D. DeCarlo and D.Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation, CVPR 1996, pp.231-238.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... [41] 2 view point based models [68] Labeled graphs [86] 56] 40] Motion Extraction Holistic Methods Local Methods Dense Optical Flow Dense ow elds [52] 2] Region based ow [57] 65] 85] Motion Models 3D motion models [33] 25] Parametric motion models [9] 84] 3D deformable models [21] 3D motion models [5] Feature Point Tracking Feature tracking [52] 80] 82] 64] 72] Di erence Images Holistic di . imgs [2] 22] 35] 34] Region based di erence images [12] Marker based Highlighted facial features [4] Dot markers [78] 44] Table 1: Facial Feature Extraction Methods: ....

....found in the images. Only frontal faces were allowed and some facial make up was used to enhance contrast. Essa and Pentland [33] employed sophisticated 3D motion and muscle models for facial expression recognition and increased tracking stability by Kalman ltering. DeCarlo and Metaxas [21] presented a formal methodology for the integration of optical ow and 3D deformable models and applied it to human face shapes and facial motion estimation. A relatively small number of parameters were used to describe a rich variety of face shapes and facial expressions. Eisert and Girod [25] ....

[Article contains additional citation context not shown here]

D. DeCarlo and D. Metaxas. The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation. In Proceedings of the International Conference on Computer Vision and Pattern Recognition (CVPR '96), pages 231-238, 1996.

....[16] While appropriate for some user interface applications, these sequence specific approaches are difficult to extend to arbitrary image sequences. Tracking approaches for generic scenes have typically used extracted edge information [8, 11, 15, 17, 30, 32] optical flow [3, 19, 48] or both [7, 43, 46]. Edges are first extracted using some standard technique and then a match metric is defined that measures the distance from predicted model edges (e.g. limb boundaries) to detected edges in the scene. Probabilistic tracking methods convert this match metric into an ad hoc probabilistic ....

....More information about the limb appearance can be derived from the assumption of temporal brightness constancy that two image locations originating from the same scene location at two consecutive time instants have the same intensity. This assumption is used widely for track ing of humans [7 38 46]. There are two problems with this assumption. First since there is no absolute model of the limb appearance any errors in the estimated motion will accumulate over time and the model may drift off the tracking target and eventually follow the background or some other object. To avoid this drift ....

[Article contains additional citation context not shown here]

DeCarlo, D. and D. Metaxas: 1996, 'The integration of optical flow and deformable models with applications to human face shape and motion estimation '. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR. pp. 231-238.

....face, and a prepared representative model of the individual. This paper focuses on stages highlighted in orange. II. BACKGROUND PDFA requires facial sensing, and computer vision research is extensive in this area [6, 7, 8, 10, 13, 30 32] One approach uses whole image optical flow measures [3, 4, 33]. These are currently too slow for use in a live performance. When live performance is important, as in virtual tele conferencing, tracked feature points can provide sparse but real time updates. We employ an existing commercial feature tracking system [11] that works without markers. It reports ....

D. DeCarlo, D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation," CVPR'96, pp.231-238.

....(page 37) The same is true of body motions. In a passage from page 103 it seems as though Hiro had been performing the same actions as his avatar, but in a later passage (page 420) his avatar goes slack while he is busy in reality. Currently, face input (from unmarked faces) is nearly real time [18, 19], and whole body motion capture from video is not far off [27] Current laser scanning systems are not close to real time. Some body and facial motions can be previously captured and replayed in context. This requires some parameterization so they can be executed in the current context and ....

D. DeCarlo and D. Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation. In Proc. CVPR, pages 231-- 238. IEEE Press, 1996.

....face, and a prepared representative model of the individual. This paper focuses on stages highlighted in orange. II. BACKGROUND PDFA requires facial sensing, and computer vision research is extensive in this area [6, 7, 8, 10, 13, 30 32] One approach uses whole image optical flow measures [3, 4, 33]. These are currently too slow for use in a live performance. When live performance is important, as in virtual tele conferencing, tracked feature points can provide sparse but real time updates. We employ an existing commercial feature tracking system [11] that works without markers. It reports ....

D. DeCarlo, D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation," CVPR'96, pp.231-238.

No context found.

