Comparison of action recognition from video and IMUs

被引：2

作者：

Podoprosvetov, A., V ^{[1
]}

Alisejchik, A. P. ^{[2
]}

Orlov, I. A. ^{[1
]}

机构：

[1] Russian Acad Sci, Keldysh Inst Appl Math, Moscow, Russia

[2] Inst Machine Sci, Moscow, Russia

来源：

14TH INTERNATIONAL SYMPOSIUM INTELLIGENT SYSTEMS | 2021年 / 186卷

关键词：

Motion reconstruction; neural networks in control systems; intelligent control system; IMU visualization; IMU sensors;

D O I：

10.1016/j.procs.2021.04.144

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

The aim of the study is to create an intelligent control system for automatic assistance to the operator of a semi-passive exoskeleton. This work solves the problem of recognizing human movements, since it is necessary to build a control system based on human movements in an exoskeleton. An experimental comparative study of two approaches to create human models for action recognition is considered. The first approach is based on the reconstruction of a simplified human kinematic model from IMU data, represented in special format. The second approach is to obtain 2D coordinates of a person's key points from video from two different views. The sets of human models obtained from these approaches are classified using the XGBoost algorithm. The resulting classifiers are compared in terms of accuracy. (C) 2021 The Authors. Published by Elsevier B.V.

引用

页码：242 / 249

页数：8

共 9 条

[1]

Aliseichik A.P., 2020, ALGORITHM CALI UNPUB

[2] Upper Limb Kinematics Using Inertial and Magnetic Sensors: Comparison of Sensor-to-Segment Calibrations [J].

Bouvier, Brice ;

Duprey, Sonia ;

Claudon, Laurent ;

Dumas, Raphael ;

Savescu, Adriana .

SENSORS, 2015, 15 (08) :18813-18833

[3] Adaptive Neural Sliding Mode Control for Singular Semi-Markovian Jump Systems Against Actuator Attacks [J].

Cao, Zhiru ;

Niu, Yugang ;

Zou, Yuanyuan .

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2021, 51 (03) :1523-1533

[4] POSITION AND ORIENTATION IN-SPACE OF BONES DURING MOVEMENT - ANATOMICAL FRAME DEFINITION AND DETERMINATION [J].

CAPPOZZO, A ;

CATANI, F ;

DELLA CROCE, U ;

LEARDINI, A .

CLINICAL BIOMECHANICS, 1995, 10 (04) :171-178

[5] XGBoost: A Scalable Tree Boosting System [J].

Chen, Tianqi ;

Guestrin, Carlos .

KDD'16: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2016, :785-794

[6]

Hidalgo G., 2015, SENSORS-BASEL, P18813

[7]

Luinge H.J., 2019, US, Patent No. [8,165,844, 8165844]

[8] R3DG features: Relative 3D geometry-based skeletal representations for human action recognition [J].

Vemulapalli, Raviteja ;

Arrate, Felipe ;

Chellappa, Rama .

COMPUTER VISION AND IMAGE UNDERSTANDING, 2016, 152 :155-166

[9]

XGBoost, 2019, ITMO U ABSTR

← 1 →