An efficient neuromorphic analog network for motion estimation

IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications

Volumn 46, Issue 2, 1999, Pages 269-280

  Torralba, Antonio B ^a   Hérault, Jeanny ^a  

^a GRENOBLE INSTITUTE OF TECHNOLOGY   (France)

Author keywords

Aperture problem; Neuromorphic systems; Optical flow estimation; Velocity tuned filters (vtf's); Very large scale integration (vlsi) motion chips

Indexed keywords

ALGORITHMS; MICROPROCESSOR CHIPS; SIGNAL FILTERING AND PREDICTION; VLSI CIRCUITS;

MOTION ESTIMATION; NEUROMORPHIC ANALOG NETWORKS;

CELLULAR NEURAL NETWORKS;

EID: 0033077779     PISSN: 10577122     EISSN: None     Source Type: Journal    
DOI: 10.1109/81.747199     Document Type: Article

References (27)

1. E. H. Adelson and J. R. BergenSpatiotemporal energy models for the perception of motionJ. Opt. Soc. Amer. A, Opt. Image Sei., vol. 2, no. 2, pp. 284-299, Feb. 1985.
2. J. Atick and A. RedlichWhat does the retina know about natural scenes?Neural Comput., vol. 4, pp. 196-210, 1992.
3. J. L. Barron, D. J. Fleet, and S.S. BeaucheminPerformances of optical flow techniquesInt. J. Comput. Vision, vol. 12, no. 1, pp. 43-77, 1994.
4. W. Beaudot, P. Palagi, and J. HéraultRealistic simulation tool for early visual processing including space, time and color dataLecture Notes in Computer Sciences, 686. Berlin, Germany: Springer-Verlag, 1993.
5. T. DelbrückSilicon retina with correlation-based, velocity-tuned pixelsIEEE Trans. Neural Networks, vol. 4, pp. 529-541, 1993.
6. R. A. Deutschmann and C. KochCompact real-time 2-D gradientbased analog VLSI motion sensorProc. Int. Conf. Advanced Focal Plane Arrays Electronic Cameras, 1998.
7. R. A. Deutschmann, C. M. Higgins, and C. Koch, Real-Time analog VLSI sensors for 2-D direction of motion in Proc. Int. Conf. Artificial Neural Networks (ICANN 97), Lausanne, Switzerland, 1997, vol. 1327 of Lecture Notes in Computer Science, Springer-Verlag, pp. 1163-1168.
8. D. J. HeegerModel for the extraction of image flowJ. Opt. Soc. Amer. A, Opt. Image Sei., vol. 4, no. 8, pp. 1455-1471, Aug. 1987.
9. C. M. Higgins and C. KochAnalog CMOS velocity sensorsProc. Electronic Imaging '97, San Jose, CA, Feb. 1997.
10. J. HéraultA model of color processing in the retina of vertebrates: From photoreceptors to color opposition and color constancy phenomenaNeurocomputing, vol. 12, no. 2-3, pp. 113-129, 1996.
11. C. Koch, J. Luo, and C. MeadComputing motion using analog binary resistive networksComputer, 52-63, Mar. 1988.
12. J. KramerCompact integrated motion sensor with three-pixel interactionIEEE Trans. Pattern Anal. Machine Intell., vol. 18, pp. 455-160, Apr. 1996.
13. C. Mead, Analog VLSI and Neural Systems. Reading, MA: AddisonWesley, 1989.
14. C. A. Mead and M. A. MahowaldA silicon model of early visual processingNeural Networks, vol. 1, pp. 91-97, 1988.
15. A. Mhani, G. Sicard, and G. BouvierAnalog vision chip for sensing edges contrasts and motionProc. ISCAS, 1997, pp. 326-329.
16. B. NabetElectronic hardware for vision modelingNonlinear Vision: Determination of Neural Receptive Fields, Function and Networks, R. B. Pinter and B. Nabet, Eds. Boca Raton, FL: CRC, 1992, ch. 18, pp. 463-74.
17. T. Poggio, V. Torre, and C. KochComputational vision and regularization theoryNature, vol. 317, no. 26, pp. 314-319, Sept. 1985.
18. R. Sarpeshkar, J. Kramer, G. Indiveri, and C. KochAnalog VLSI architectures for motion processing: From fundamentals limits to system applicationsProc. IEEE, vol. 84, pp. 969-987, July 1996.
19. R. Sarpeshkar, R. F. Lyon, and C. MeadA low-power wide-linearrange transconductance amplifierAnalog Integrated Circuits Signal Processing, vol. 13, nos. 1/2, 1997.
20. B. E. ShiGabor-type filtering in space and time with cellular neural networksIEEE Trans. Circuits Syst. I, vol. 45, pp. 121-132, Feb. 1998.
21. B. E. Shi, T. Roska, and L. O. ChuaDesign of linear cellular neural networks for motion sensitive filteringIEEE Trans. Circuits Syst. II, vol. 40, pp. 320-331, May 1993.
22. _Hyperacuity in cellular neural networks and the measurement of optical flowInt. J. Circuit Theory Appl., vol. 26, pp. 343-364, 1998.
23. A. Spinel, D. Pellerin, and J. HéraultSpatiotemporal energy-based method for velocity estimationSignal Processing, vol. 65, no. 3, pp. 347-362, 1998.
24. J. Tanner and C. MeadAn integrated analog optical motion sensorVLSI Signal Processing, R. W. Brodersen and H. S. Moscovitz, Eds. New York: IEEE, 1988, 2, pp. 59-87.
25. A. Torralba and J. HéraultFrom retinal circuits to motion processing: A neuromorphic approach to velocity estimationProc. ESANN'97, Brussels, Belgium, Apr. 1997, pp. 47-54.
26. _Minimal complexity velocity-tuned filters with analogue neuromorphic networks: A theoretical approach for efficient designNeural Processing Lett., vol. 8, no. 3, pp. 1-11, 1998.
27. A. B. Watson and A. J. AhumadaA look at motion in the frequency domainMotion: Perception and Representation, J. K. Tsotsos, Ed. New York: Association for Computing Machinery, 1983, pp. 1-10.