English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Spiking Neural Networks (SNNs) are bio-inspired networks that processinformation conveyed as temporal spikes rather than numeric values. A spikingneuron of an SNN only produces a spike whenever a significant number of spikesoccur within a short period of time. Due to their spike-based computationalmodel, SNNs can process output from event-based, asynchronous sensors withoutany pre-processing at extremely lower power unlike standard artificial neuralnetworks. This is possible due to specialized neuromorphic hardware thatimplements the highly-parallelizable concept of SNNs in silicon. Yet, SNNs havenot enjoyed the same rise of popularity as artificial neural networks. This notonly stems from the fact that their input format is rather unconventional butalso due to the challenges in training spiking networks. Despite their temporalnature and recent algorithmic advances, they have been mostly evaluated onclassification problems. We propose, for the first time, a temporal regressionproblem of numerical values given events from an event camera. We specificallyinvestigate the prediction of the 3-DOF angular velocity of a rotating eventcamera with an SNN. The difficulty of this problem arises from the predictionof angular velocities continuously in time directly from irregular,asynchronous event-based input. Directly utilising the output of event cameraswithout any pre-processing ensures that we inherit all the benefits that theyprovide over conventional cameras. That is high-temporal resolution,high-dynamic range and no motion blur. To assess the performance of SNNs onthis task, we introduce a synthetic event camera dataset generated fromreal-world panoramic images and show that we can successfully train an SNN toperform angular velocity regression.

Event-Based Angular Velocity Regression with Spiking Networks