COMPUTATIONAL MODEL BASED ON NEURAL NETWORK OF VISUAL CORTEX FOR HUMAN ACTION RECOGNITION

Abstract

In this paper, we propose a bioinspired model for human action recognition through modeling neural mechanisms of information processing in two visual cortical areas: the primary visual cortex (V1) and the middle temporal cortex (MT) dedicated to motion. This model, named V1-MT, is composed of V1 and MT models (layers) corresponding to their cortical areas, which are built with layered spiking neural networks (SNNs). Some neuron properties in V1 and MT, such as direction and speed selectivity, spatiotemporal inseparability, and center surround suppression, are integrated into SNNs. Based on speed and direction selectivity, V1 and MT models contain multiple SNN channels, each of which processes motion information in sequences with spatiotemporal tunings of neurons at a certain speed and different directions. Therefore, we propose two operations, input signal perceiving with 3-D Gabor filters and surround inhibition processing with 3-D differences of Gaussian functions, to perform this task according to the spatiotemporal inseparability and center surround suppression of neurons. Then, neurons are modeled with our simplified integrate-and fire model and motion information is transformed into spike trains. Afterward, we define a new feature vector: a mean motion map computed from spike trains in all channels to represent human actions. Finally, a support vector machine is trained to classify actions represented by the feature vectors. We conducted extensive experiments on public action databases, and the results show that our model outperforms other bioinspired models and rivals the state-of-the-art approaches.

Existing System

Slow feature analysis (SFA), feedforward hierarchical model (HMAX model)

Proposed System

An action is represented by the mean motion maps built with the mean firing rates of all MT neurons, and recognized with a support vector machine (SVM) classifier. Based on the above analysis, the key contributions of this paper are summarized as follows. 1) We propose a multichannel 3-D SNN architecture based on the direction and speed selectivity of V1 and MT neurons to adapt to speed changes in actions. 2) We integrate neuron properties into the 3-D SNNs, and propose two operations so to effectively detect motion information in video streams for action recognition. 3) We also propose a new feature vector to represent human action. This feature vector is built with the mean motion maps by combining spike trains from all channels.

Architecture

ARCHITECTURE