Convolutional Spiking Neural Networks for Spatio-Temporal Feature Extraction (2020-03-27T00:00:00.000000Z)

TL;DR

It is shown that the shallow convolutional SNN outperforms spatio-temporal feature extractor methods such as C3D, ConvLstm, and cascaded Conv and LSTM and is presented a new deep spiking architecture to tackle real-world classification and activity recognition tasks.

Abstract

Spiking neural networks (SNNs) can be used in low-power and embedded systems e.g. neuromorphic chips due to their event-based nature. They preserve conventional artificial neural networks (ANNs) properties with lower computation and memory costs. The temporal coding in layers of convolutional SNNs has not yet been studied. In this paper, we exploit the spatio-temporal feature extraction property of convolutional SNNs. Based on our analysis, we have shown that the shallow convolutional SNN outperforms spatio-temporal feature extractor methods such as C3D, ConvLstm, and cascaded Conv and LSTM. Furthermore, we present a new deep spiking architecture to tackle real-world classification and activity recognition tasks. This model is trained with our proposed hybrid training method. The proposed architecture achieved superior performance compared to other SNN methods on NMNIST (99.6%), DVS-CIFAR10 (69.2%), and DVS-Gesture (96.7%). Also, it achieves comparable results compared to ANN methods on UCF-101 (42.1%) and HMDB-51 (21.5%) datasets.

Authors

A. Nickabadi

3 papers

Alireza Samadzadeh

1 papers

Fatemeh Sadat Tabatabaei Far

1 papers

TL;DR

Abstract

Authors

References69 items

RecDis-SNN: Rectifying Membrane Potential Distribution for Directly Training Spiking Neural Networks

AutoSNN: Towards Energy-Efficient Spiking Neural Networks

Event-based Video Reconstruction via Potential-assisted Spiking Neural Network

Leaky Integrate-and-Fire Spiking Neuron with Learnable Membrane Time Parameter

Comparing SNNs and RNNs on Neuromorphic Vision Datasets: Similarities and Differences

RMP-SNN: Residual Membrane Potential Neuron for Enabling Deeper High-Accuracy and Low-Latency Spiking Neural Network

Effective AER Object Classification Using Segmented Probability-Maximization Learning in Spiking Neural Networks

Temporal Pulses Driven Spiking Neural Network for Fast Object Recognition in Autonomous Driving

Event-driven implementation of deep spiking convolutional neural networks for supervised classification using the SpiNNaker neuromorphic platform

Rethinking the performance comparison between SNNS and ANNS

A Comprehensive Analysis on Adversarial Robustness of Spiking Neural Networks

Speed Invariant Time Surface for Learning to Detect Corner Points With Event-Based Cameras

Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-based optimization to spiking neural networks

A Biologically Plausible Supervised Learning Method for Spiking Neural Networks Using the Symmetric STDP Rule

A Spiking Neural Network Framework for Robust Sound Classification

Direct Training for Spiking Neural Networks: Faster, Larger, Better

Training Deep Spiking Convolutional Neural Networks With STDP-Based Unsupervised Pre-training Followed by Supervised Fine-Tuning

Is Neuromorphic MNIST Neuromorphic? Analyzing the Discriminative Power of Neuromorphic Datasets in the Time Domain

How Does Batch Normalization Help Optimization? (No, It Is Not About Internal Covariate Shift)

Spiking Deep Residual Network

Deep Learning in Spiking Neural Networks

HATS: Histograms of Averaged Time Surfaces for Robust Event-Based Object Classification

Going Deeper in Spiking Neural Networks: VGG and Residual Architectures

Conversion of Continuous-Valued Deep Networks to Efficient Event-Driven Networks for Image Classification

BP-STDP: Approximating Backpropagation using Spike Timing Dependent Plasticity

A Low Power, Fully Event-Based Gesture Recognition System

An Event-Driven Classifier for Spiking Neural Networks Fed with Synthetic or Dynamic Vision Sensor Data

Spatio-Temporal Backpropagation for Training High-Performance Spiking Neural Networks

CIFAR10-DVS: An Event-Stream Dataset for Object Classification

A Survey of Neuromorphic Computing and Neural Networks in Hardware

EVO: A Geometric Approach to Event-Based 6-DOF Parallel Tracking and Mapping in Real Time

STDP-based spiking deep neural networks for object recognition

Real-Time 3D Reconstruction and 6-DoF Tracking with an Event Camera

Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization

Low-latency visual odometry using event-based feature tracks

Event-Based, 6-DOF Camera Tracking from Photometric Depth Maps

Supervised Learning Based on Temporal Coding in Spiking Neural Networks

Using patterns of firing neurons in spiking neural networks for learning and early recognition of spatio-temporal patterns

Backpropagation for Energy-Efficient Neuromorphic Computing

Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks

TrueNorth: Design and Tool Flow of a 65 mW 1 Million Neuron Programmable Neurosynaptic Chip

HFirst: A Temporal Approach to Object Recognition

Converting Static Image Datasets to Spiking Neuromorphic Datasets Using Saccades

Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing

Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting

Spiking Deep Convolutional Neural Networks for Energy-Efficient Object Recognition

Unsupervised Learning of Video Representations using LSTMs

Spiking neural network methodology for modelling, classification and understanding of EEG spatio-temporal data measuring cognitive processes

Learning Spatiotemporal Features with 3D Convolutional Networks

Event-based, 6-DOF pose tracking for high-speed maneuvers

A million spiking-neuron integrated circuit with a scalable communication network and interface

Scalable Hierarchical Network-on-Chip Architecture for Spiking Neural Network Hardware Implementations

Spike-based indirect training of a spiking neural network-controlled virtual insect

UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild

Advancing interconnect density for spiking neural network hardware implementations using traffic-aware adaptive network-on-chip routers

A 45nm CMOS neuromorphic chip with a scalable architecture for learning in networks of spiking neurons

A History of Spike-Timing-Dependent Plasticity

Cognitive processing using spiking neural networks

Spike timing-dependent plasticity: a Hebbian learning rule.

Networks of spiking neurons: the third generation of neural network models

Neuromorphic electronic systems

N-imagenet: towards robust, ﬁne-grained

Space-timeeventcloudsforgesturerecognition:fromrgbcameras

Can spatiotemporal 3d cnns retrace

DiehlPUetal

Hmdb: a large video database

Networks of Spiking Neurons: The Third Generation of Neural Network Models

2022) Spiking transformers

Networks for Spatio-Temporal… 6995

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type