Visual Motion Computation in Analog VLSI Using Pulses
Abstract
The real time computation of motion from real images using a single chip with integrated sensors is a hard prob(cid:173) lem. We present two analog VLSI schemes that use pulse domain neuromorphic circuits to compute motion. Pulses of variable width, rather than graded potentials, represent a natural medium for evaluating temporal relationships. Both algorithms measure speed by timing a moving edge in the image. Our first model is inspired by Reichardt's algorithm in the fiy and yields a non-monotonic response vs. velocity curve. We present data from a chip that implements this model. Our second algorithm yields a monotonic response vs. velocity curve and is currently being translated into silicon.
Cite
Text
Sarpeshkar et al. "Visual Motion Computation in Analog VLSI Using Pulses." Neural Information Processing Systems, 1992.Markdown
[Sarpeshkar et al. "Visual Motion Computation in Analog VLSI Using Pulses." Neural Information Processing Systems, 1992.](https://mlanthology.org/neurips/1992/sarpeshkar1992neurips-visual/)BibTeX
@inproceedings{sarpeshkar1992neurips-visual,
title = {{Visual Motion Computation in Analog VLSI Using Pulses}},
author = {Sarpeshkar, Rahul and Bair, Wyeth and Koch, Christof},
booktitle = {Neural Information Processing Systems},
year = {1992},
pages = {781-788},
url = {https://mlanthology.org/neurips/1992/sarpeshkar1992neurips-visual/}
}