Y. Wu and T. Huang, "Capturing articulated human hand motion: A divide--and--conquer approach," in International Conference on Computer Vision, 1999, pp. 606--611.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....lie in the high number of degrees of freedom and the problem of occlusions. Two general approaches have been suggested for this problem. Model based approaches try to estimate the position of a hand by projecting a 3 D hand model to image space and comparing it with image features (fingertips [25, 24, 28], line segments [25] Spline and quadricsbased hand shape models were used in [23, 27] to minimize differences between the silhouette of the projected model and the data. Others [31, 25] have used stereo to avoid occlusions. Appearance based approaches estimate hand postures directly from the ....

Ying Wu and T. S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In ICCV '99, pg 606-611, Corfu, Greece, Sept. 1999.

....user, make a hierarchical estimation of the model configuration and conserve all possible configurations when ambiguous solutions appear. However, their system is di#cult to evaluate since no performance measurements are given and the only example is a 3 dof finger moving in a plane. Wu and Huang [11] track separately hand global (palm position and orientation) and local (finger configurations) features. They combine inverse kinematics, genetic algorithms and least median squares in order to track a 21 dof hand with a single camera. However, their results seem no more precise than others, ....

Wu, Y., Huang, T.: Capturing articulated human hand motion: A divide--and-- conquer approach. In: International Conference on Computer Vision. (1999)

....mesh which is constructed via PCA from training examples. Real time tracking is achieved by finding the closest possibly deformed model matching the image. In [7] a stereo hand tracking system using a 2D deformable model was presented. A two step algorithm to estimate the hand pose is proposed in [25], first estimating the global pose and subsequently finding the configuration of the joints. However, the algorithm relies on the assumption that all fingertips are visible. Recently, a vision based drawing system was proposed in [16] The 2D shape of a hand is modeled with Bsplines and ....

Y. Wu and T. S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, Sept. 1999.

....mesh which is constructed via PCA from training examples. Real time tracking is achieved by finding the closest possibly deformed model matching the image. In [4] Cipolla and Hollinghurst presented a stereo handtracking system using a 2D model deformable by affine transformations. Wu and Huang [15] proposed a two step algorithm to estimate the hand pose, first estimating the global pose and subsequently finding the configuration of the joints. However, their algorithm relies on the assumption that all fingertips are visible. Recently, Isard and MacCormick [9] have presented a vision based ....

Y. Wu and T. Huang. Capturing articulated human hand motion: A divide-andconquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, September 1999.

....to this reconstruction. A physically motivated force model is used to attract the 3D model to the reconstruction. However, the disparity maps are not very accurate and the costs to obtain a 3D depth map are very high. A two step algorithm to estimate the hand pose is proposed by Wu and Huang in [42], first estimating the global pose and subsequently finding the configuration of the joints. The problem of determining the hand pose is formulated as a least median of squares problem, and finding local finger motion is solved as an inverse kinematics problem. A genetic algorithm is used to ....

Y. Wu and T. S. Huang. Capturing articulated human hand motion: A divide-and- Bibliography 63 conquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, September 1999.

....is the model based approach, which takes advantage of 3D hand models. Hand states could be recovered by matching the projected 3D model and observed image features, so that the problem becomes a search problem in a high dimensional space. Di#erent image observations have been studied. Fingertips [7, 15, 16] could be used to construct the correspondences between the model and the images. However, the robustness and accuracy largely depends on the performance of fingertip detection. Line features were employed in [12, 14] to enhance the robustness. An exact hand shape model was built by splines in ....

....hand configuration space could be a small constrained subspace in the state space R D . The constraints could dramatically reduce the search space in capturing hand articulation. Although some simple and closed form constraints have been found in biometrics and applied to hand motion analysis [7, 6, 16], more investigations on the representations and utilizations of the constraints need to be conducted. In this paper, we propose an e#ective method to capture hand articulation by integrating constraints of natural hand motion. Our study of natural hand motion shows that the hand configuration ....

Ying Wu and Thomas S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In Proc. IEEE Int'l Conf. on Computer Vision, pages 606--611, Corfu, Greece, Sept. 1999.

....However, those devices are either inconvenient or unnatural in the sense of 3D or high DOF inputs. Human body, such as hand, is being considered as a natural input device in human computer interaction (HCI) which motivates the research of tracking, analyzing and recognizing human body movements [7, 12, 13]. Although the goal of natural user interfaces is to recognize and understand the movements of human body, the first step to achieve this goal is to reliably localize and track human body parts, such as face and hand. Magnetic sensors have been used to supply some motion information directly, ....

....a large variety of pictures. And our localization system that integrates this color segmentation algorithm has run under a wide range of operating conditions. The global hand tracking system based on our color segmentation supplies some inputs for our articulated hand motion capturing algorithm[12]. Experiments show that our color segmentation algorithm is fast, automatic and accurate, and the proposed localization system is robust, real time and reliable. This color segmentation algorithm can also be applied to other segmentation tasks. 5.1. Performance of Segmentation One parameter we ....

