Human Pose Estimation from Monocular Images

Research output: Book/Report › Monograph › Research

Authors

  • Bastian Wandt

Research Organisations

View graph of relations

Details

Original languageGerman
Place of PublicationDüsseldorf
Number of pages112
Publication statusPublished - 2020

Publication series

NameFortschritt-Berichte VDI
PublisherVDI Verlag
Volume869
ISSN (Print)0178-9627

Abstract

Abstract This dissertation deals with the problem of capturing human motions and poses using a single camera. The frst part of the thesis proposes two closely related approaches for the 3D reconstruction of human motions from image sequences. To resolve inherent ambiguities in monocular 3D reconstruction the main idea of this part is to exploit temporal properties of human motions in combination with a human body model learned from training data. The second part of the thesis tackles the problem of reconstructing a human pose from a single image. A human body model is learned by training a deep neural network that covers nonlinearities and anthropometric constraints. C O N T E N T S 1 Introduction ….. 1 1.1 Applications and Commercial Systems . . . . . . . . . . . 1 1.2 Image-based Motion Capture . . . . . . . . . . . . . . . . 2 1.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.3.1 Time Consistent Human Motion Reconstruction . 6 1.3.2 RepNet

Cite this

Human Pose Estimation from Monocular Images. / Wandt, Bastian.
Düsseldorf, 2020. 112 p. (Fortschritt-Berichte VDI; Vol. 869).

Research output: Book/Report › Monograph › Research

Wandt, B 2020, Human Pose Estimation from Monocular Images. Fortschritt-Berichte VDI, vol. 869, Düsseldorf. https://doi.org/10.51202/9783186869104
Wandt, B. (2020). Human Pose Estimation from Monocular Images. (Fortschritt-Berichte VDI; Vol. 869). https://doi.org/10.51202/9783186869104
Wandt B. Human Pose Estimation from Monocular Images. Düsseldorf, 2020. 112 p. (Fortschritt-Berichte VDI). doi: 10.51202/9783186869104
Wandt, Bastian. / Human Pose Estimation from Monocular Images. Düsseldorf, 2020. 112 p. (Fortschritt-Berichte VDI).
Download
@book{41c6305f50d44a9191057135b47b8f5f,
title = "Human Pose Estimation from Monocular Images",
abstract = "Abstract This dissertation deals with the problem of capturing human motions and poses using a single camera. The frst part of the thesis proposes two closely related approaches for the 3D reconstruction of human motions from image sequences. To resolve inherent ambiguities in monocular 3D reconstruction the main idea of this part is to exploit temporal properties of human motions in combination with a human body model learned from training data. The second part of the thesis tackles the problem of reconstructing a human pose from a single image. A human body model is learned by training a deep neural network that covers nonlinearities and anthropometric constraints. C O N T E N T S 1 Introduction ….. 1 1.1 Applications and Commercial Systems . . . . . . . . . . . 1 1.2 Image-based Motion Capture . . . . . . . . . . . . . . . . 2 1.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.3.1 Time Consistent Human Motion Reconstruction . 6 1.3.2 RepNet",
keywords = "3D reconstruction, Universit{\"a}t Hannover, Structure From Motion, TNT, Monocular Cameras, Human Pose Estimation",
author = "Bastian Wandt",
year = "2020",
doi = "10.51202/9783186869104",
language = "Deutsch",
isbn = "9783186869104",
series = "Fortschritt-Berichte VDI",
publisher = "VDI Verlag",

}

Download

TY - BOOK

T1 - Human Pose Estimation from Monocular Images

AU - Wandt, Bastian

PY - 2020

Y1 - 2020

N2 - Abstract This dissertation deals with the problem of capturing human motions and poses using a single camera. The frst part of the thesis proposes two closely related approaches for the 3D reconstruction of human motions from image sequences. To resolve inherent ambiguities in monocular 3D reconstruction the main idea of this part is to exploit temporal properties of human motions in combination with a human body model learned from training data. The second part of the thesis tackles the problem of reconstructing a human pose from a single image. A human body model is learned by training a deep neural network that covers nonlinearities and anthropometric constraints. C O N T E N T S 1 Introduction ….. 1 1.1 Applications and Commercial Systems . . . . . . . . . . . 1 1.2 Image-based Motion Capture . . . . . . . . . . . . . . . . 2 1.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.3.1 Time Consistent Human Motion Reconstruction . 6 1.3.2 RepNet

AB - Abstract This dissertation deals with the problem of capturing human motions and poses using a single camera. The frst part of the thesis proposes two closely related approaches for the 3D reconstruction of human motions from image sequences. To resolve inherent ambiguities in monocular 3D reconstruction the main idea of this part is to exploit temporal properties of human motions in combination with a human body model learned from training data. The second part of the thesis tackles the problem of reconstructing a human pose from a single image. A human body model is learned by training a deep neural network that covers nonlinearities and anthropometric constraints. C O N T E N T S 1 Introduction ….. 1 1.1 Applications and Commercial Systems . . . . . . . . . . . 1 1.2 Image-based Motion Capture . . . . . . . . . . . . . . . . 2 1.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.3.1 Time Consistent Human Motion Reconstruction . 6 1.3.2 RepNet

KW - 3D reconstruction

KW - Universität Hannover

KW - Structure From Motion

KW - TNT

KW - Monocular Cameras

KW - Human Pose Estimation

U2 - 10.51202/9783186869104

DO - 10.51202/9783186869104

M3 - Monografie

SN - 9783186869104

T3 - Fortschritt-Berichte VDI

BT - Human Pose Estimation from Monocular Images

CY - Düsseldorf

ER -