reliably correspond to people and can be used to identify them. Our trick of projecting classifiers is effective at pruning an otherwise completely unmanageable correspondence search. Future issues include: fusing responses from face finders (such as those of [11, 9]; exploiting patterns of shading on human limbs to get better selectivity (as in [8] determining the configuration of the person, which might tell what they are doing; and exploiting the kinematic similarities between humans and many animals to build systems that can find many different types of ....

K-K Sung and T. Poggio. Example based learning for view based face detection. Ai memo 1521, MIT, 1994.

....The problem of detecting instances from a generic object class without information due to color, depth or motion has been widely studied in the computer vision literature. For example, in the case of faces, a variety of methods have been proposed, including artificial neural networks [12] [13], support vector machines [11] graphmatching [10] Bayesian inference [5] deformable templates [17] and the precursors of our methodology already cited. One of the principal difficulties is the variation in the appearance of faces due to the vagaries of lighting; see for example the discussion ....

K. K. Sung and T. Poggio. Example-based learning for view-based face detection. IEEE Trans. PAMI, 20:39--51, 1998.

No context found.

K. K. Sung and T. Poggio. Example-based learning for viewbased face detection. IEEE Trans. PAMI, 20:39--51, 1998.

No context found.

K. Sung and T. Poggio. Example-based learning for view-based face detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20:39--51, 1998.

No context found.

K. Sung & T.Poggio (1998), "Example-based learning for view-based face Detection", IEEE-PAMI, v20, pp. 39-51.

No context found.

K. Sung and T. Poggio. Example-based learning for viewbased face detection. In IEEE Patt. Anal. Mach. Intell., volume 20, pages 39--51, 1998.

No context found.

K. Sung and T. Poggio. Example-based learning for viewbased face detection. In IEEE Patt. Anal. Mach. Intell.,vol- ume 20, pages 39--51, 1998.

No context found.

K. Sung and T. Poggio, "Example-based learning for viewbased face detection," In IEEE Patt. Anal. Mach. Intell., vol. 20, no. 1, pp. 39--51, 1998.

No context found.

K. Sung and T. Poggio. Example-based learning for view-based face detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20:39--51, 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC