Volker Kruger, Sven Bruns, and Gerald Sommer. Efficient head pose estimation with gabor wavelet networks. Proc. British Machine Vision Conference, Bristol,UK, 2000. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Robust 3D Head Pose Classification Using Wavelets - Motwani (2003) (Correct)
....semi profile, and right profile) with very little labeled data. Azarbayejani et al. 1] used an extended Kalman filter to recover head pose from between ten and twenty tracked feature points. However, the estimation accuracy was found to degrade as the rotation angle increased. Kruger et al. [20] used computationally intensive elastic graph matching to locate and estimate the pose of faces. The graphs consisted of connected nodes of Gabor filter jets. Different graph models were needed for different poses, leading to a poorly integrated and computationally expensive approach. Borovikov ....
Volker Kruger, Sven Bruns, and Gerald Sommer. Efficient head pose estimation with gabor wavelet networks. Proc. British Machine Vision Conference, Bristol,UK, 2000.
Efficient Real-Time Face Tracking in Wavelet Subspace - Feris, Cesar, Jr., Krueger (2001) (Correct)
....applications, as well as for tele conferencing and teleteaching tasks. For gesture, gaze and pose estimation applications, tracking has to be not only fast, but also precise. A variety of tracking approaches already exists [6, 2, 1] In this work, we will use Gabor Wavelet Networks (GWNs) [4, 5] in order to represent the face to be tracked. It is worth saying that we have already discussed a GWN based realtime tracking method in [6] The method discussed in the present paper differs from that approach in the sense that it is performed in a low dimensional wavelet subspace, posing a ....
....analysis and for motion estimation, allows to map an image into the low dimensional space R , where all subsequent computations can be carried out. Furthermore, the careful selection of the wavelets for defining a GWN ensures that almost all visual information is preserved (see fig. 1 and [5]) The method has the further advantage, that its precision and computation time can be adapted to the needs of a given task. When fast and only approximate tracking is needed, a small number of filtrations is usually sufficient to realize tracking. When high precision tracking is needed, the ....
V. Kr uger, S. Bruns, and G. Sommer. Efficient head pose estimation with gabor wavelet networks. In Proc. British Machine Vision Conference, Bristol, UK, Sept. 12-14, 2000.
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