Graphical Model MAP Inference with Continuous Label Space in Computer Vision

Publikation: Buch/Bericht/Sammelwerk/KonferenzbandMonografieForschungPeer-Review

Autoren

  • Oliver Müller
Forschungs-netzwerk anzeigen

Details

OriginalspracheDeutsch
ErscheinungsortDüsseldorf
Seitenumfang141
Auflage1. Auflage
ISBN (elektronisch)9783186860101
PublikationsstatusVeröffentlicht - 2018

Publikationsreihe

NameInformatik/ Kommunikation
Band860
ISSN (Print)0178-9627

Abstract

This thesis deals with monocular object tracking from video sequences. The goal is to improve tracking of previously unseen non-rigid objects under severe articulations without relying on prior information such as detailed 3D models and without expensive offline training with manual annotations. The proposed framework tracks highly articulated objects by decomposing the target object into small parts and apply online tracking. Drift, which is a fundamental problem of online trackers, is reduced by incorporating image segmentation cues and by using a novel global consistency prior. Joint tracking and segmentation is formulated as a high-order probabilistic graphical model over continuous state variables. A novel inference method is proposed, called S-PBP, combining slice sampling and particle belief propagation. It is shown that slice sampling leads to fast convergence and does not rely on hyper-parameter tuning as opposed to competing approaches based on Metropolis-Hastings or heuristi

Schlagwörter

    monocular object tracking, 3 D models, framework tracks, MAPInferenz, Artikulierte Objektverfolgung, Markov-chain Monte-Carlo, Maschinelles Sehen, Produkt-Slice-Sampling, Visuelle Objektverfolgung, Slice-Sampling, Probabilistisch graphische Modelle, Poseschätzung

Zitieren

Graphical Model MAP Inference with Continuous Label Space in Computer Vision. / Müller, Oliver.
1. Auflage Aufl. Düsseldorf, 2018. 141 S. (Informatik/ Kommunikation; Band 860).

Publikation: Buch/Bericht/Sammelwerk/KonferenzbandMonografieForschungPeer-Review

Müller, O 2018, Graphical Model MAP Inference with Continuous Label Space in Computer Vision. Informatik/ Kommunikation, Bd. 860, 1. Auflage Aufl., Düsseldorf. https://doi.org/10.51202/9783186860101
Müller, O. (2018). Graphical Model MAP Inference with Continuous Label Space in Computer Vision. (1. Auflage Aufl.) (Informatik/ Kommunikation; Band 860). https://doi.org/10.51202/9783186860101
Müller O. Graphical Model MAP Inference with Continuous Label Space in Computer Vision. 1. Auflage Aufl. Düsseldorf, 2018. 141 S. (Informatik/ Kommunikation). doi: 10.51202/9783186860101
Müller, Oliver. / Graphical Model MAP Inference with Continuous Label Space in Computer Vision. 1. Auflage Aufl. Düsseldorf, 2018. 141 S. (Informatik/ Kommunikation).
Download
@book{0ad242b45f1d430e98a0093b1a55bea4,
title = "Graphical Model MAP Inference with Continuous Label Space in Computer Vision",
abstract = "This thesis deals with monocular object tracking from video sequences. The goal is to improve tracking of previously unseen non-rigid objects under severe articulations without relying on prior information such as detailed 3D models and without expensive offline training with manual annotations. The proposed framework tracks highly articulated objects by decomposing the target object into small parts and apply online tracking. Drift, which is a fundamental problem of online trackers, is reduced by incorporating image segmentation cues and by using a novel global consistency prior. Joint tracking and segmentation is formulated as a high-order probabilistic graphical model over continuous state variables. A novel inference method is proposed, called S-PBP, combining slice sampling and particle belief propagation. It is shown that slice sampling leads to fast convergence and does not rely on hyper-parameter tuning as opposed to competing approaches based on Metropolis-Hastings or heuristi",
keywords = "monocular object tracking, 3 D models, framework tracks, MAPInferenz, Artikulierte Objektverfolgung, Markov-chain Monte-Carlo, Maschinelles Sehen, Produkt-Slice-Sampling, Visuelle Objektverfolgung, Slice-Sampling, Probabilistisch graphische Modelle, Posesch{\"a}tzung",
author = "Oliver M{\"u}ller",
year = "2018",
doi = "10.51202/9783186860101",
language = "Deutsch",
isbn = "978-3-18-386010-4",
series = "Informatik/ Kommunikation",
edition = "1. Auflage",

}

Download

TY - BOOK

T1 - Graphical Model MAP Inference with Continuous Label Space in Computer Vision

AU - Müller, Oliver

PY - 2018

Y1 - 2018

N2 - This thesis deals with monocular object tracking from video sequences. The goal is to improve tracking of previously unseen non-rigid objects under severe articulations without relying on prior information such as detailed 3D models and without expensive offline training with manual annotations. The proposed framework tracks highly articulated objects by decomposing the target object into small parts and apply online tracking. Drift, which is a fundamental problem of online trackers, is reduced by incorporating image segmentation cues and by using a novel global consistency prior. Joint tracking and segmentation is formulated as a high-order probabilistic graphical model over continuous state variables. A novel inference method is proposed, called S-PBP, combining slice sampling and particle belief propagation. It is shown that slice sampling leads to fast convergence and does not rely on hyper-parameter tuning as opposed to competing approaches based on Metropolis-Hastings or heuristi

AB - This thesis deals with monocular object tracking from video sequences. The goal is to improve tracking of previously unseen non-rigid objects under severe articulations without relying on prior information such as detailed 3D models and without expensive offline training with manual annotations. The proposed framework tracks highly articulated objects by decomposing the target object into small parts and apply online tracking. Drift, which is a fundamental problem of online trackers, is reduced by incorporating image segmentation cues and by using a novel global consistency prior. Joint tracking and segmentation is formulated as a high-order probabilistic graphical model over continuous state variables. A novel inference method is proposed, called S-PBP, combining slice sampling and particle belief propagation. It is shown that slice sampling leads to fast convergence and does not rely on hyper-parameter tuning as opposed to competing approaches based on Metropolis-Hastings or heuristi

KW - monocular object tracking

KW - 3 D models

KW - framework tracks

KW - MAPInferenz

KW - Artikulierte Objektverfolgung

KW - Markov-chain Monte-Carlo

KW - Maschinelles Sehen

KW - Produkt-Slice-Sampling

KW - Visuelle Objektverfolgung

KW - Slice-Sampling

KW - Probabilistisch graphische Modelle

KW - Poseschätzung

U2 - 10.51202/9783186860101

DO - 10.51202/9783186860101

M3 - Monografie

SN - 978-3-18-386010-4

T3 - Informatik/ Kommunikation

BT - Graphical Model MAP Inference with Continuous Label Space in Computer Vision

CY - Düsseldorf

ER -