6 Conclusions

The experimental results and the theoretical arguments exposed above seem to claim that the HMM paradigm deserves to be confirmed, whereas future work should concentrate upon a better shape representation. The measuring functions proposed in [ 9 ] must be improved since they are particularly sensitive to small changes in the position of the center of mass of the images. Further developments could come, from integrating the qualitative modeling philosophy presented in this paper with quantitative techniques such as those proposed in [ 7 , 6 ] which describe the 27 degrees of freedom kinematics of the hand.