Analog VLSI and neural systems
Analog VLSI and neural systems
What does the retina know about natural scenes?
Neural Computation
Characterization of subthreshold MOS mismatch in transistors for VLSI systems
Analog Integrated Circuits and Signal Processing - Joint special issue on analog VLSI computation
Translinear circuits in subthreshold MOS
Analog Integrated Circuits and Signal Processing - Special issue: translinear circuits
Retinomorphic vision systems: reverse engineering the vertebrate retina
Retinomorphic vision systems: reverse engineering the vertebrate retina
Silicon retina with adaptive filtering properties
NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
The Retinomorphic Approach: Pixel-Parallel Adaptive Amplification,Filtering, and Quantization
Analog Integrated Circuits and Signal Processing
A Throughput-On-Demand Address-Event Transmitter for Neuromorphic Chips
ARVLSI '99 Proceedings of the 20th Anniversary Conference on Advanced Research in VLSI
MICRONEURO '96 Proceedings of the 5th International Conference on Microelectronics for Neural Networks and Fuzzy Systems
Synthetic neural circuits using current-domain signal representations
Neural Computation
A Modular Multi-Chip Neuromorphic Architecture for Real-Time Visual Motion Processing
Analog Integrated Circuits and Signal Processing
Focal-plane analog image processing
CMOS imagers
Modeling Selective Attention Using a Neuromorphic Analog VLSI Device
Neural Computation
IEEE Transactions on Neural Networks
IEEE Transactions on Circuits and Systems Part I: Regular Papers
| Hi-index | 0.00 |
Retinomorphic Chips may improve their spike-coding efficiency by emulating the primate retina's parallel pathways. To this end, I recreated retinal microcircuits in a chip, Visio1, that models the four predominant ganglion-cells, a die size of 9.25 脳 9.67mm虏 in 1.2µm 5V CMOS, and consumes 11.5mW at 5 spikes/seconds/neuron. Visio1 includes novel subthreshold current-mode circuits that use horizontal-cell autofeedback to decouple spatiotemporal bandpass filtering from local gain control and use amacrine-cell loop-gain modulation to adapt highpass and lowpass temporal filtering. Different ganglion cells respond to motion in a stereotyped sequence, making it possible to detect edges of one contrast or the other moving in one direction or the other. I present results from a multichip 2-D motion architecture, which implements Watson and Ahumada's model of human visual-motion sensing.