Loading [MathJax]/extensions/tex2jax.js

Human Motion Capture with Sparse Inertial Sensors and Video

Research output: Thesis › Doctoral thesis

Authors

  • Timo von Marcard

Research Organisations

Details

Original languageGerman
QualificationDoctor of Engineering
Awarding Institution
Supervised by
Date of Award16 Oct 2019
Place of PublicationHannover
Publication statusPublished - 2019

Abstract

This thesis explores approaches to capture human motions with a small number of sensors. In the first part of this thesis an approach is presented that reconstructs the body pose from only six inertial sensors. Instead of relying on pre-recorded motion databases, a global optimization problem is solved to maximize the consistency of measurements and model over an entire recording sequence. The second part of this thesis deals with a hybrid approach to fuse visual information from a single hand-held camera with inertial sensor data. First, a discrete optimization problem is solved to automatically associate people detections in the video with inertial sensor data. Then, a global optimization problem is formulated to combine visual and inertial information. The propose approach enables capturing of multiple interacting people and works even if many more people are visible in the camera image. In addition, systematic inertial sensor errors can be compensated, leading to a substantial in

Cite this

Human Motion Capture with Sparse Inertial Sensors and Video. / von Marcard, Timo.
Hannover, 2019. 124 p.

Research output: Thesis › Doctoral thesis

von Marcard, T 2019, 'Human Motion Capture with Sparse Inertial Sensors and Video', Doctor of Engineering, Leibniz University Hannover, Hannover.
von Marcard, T. (2019). Human Motion Capture with Sparse Inertial Sensors and Video. [Doctoral thesis, Leibniz University Hannover].
von Marcard T. Human Motion Capture with Sparse Inertial Sensors and Video. Hannover, 2019. 124 p.
von Marcard, Timo. / Human Motion Capture with Sparse Inertial Sensors and Video. Hannover, 2019. 124 p.
Download
@phdthesis{f821f21d5c834b8cb2e25d5177f9db20,
title = "Human Motion Capture with Sparse Inertial Sensors and Video",
abstract = "This thesis explores approaches to capture human motions with a small number of sensors. In the first part of this thesis an approach is presented that reconstructs the body pose from only six inertial sensors. Instead of relying on pre-recorded motion databases, a global optimization problem is solved to maximize the consistency of measurements and model over an entire recording sequence. The second part of this thesis deals with a hybrid approach to fuse visual information from a single hand-held camera with inertial sensor data. First, a discrete optimization problem is solved to automatically associate people detections in the video with inertial sensor data. Then, a global optimization problem is formulated to combine visual and inertial information. The propose approach enables capturing of multiple interacting people and works even if many more people are visible in the camera image. In addition, systematic inertial sensor errors can be compensated, leading to a substantial in",
keywords = "Sparse Sensors, Modelbased Optimization, video, Inertial Sensors, Non-static Camera, Human Pose Estimation",
author = "{von Marcard}, Timo",
year = "2019",
language = "Deutsch",
school = "Gottfried Wilhelm Leibniz Universit{\"a}t Hannover",

}

Download

TY - BOOK

T1 - Human Motion Capture with Sparse Inertial Sensors and Video

AU - von Marcard, Timo

PY - 2019

Y1 - 2019

N2 - This thesis explores approaches to capture human motions with a small number of sensors. In the first part of this thesis an approach is presented that reconstructs the body pose from only six inertial sensors. Instead of relying on pre-recorded motion databases, a global optimization problem is solved to maximize the consistency of measurements and model over an entire recording sequence. The second part of this thesis deals with a hybrid approach to fuse visual information from a single hand-held camera with inertial sensor data. First, a discrete optimization problem is solved to automatically associate people detections in the video with inertial sensor data. Then, a global optimization problem is formulated to combine visual and inertial information. The propose approach enables capturing of multiple interacting people and works even if many more people are visible in the camera image. In addition, systematic inertial sensor errors can be compensated, leading to a substantial in

AB - This thesis explores approaches to capture human motions with a small number of sensors. In the first part of this thesis an approach is presented that reconstructs the body pose from only six inertial sensors. Instead of relying on pre-recorded motion databases, a global optimization problem is solved to maximize the consistency of measurements and model over an entire recording sequence. The second part of this thesis deals with a hybrid approach to fuse visual information from a single hand-held camera with inertial sensor data. First, a discrete optimization problem is solved to automatically associate people detections in the video with inertial sensor data. Then, a global optimization problem is formulated to combine visual and inertial information. The propose approach enables capturing of multiple interacting people and works even if many more people are visible in the camera image. In addition, systematic inertial sensor errors can be compensated, leading to a substantial in

KW - Sparse Sensors

KW - Modelbased Optimization

KW - video

KW - Inertial Sensors

KW - Non-static Camera

KW - Human Pose Estimation

M3 - Dissertation

CY - Hannover

ER -

By the same author(s)