D. DeCarlo and D.Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation, CVPR 1996, pp.231-238.

....by Zheng and Chellapa[44] Keywords: Physics based modeling, shape from shading, deformable models, illuminant estimation, diffuse reflectance. 1. Introduction The integration of visual cues within a physics based deformable model framework has been attempted recently by several researchers [6, 26, 9] due to its potential for improved shape estimation. In all previous attempts, illumination constraints such as those appearing in the shape from shading problem, have never been considered. This is due to the nonlinear nature of the constraints and the fact that numerically robust methods for ....

....holonomic constraints. Once a constraint is satisfied, its derivative must remain zero, for the constraint to remain satisfied. We incorporate these constraints in a deformable model formulation, using the method of Lagrange multipliers. The case of linear non holonomic constraints was treated in [6], and the case of linear holonomic constraints was treated in [25] Lagrange Multipliers. In order to recover the shape parameters based on the constraint information at m points of our model, we will have an m 1 constraint vector C, which is incorporated in (3) based on the theory of Lagrange ....

D. DeCarlo and D. Metaxas. The integration of optical flow and deformable models: Applications to human face shape and motion estimation. In CVPR96, pages 231--238, 1996.

....lead to difficult problems when attempting to integrate those features within a particular representation or optimization procedure. In the area of constraint integration, particularly in a deformable model setting, contours and stereo ( 26] shading and stereo ( 7] contours and optical flow ([3, 4]) and shading ( 18] have been introduced, with good results. Beyond the particular choice of sources of information to use, the approaches differ in the way they fuse their component information. Some combine the information in a symmetric manner and weight them statistically (soft constraints) ....

D. DeCarlo and D.Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation, CVPR 1996, pp.231-238.

....same arm tracking experiment and each column to the views captured from one of the cameras for each of the three experiments. 6. 2 Integration of Visual Cues for Facial Tracking We have recently proposed a theory for the integration of optical flow information within a deformable model framework [1, 2, 5]. Based on this approach, the optical flow is treated as a nonholonomic constraint, i.e. it constrains the velocity of the model parame BMVC99 6 (a) b) c) d) e) f) Figure 5: A face motion and expression tracking example. ters, within a deformable model framework. As a result, we can ....

D. DeCarlo and D. Metaxas. "The integration of optical flow and deformable models with applications to human face shape and motion estimation". Procs. of the IEEE Computer Society on Computer Vision and Pattern Recognition, pp. 231-238, June 1996.

No context found.

D. DeCarlo and D. Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation. In Computer Vision and Pattern Recognition, 1996.

No context found.

D. DeCarlo and D. Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation. In Proceedings, CVPR96, pages 231-- 238, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation, " Proc. CVPR '96, pp. 231-238, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The integration of optical flow and deformable models with applications to human face shape and motion estimation," CVPR, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation, " Proceedings CVPR '96, pp. 231-238, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The integration of optical flow and deformable models with applications to human face shape and motion estimation", in Proc. Computer Vision and Pattern Recognition, San Francisco, CA, June 1996, pp. 231--238.

No context found.

D. DeCarlo and D. Metaxas, "The integration of optical flow and deformable models with applications to human face shape and motion estimation," in Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, June 1996, pp. 231--238.

No context found.

D. DeCarlo and D. Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation. In Proceedings, CVPR96, pages 231-- 238, 1996.

No context found.

DeCarlo D. and Metaxas D., "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation", Proc. CVPR'96, IEEE Computer Society Press, pp. 231-238, 1996.

No context found.

D. DeCarlo and D. Metaxas. The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation. In Conference on Computer Vision and Pattern Recognition, pages 231#238, 1996.

No context found.

DeCarlo D., Metaxas D., "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation" In Proceedings CVPR 1996.

No context found.

D. DeCarlo and D. Metaxas. The integration of optical flow and deformable models with applications to human face shape and motion estimation. In CVPR, pages 231--238, 1996.

No context found.

D. DeCarlo and D. Metaxas. The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation. In CVPR'96, pp. 231-238, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation, " Proc. CVPR '96, pp. 231-238, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation, " Proc. CVPR '96, pp. 231-238, 1996.

No context found.

D. DeCarlo and D. Metaxas, "The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation," Proc. Computer Vision and Pattern Recognition (CVPR) 96, IEEE CS Press, Los Alamitos, Calif., 1996, pp. 231-238.

First 50 documents  Next 50

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