T. Maurer and C. v.d. Malsburg. Tracking and learning graphs on image sequences of faces. In Proc. of the Int. Conf. on Arti cial Neural Networks, ICANN, pages 323-328, Bochum, Germany, Jul. 16-19. C. v.d. Malsburg, W. v. Seelen, J. Vorbruggen, B. Sendho (eds.), Springer-Verlag, Berlin, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....emphasis of this approach is robustness on the expense of precision. In [7] a stochastic approach for simultanious tracking and veri cation has been presented. The employed method uses sequential importance sampling (SIS) As templates, discrete gray value templats or bunch graphs are used. In [9] Gabor wavelets and bunch graphs were used for tracking faces. Because of the large number of used wavelets, this approach was able to run with approximately one Hz. In [5] tracking method is presented that is based on Gabor wavelet networks. The method presented here is a generalization of [5] ....

T. Maurer and C. v.d. Malsburg. Tracking and learning graphs on image sequences of faces. In Proc. of the Int. Conf. on Arti cial Neural Networks, ICANN, pages 323-328, Bochum, Germany, Jul. 16-19. C. v.d. Malsburg, W. v. Seelen, J. Vorbruggen, B. Sendho (eds.), Springer-Verlag, Berlin, 1996.

....the face, 1 on the nose, and 8 on the lips (figure 1) To ensure the convexity of the RBF interpolation, the four corners of the mesh are also used as feature points. If viseme images are extracted from the video sequences, feature points selections can be automated using feature tracking systems [16][27] A 3 x 4 mesh grid for the face movements and a 14 x 13 mesh grid for the mouth movements are defined implicitly on the image (figure 2) One face mesh grid consists of 60 x 60 pixels and one mouth mesh grid consists of 6 x 6 pixels. Figure 2. A coarse mesh grid is placed on the face (a) ....

T. Maurer, C. von der Malsburg, Tracking and learning graphs of image sequence of faces, In Proceedings of International Conference on Artificial Neural Networks, Bochum, Germany, 1996

.... for static face analysis with elastic graph matching [8] and coarse to ne correspondence matching [14] Graphs with di erent topologies were needed in order to handle di erent head poses [7] Gabor feature jets have recently been used to track facial feature points on faces rotating in depth [10]. However, each point was treated independently with no global shape model to constrain the tracking. The method was, therefore, susceptible to the inevitable tracking errors which occur due to aperture problems, noise and occlusions. Research carried out at the University of Manchester has ....

T. Maurer and C. von der Malsburg. Tracking and learning graphs on image sequences of faces. In Proc. Int. Conf. on Articial Neural Networks, Bochum, 1996.

....active contour, etc) while allowing affine variations of the facial image. These systems track precisely as shown in the excellent work of [3] but the templates are either of individual persons [3] or are computationally expensive [2; 4] and therefore slow so that tracking is not in RT. In [6] a system is presented that is able to track faces independent of face orientation and gesture. The system uses a wavelet jet bunch graph approach and tracks with less than 1 fps. In this paper we present an approach for RT face tracking that allows arbitrary affine deformations of the facial ....

T. Maurer and C. v.d. Malsburg. Tracking and learning graphs on image sequences of faces. In Int. Conf. on Automatic Face- and Gesture-Recognition, pages 176--181, Killington, Vermont, USA, Oct. 14-16, 1996.

....of individual persons or are computationally expensive [7; 5] so that tracking is not in RT. Other systems use a separate neural network or eigenfaces for verifying for the face during tracking. However, these systems are not invariant with respect to a ne variations of the facial image [13] In [12] a system is presented that is able to track faces independently of face orientation and gesture. The systems uses the wavelet jet bunch graph approach [18] but tracks with less than 1 fps. In this work we will present an approach for RT face tracking that allows arbitrary a ne deformations of the ....

T. Maurer and C. v.d. Malsburg. Tracking and learning graphs on image sequences of faces. In Int. Conf. on Automatic Face- and GestureRecognition, pages 176-181, Killington, Vermont, USA, Oct. 14-16, 1996.

.... for static face analysis with elastic graph matching [8] and coarse to fine correspondence matching [14] Graphs with different topologies were needed in order to handle different head poses [7] Gabor feature jets have recently been used to track facial feature points on faces rotating in depth [10]. However, each point was treated independently with no global shape model to constrain the tracking. The method was, therefore, susceptible to the inevitable tracking errors which occur due to aperture problems, noise and occlusions. Research carried out at the University of Manchester has ....

T. Maurer and C. von der Malsburg. Tracking and learning graphs on image sequences of faces. In Proc. Int. Conf. on Artificial Neural Networks, Bochum, 1996.

.... of nodes for salient points on the basis of common motion has been demonstrated in (Manjunath et al. 1992) Monitoring a rotating object by continuously applying EBGM can then reveal which nodes refer to corresponding fiducial points in different poses (cf. von der Malsburg Reiser 1995; Maurer von der Malsburg 1996). In a boot strapping fashion, object specific grids and object bunch graphs could be established by starting with a crude or general grid, such as utilized in (Lades et al. 1993) to first recognize object classes and establish approximate correspondences between images within a class, on the ....

Maurer, T. and von der Malsburg, C. (1996). Tracking and learning graphs on image sequences of faces. Submitted to the International Conference on Artificial Neural Networks ICANN'96, Bochum.

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