An Event-Driven Multi-Kernel Convolution Processor Module for Event-Driven Vision Sensors

Camuñas-Mesa, Luis A.; Zamarreno-Ramos, C.; Linares-Barranco, A.; Acosta, Antonio; Serrano-Gotarredona, Teresa; Linares-Barranco, B.

doi:10.1109/jssc.2011.2167409

Cited by 102 publications

(100 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this prototype the visual processing (visual stimulus orientation detection) was performed by real-time address-event processing software (57). In future versions this processing could also be performed in neuromorphic hardware by using simple-cell orientation selectivity hardware models (58,59) or event-driven convolution chips (60).…”

Section: Discussionmentioning

confidence: 99%

Synthesizing cognition in neuromorphic electronic systems

Neftci

Binas

Rutishauser

et al. 2013

Proc. Natl. Acad. Sci. U.S.A.

130

122

View full text Add to dashboard Cite

The quest to implement intelligent processing in electronic neuromorphic systems lacks methods for achieving reliable behavioral dynamics on substrates of inherently imprecise and noisy neurons. Here we report a solution to this problem that involves first mapping an unreliable hardware layer of spiking silicon neurons into an abstract computational layer composed of generic reliable subnetworks of model neurons and then composing the target behavioral dynamics as a "soft state machine" running on these reliable subnets. In the first step, the neural networks of the abstract layer are realized on the hardware substrate by mapping the neuron circuit bias voltages to the model parameters. This mapping is obtained by an automatic method in which the electronic circuit biases are calibrated against the model parameters by a series of population activity measurements. The abstract computational layer is formed by configuring neural networks as generic soft winner-take-all subnetworks that provide reliable processing by virtue of their active gain, signal restoration, and multistability. The necessary states and transitions of the desired high-level behavior are then easily embedded in the computational layer by introducing only sparse connections between some neurons of the various subnets. We demonstrate this synthesis method for a neuromorphic sensory agent that performs real-time context-dependent classification of motion patterns observed by a silicon retina.decision making | sensorimotor | working memory | analog very large-scale integration | artificial neural systems U nlike digital simulations, in which the dynamics of neuronal models are encoded and calculated on general purpose digital hardware, "neuromorphic" emulations express the dynamics of the neural systems directly on an analogous physical substrate (1). Digital simulations have the advantage that they can be exactly and reliably programmed using numerical operations of very high precision. However, they suffer the disadvantage that they are cast on abstract binary electronic circuits whose operation is entirely divorced from the physical processes being simulated. Consequently, such simulations do not readily advance our understanding of how biological neural systems are able to attain their extraordinary physical performance, using only large numbers of apparently unreliable, slow, and imprecise neural components, a problem first recognized by von Neumann more than a half century ago (2). Furthermore, the reliability of digital systems comes at high cost of the circuit complexity necessary for the orchestration of communication and processing, an overhead that declares itself also in the costs of system construction and power dissipation (3).The alternative, neuromorphic, approach to information processing strives to capture in complementary metal-oxide semiconductor (CMOS) very large-scale integration (VLSI) electronic technology the more distributed, asynchronous, and limited precision nature of biological intelligent systems (1, 4).†, ‡ Research...

show abstract

Section: Discussionmentioning

confidence: 99%

Synthesizing cognition in neuromorphic electronic systems

Neftci

Binas

Rutishauser

et al. 2013

Proc. Natl. Acad. Sci. U.S.A.

130

122

View full text Add to dashboard Cite

show abstract

“…Alternatively, serial AER schemes have also been proposed where a differential microstrip communicates (x, y, p) events bit-serially and asynchronously [71][72][73]. The availability of ED sensing and processing chips has allowed the implementation of first ED sensory systems [50,51,55] that show the unique pseudosimultaneity property, where the input and output event flows of a processing stage are (in practice) simultaneous or coincident in time. This is illustrated in Fig.…”

Section: Spiking Neural Network For Event-driven Sensing and Processingmentioning

confidence: 99%

“…12.1e3 shows the situation for an ED implementation. An ED processor module processes events as they flow in, with a delay typically in the 100 ns range per event [51]. The system does not need to wait for collecting image frames, but output events are emitted while the input events are processed as soon as enough input events are received, as is in cortical circuits.…”

Section: Spiking Neural Network For Event-driven Sensing and Processingmentioning

confidence: 99%

Spike-Timing-Dependent-Plasticity in Hybrid Memristive-CMOS Spiking Neuromorphic Systems

Serrano-Gotarredona

Linares-Barranco

2013

Memristors and Memristive Systems

Self Cite

View full text Add to dashboard Cite

“…They addressed the need in some applications of higher speed and sensitivity by realizing that the best improvement in performance results from adding more gain and bandwidth to the photoreceptor that precedes the differencing amplifier. They have taken two approaches to this improvement but only the first is published [19]. In their pixel, they interposed two non-inverting voltage gain amplifiers between the logarithmic photoreceptor and the capacitive differencing amplifier.…”

Section: Faster and More Sensitive Dvs Pixelsmentioning

confidence: 99%

“…Instead of supplying photocurrent from the source of an nfet with feedback to the gate of the nfet, they use the photoreceptor from Oliver Landolt [19], where the feedback photocurrent is supplied from the drain of a pfet, with feedback applied to the source of the pfet. The gate of the pfet is tied to a fixed voltage, which determines the clamped photodiode voltage.…”

Section: Faster and More Sensitive Dvs Pixelsmentioning

confidence: 99%

Fun with Asynchronous Vision Sensors and Processing

Delbrück

2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

View full text Add to dashboard Cite

Abstract. This paper provides a personal perspective on our group's efforts in building event-based vision sensors, algorithms, and applications over the period [2002][2003][2004][2005][2006][2007][2008][2009][2010][2011][2012]. Some recent advances from other groups are also briefly described.When Mahowald and Mead built the first silicon retina with asynchronous digital output around 1992 [1], conventional CMOS active pixel sensors (APS) were still research chips. It required the investment by industry of about a billion dollars to bring CMOS APS to high volume production. So it is no surprise that while the imager community has been consumed by the megapixel race to make nice photos, cameras that mimic more closely how the eye works have taken a long time to come to a useful form. These "silicon retinas" are much more complex at the pixel level than APS cameras and they pay the price in terms of fill factor and pixel size; machine vision cameras with capability of synchronous global electronic shutter are about 5um. Silicon retina pixels are roughly 10 times the area of a machine vision camera pixel. So why are silicon retinas still interesting? Mostly because of the high cost at the system level of processing the highly redundant data from conventional cameras, and the fixed latencies imposed by the frame intervals. High performance activity driven event-based sensors could greatly benefit applications in real time robotics, where just as in nature, latency and power are very important [2,5,9,10].

show abstract

An Event-Driven Multi-Kernel Convolution Processor Module for Event-Driven Vision Sensors

Cited by 102 publications

References 41 publications

Synthesizing cognition in neuromorphic electronic systems

Synthesizing cognition in neuromorphic electronic systems

Spike-Timing-Dependent-Plasticity in Hybrid Memristive-CMOS Spiking Neuromorphic Systems

Fun with Asynchronous Vision Sensors and Processing

Contact Info

Product

Resources

About