Machine Learning and Robot Perception - Bruno Apolloni et al (Eds) Part 4 potx

... Mueller R. Etienne-Cummings, J.Van der Spiegel and Mao-Zhu Zhang, “A foveated silicon retina for two-dimensional tracking”, IEEE Trans. on Circuits and Systems II: Analog and Digital Signal Processing, ... spatial domain vision op- erations. Next section presents classic vision algorithms (space-variant op- tical flow, stereo disparity, anisotropic diffusion, corner detection and et...

Ngày tải lên: 10/08/2014, 04:22

... region-growing method, projection method, and sculpting-based method. In the next sub-sections we will ex- plain each of these categories. 6 .4. 1 Volumetric Methods The volumetric method consists of two stages. ... α-shape of a finite point set S is a polytope uniquely determined by S and the param- eter α that controls the level-of-detail. A subset T ⊆ S of size |T | = k +1with 0 ≤ k ≤ 2...

Ngày tải lên: 10/08/2014, 04:22

... of 3-D parameters from calibrated sets of 2-D parameters is computationally very efficient, requiring only a small fraction of computational power as compared to the low-level segmentation algorithms[5]. ... typical value for β is 4 Ns m . Distributed Integration Once the global forces are projected back into the allowable subspace and corrected for discretization error, all further com- p...

Ngày tải lên: 10/08/2014, 04:22

... global and/ or local object specific detail. Global details do not have a precise geometric location in- side the object, such as statistics of gray levels or colors, textures, etc. Lo- cal details ... d Fig. 1. 14. Examples of target box This function provides a measure for each d i value; it is evaluated along the diagonals for each target box, and averaged through all target boxes...

Ngày tải lên: 10/08/2014, 04:22

... la Escalera, A., Salichs, M.A., (1998) “Landmark perception planning for mobile robot localization”. IEEE International Conference on Robotics and Automation, vol. 3, pp. 342 5-3 0. 4 Balkenius, ... scene image using genetic algorithms and neural networks”. Interna- tional Conference on Industrial Electronics, Control and Instrumenta- tion, pp.183 8-1 843 . 2 Argamon-Engels...

Ngày tải lên: 10/08/2014, 04:22

... Ve- gas,USA, Oct. 19 94. 40 . E.R. Kandel, J.H. Schawartz and T.M. Jessell, “Principles of Neural Sci- ence, 4/ e”, McGraw-Hill, New York, 2000. 41 . M. Bouldac and M. D. Levine, “A real-time foveated ... cmos tech- nology”, Tech. Rep., LIRA-TR 6/ 94, 19 94. 63. E. Martinuzzi and F. Pardo, “FG II new version of the ccd retinal sensor frame grabber”, Tech. Rep., LIRA-TR 1/ 94, 19 94...

Ngày tải lên: 10/08/2014, 04:22

... Length (m) )( ˆ mL error )( ˆ mL error 0.93 0.0290 -9 6% 1.0278 10.5% 1. 14 0.128 -8 9.3% 1.061 -7 .0% 1.31 0.1213 -9 0% 1 .41 5 7.9% 1 .46 0.1577 -8 9% 1.50 2.6% 4 Continuous Reinforcement Learning Algorithm for Skills Learning in an ... International Journal of Robotics Research, pages 36 6-3 81, vol. 14, No .4, 1995. 4 Choi A (1997), Real-Time fundamental fre...

Ngày tải lên: 10/08/2014, 04:22

... control [40 ], etc. One of the applications of the reinforcement learning in robotics focuses on behaviors’ learning [41 ] [42 ] and behavior coordination’s learning [43 ] [44 ] [45 ] [46 ]. 4. 5 Continuous ... is parallel to the wall. The reinforcement is maximum when the distance between the robot and the robot coincides with distLim and the robot is parallel to the w...

Ngày tải lên: 10/08/2014, 04:22

... 1 94 G. Unal et al. Fig. 5.15. A walking person (Frames shown L-R-top-bottom) is tracked by the polygonal tracker 180 G. Unal et al. Fig. 5.5. Two rays-swimming video noisy ... prediction using optical flow normal component Fig. 5. 14. A swimming fish in a rocky terrain in the sea (Frames 1, 10, 20, 30, 40 , 70, 110, 143 are shown left-right-top-bottom). Rows 1 and 2: Continu...

Ngày tải lên: 10/08/2014, 04:22