Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework

D. Cremers; C. Schnörr; J. Weickert

doi:10.1109/VLSM.2001.938892

Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework

D. Cremers, C. Schnörr, J. Weickert

University of Mannheim

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

65 Scopus citations

Abstract

We present a modification of the Mumford-Shah functional and its cartoon limit which allows the incorporation of statistical shape knowledge in a single energy functional. We show segmentation results on artificial and real-world images with and without prior shape information. In the case of occlusion and strongly cluttered background the shape prior significantly improves segmentation. Finally we compare our results to those obtained by a level-set implementation of geodesic active contours.

Original language	English
Title of host publication	Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	137-144
Number of pages	8
ISBN (Electronic)	076951278X, 9780769512785
DOIs	https://doi.org/10.1109/VLSM.2001.938892
State	Published - 2001
Externally published	Yes
Event	IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001 - Vancouver, Canada Duration: 13 Jul 2001 → …

Publication series

Name	Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001

Conference

Conference	IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001
Country/Territory	Canada
City	Vancouver
Period	13/07/01 → …

Keywords

diffusion
diffusion-snake
geodesic active contours
image segmentation
shape recognition
statistical learning
variational methods

Access to Document

10.1109/VLSM.2001.938892

Cite this

Cremers, D., Schnörr, C., & Weickert, J. (2001). Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework. In Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001 (pp. 137-144). Article 938892 (Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VLSM.2001.938892

Cremers, D. ; Schnörr, C. ; Weickert, J. / Diffusion-snakes : Combining statistical shape knowledge and image information in a variational framework. Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001. Institute of Electrical and Electronics Engineers Inc., 2001. pp. 137-144 (Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001).

@inproceedings{baece93a77f147138f5c7a75b4403195,

title = "Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework",

abstract = "We present a modification of the Mumford-Shah functional and its cartoon limit which allows the incorporation of statistical shape knowledge in a single energy functional. We show segmentation results on artificial and real-world images with and without prior shape information. In the case of occlusion and strongly cluttered background the shape prior significantly improves segmentation. Finally we compare our results to those obtained by a level-set implementation of geodesic active contours.",

keywords = "diffusion, diffusion-snake, geodesic active contours, image segmentation, shape recognition, statistical learning, variational methods",

author = "D. Cremers and C. Schn{\"o}rr and J. Weickert",

note = "Publisher Copyright: {\textcopyright} 2001 IEEE.; IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001 ; Conference date: 13-07-2001",

year = "2001",

doi = "10.1109/VLSM.2001.938892",

language = "English",

series = "Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "137--144",

booktitle = "Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001",

}

Cremers, D, Schnörr, C & Weickert, J 2001, Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework. in Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001., 938892, Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001, Institute of Electrical and Electronics Engineers Inc., pp. 137-144, IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001, Vancouver, Canada, 13/07/01. https://doi.org/10.1109/VLSM.2001.938892

Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework. / Cremers, D.; Schnörr, C.; Weickert, J.
Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001. Institute of Electrical and Electronics Engineers Inc., 2001. p. 137-144 938892 (Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Diffusion-snakes

T2 - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001

AU - Cremers, D.

AU - Schnörr, C.

AU - Weickert, J.

PY - 2001

Y1 - 2001

N2 - We present a modification of the Mumford-Shah functional and its cartoon limit which allows the incorporation of statistical shape knowledge in a single energy functional. We show segmentation results on artificial and real-world images with and without prior shape information. In the case of occlusion and strongly cluttered background the shape prior significantly improves segmentation. Finally we compare our results to those obtained by a level-set implementation of geodesic active contours.

AB - We present a modification of the Mumford-Shah functional and its cartoon limit which allows the incorporation of statistical shape knowledge in a single energy functional. We show segmentation results on artificial and real-world images with and without prior shape information. In the case of occlusion and strongly cluttered background the shape prior significantly improves segmentation. Finally we compare our results to those obtained by a level-set implementation of geodesic active contours.

KW - diffusion

KW - diffusion-snake

KW - geodesic active contours

KW - image segmentation

KW - shape recognition

KW - statistical learning

KW - variational methods

UR - http://www.scopus.com/inward/record.url?scp=84967016941&partnerID=8YFLogxK

U2 - 10.1109/VLSM.2001.938892

DO - 10.1109/VLSM.2001.938892

M3 - Conference contribution

AN - SCOPUS:84967016941

T3 - Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001

SP - 137

EP - 144

BT - Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 13 July 2001

ER -

Cremers D, Schnörr C, Weickert J. Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework. In Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001. Institute of Electrical and Electronics Engineers Inc. 2001. p. 137-144. 938892. (Proceedings - IEEE Workshop on Variational and Level Set Methods in Computer Vision, VLSM 2001). doi: 10.1109/VLSM.2001.938892

Diffusion-snakes: Combining statistical shape knowledge and image information in a variational framework

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this