On parameterizing higher-order motion for behaviour recognition

Abstract

Human behaviours consist different types of motion; we show how they can be disambiguated into their components in a richer way than that currently possible. Studies on optical flow have concentrated on motion alone without the higher order components: snap, jerk and acceleration. We are the first to show how the acceleration, jerk, snap and their constituent parts can be obtained from image sequences, and can be deployed for analysis, especially of behaviour. We demonstrate the estimation of acceleration in sport, human motion, traffic and in scenes of violent behaviour to demonstrate the wide potential for application of analysis of acceleration. Determining higher order components is suited to the analysis of scenes which contain them: higher order motion is innate to scenes containing acts of violent behaviour, but it is not just for behaviour which contains quickly changing movement: human gait contains acceleration though approaches have yet to consider radial and tangential acceleration, since they concentrate on motion alone. The analysis of synthetic and real-world images illustrates the ability of higher order motion to discriminate different objects under different motion. Then the new approaches are applied in heel strike detection in the analysis of human gait. These results demonstrate that the new approach is ready for developing new applications in behaviour recognition and provides a new basis for future research and applications of higher-order motion analysis.

Evaluating Expressway Safety Based on Fuzzy Comprehensive Evaluation with AHP–Entropy Method: A Case Study of Jinliwen Expressway in Zhejiang Province, China

“…In this section, comprehensive weights are computed by combinng the two types of weights, which not only reflect the preference of experts but also the influence of objective indicators. According to the minimum relative entropy [50,51], the weights of each indicator obtained from different methods should be close. Therefore, the objective function is shown in Equation (13).…”

Hydropower Unit State Evaluation Model Based on AHP and Gaussian Threshold Improved Fuzzy Comprehensive Evaluation

“…The comprehensive weight should be as close as possible to the subjective weight and objective weight, but not to any of them. This paper determines the comprehensive weight based on the principle of minimum discrimination information [21]. This principle will select a new distribution to replace the original distribution and ensure that the difference between the new distribution and the original distribution is as small as possible.…”

Enhancing Aerial Human Action Recognition through GAN-boosted ResNeXt Architecture with Squeeze-and-Excitation Network

“…Human Action Recognition (HAR) is a thriving domain of research that is still progressing on account of its versatile applicability in various elds like video surveillance [1], health care [2], human behavior understanding [3], crowd assessment [4], anomaly detection [5] etc. Video surveillance through drones is one of the intensively focused applications of action recognition as drones offer a perfect solution to video surveillance owing to its low cost, mobility and wide area coverage compared to traditional cameras.…”