Learning to Recognize Actions from Limited Training Examples Using a Recurrent Spiking Neural Model

TL;DR

This paper introduces a reservoir-based spiking neural model with a novel encoding method for recognizing actions from limited video data, achieving high accuracy with few training examples.

Contribution

It presents a new encoding inspired by microsaccades and demonstrates effective action recognition with limited data using a reservoir spiking neural network.

Findings

Achieves 81.3% Top-1 accuracy on UCF-101 with only 8 examples per class

Introduces a novel encoding inspired by visual microsaccades

Sets a new benchmark for limited-data action recognition in spiking neural models

Abstract

A fundamental challenge in machine learning today is to build a model that can learn from few examples. Here, we describe a reservoir based spiking neural model for learning to recognize actions with a limited number of labeled videos. First, we propose a novel encoding, inspired by how microsaccades influence visual perception, to extract spike information from raw video data while preserving the temporal correlation across different frames. Using this encoding, we show that the reservoir generalizes its rich dynamical activity toward signature action/movements enabling it to learn from few training examples. We evaluate our approach on the UCF-101 dataset. Our experiments demonstrate that our proposed reservoir achieves 81.3%/87% Top-1/Top-5 accuracy, respectively, on the 101-class data while requiring just 8 video examples per class for training. Our results establish a new benchmark for action recognition from limited video examples for spiking neural models while yielding competetive accuracy with respect to state-of-the-art non-spiking neural models.