Ying Wu, Thomas S. Huang, "Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach ", IEEE Int'l Conf. on Computer Vision, Corfu, Greece, 1999

....to recognize hand signs even from different viewing directions. One approach is the 3D model based approach, in which the hand configuration is estimated by taking advantage of 3D hand models (Davis and Shah 1994; Heap and Hogg 1996; Kuch and Huang 1995; Lee and Kunii 1995; Rehg and Kanade 1995; Wu and Huang 1999). Since hand configurations are independent to view directions, these methods could directly achieve view independent recognition. However, since a classification of hand postures is often enough in many other applications such as commands switching, an alternative approach is appearance based ....

Wu, Y., and Huang, T. S. 1999. Capturing articulated human hand motion: A divide-and-conquer approach. In Proc. IEEE Int'l Conf. on Computer Vision, 606--611.

....mice, wands and joysticks are the fundamental controlling and navigating devices. However, those devices are unnatural. Human body are being considered as a natural device in human computer interaction (HCI) which motivates the research of tracking, analyzing and recognizing human body motion[7, 12, 13]. Although the goal is to understand the human body movements, the first step to achieve this goal is to reliably localize and track human faces and hands in image sequences. The di#culties in visual tracking come from clutter background and unknown lighting conditions. When it needs to track ....

Ying Wu, Thomas S. Huang, "Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach", IEEE Int'l Conf. on Computer Vision, Corfu, Greece, 1999

....of the fingertip in its local frame and reference frame, the inverse kinematics problem is to find the joint angles. Generally, gradient based methods are taken to solve this problem by deriving the kinematical Jacobian [13] There are other alternatives in the literature such as genetic algorithm [20] and neural network methods [2] Since it is an ill posed problem, unique solution can only be found by adding some constraints. Joint angles should be in certain range such as 0 # # MCP # 90 0 and 15 0 # # MCP AA # 15 0 [10] At the same time, DIP and PIP joints are not independent ....

....motion can be expressed as: M = M(R, t, #) 4) One possible way to analyze hand motion is the appearance based approach which emphasis the analysis of hand shapes in images [12] However, local hand motion is very hard to estimate by this means. Another possible way is the model based approach [3, 6, 9, 10, 13, 14, 18, 20]. With single calibrated camera, local hand motion parameters can be estimated by fitting the 3D model to the observation images. Multiple camera settings are helpful to deal with occlusion [10, 13, 18] The use of a 3D model can largely alleviate the problem of depth ambiguity since the structure ....

[Article contains additional citation context not shown here]

Ying Wu, Thomas S. Huang, "Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach ", Proc. of the IEEE Int'l Conf. on Computer Vision, 1999.

No context found.

Y. Wu and T. Huang, "Capturing articulated human hand motion: A divide--and--conquer approach," in International Conference on Computer Vision, 1999, pp. 606--611.

No context found.

Y. Wu and T. S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, September 1999.

No context found.

Wu, Y., Huang, T.: Capturing articulated human hand motion: A divide--and--conquer approach. In: International Conference on Computer Vision. (1999) 606--611

No context found.

Wu, Y., Huang, T.: Capturing articulated human hand motion: A divide--and-- conquer approach. In: International Conference on Computer Vision. (1999) 606-- 611

No context found.

Wu, Y., Huang, T. S., "Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach", in Proceedings of the IEEE International Conference on Computer Vision, Greece, 1999, pp. 606-611.

No context found.

Y. Wu and T. Huang. Capturing articulated human hand motion: A divide-andconquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, September 1999.

No context found.

Ying Wu and Thomas S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In ICCV (1), pages 606--611, 1999.

No context found.

Y. Wu and T. S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, Sept. 1999.

No context found.

Y. Wu and T. Huang. Capturing articulated human hand motion: A divide-andconquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, September 1999.

No context found.

Y. Wu and T. Huang. Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach. Proc. of the International Conference on Computer Vision, 1999.

No context found.

Y. Wu and T. S. Huang, "Capturing articulated human hand motion: A divide-andconquer approach," in Proc. IEEE Int'l Conf. on Computer Vision, Sept. 1999, pp. 606-- 611.

No context found.

Y. Wu and T. S. Huang. Capturing articulated human hand motion: A divide-and-conquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, pages 606--611, Corfu, Greece, September 1999.

No context found.

Y. Wu and T. S. Huang. Capturing articulated human hand motion: A divide-andconquer approach. In Proc. 7th Int. Conf. on Computer Vision, volume I, 606--611, Corfu, Greece, September 1999.

No context found.

Ying Wu Huang, T.S. Capturing Articulated Human Hand Motion: A Divide-andConquer Approach. In IEEE Computer Vision, 1999. Vol. 1 pp. 606-611.

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