Quickshift++: Provably good initializations for sample-based mean Shift

Heinrich Jiang; Jennifer Jang; Samory Kpotufe

Quickshift++: Provably good initializations for sample-based mean Shift

Heinrich Jiang, Jennifer Jang, Samory Kpotufe

Operations Research & Financial Engineering

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

11 Scopus citations

Abstract

We provide initial seedings to the Quick Shift clustering algorithm, which approximate the locally high-density regions of the data. Such seedings act as more stable and expressive cluster-cores than the singleton modes found by Quick Shift. We establish statistical consistency guarantees for this modification. We then show strong clustering performance on real datasets as well as promising applications to image segmentation.

Original language	English (US)
Title of host publication	35th International Conference on Machine Learning, ICML 2018
Editors	Jennifer Dy, Andreas Krause
Publisher	International Machine Learning Society (IMLS)
Pages	3591-3600
Number of pages	10
ISBN (Electronic)	9781510867963
State	Published - 2018
Event	35th International Conference on Machine Learning, ICML 2018 - Stockholm, Sweden Duration: Jul 10 2018 → Jul 15 2018

Publication series

Name	35th International Conference on Machine Learning, ICML 2018
Volume	5

Other

Other	35th International Conference on Machine Learning, ICML 2018
Country/Territory	Sweden
City	Stockholm
Period	7/10/18 → 7/15/18

All Science Journal Classification (ASJC) codes

Computational Theory and Mathematics
Human-Computer Interaction
Software

Cite this

@inproceedings{f7faa8f9d946497188c4c970abff2aee,

title = "Quickshift++: Provably good initializations for sample-based mean Shift",

abstract = "We provide initial seedings to the Quick Shift clustering algorithm, which approximate the locally high-density regions of the data. Such seedings act as more stable and expressive cluster-cores than the singleton modes found by Quick Shift. We establish statistical consistency guarantees for this modification. We then show strong clustering performance on real datasets as well as promising applications to image segmentation.",

author = "Heinrich Jiang and Jennifer Jang and Samory Kpotufe",

year = "2018",

language = "English (US)",

series = "35th International Conference on Machine Learning, ICML 2018",

publisher = "International Machine Learning Society (IMLS)",

pages = "3591--3600",

editor = "Jennifer Dy and Andreas Krause",

booktitle = "35th International Conference on Machine Learning, ICML 2018",

}

Jiang, H, Jang, J & Kpotufe, S 2018, Quickshift++: Provably good initializations for sample-based mean Shift. in J Dy & A Krause (eds), 35th International Conference on Machine Learning, ICML 2018. 35th International Conference on Machine Learning, ICML 2018, vol. 5, International Machine Learning Society (IMLS), pp. 3591-3600, 35th International Conference on Machine Learning, ICML 2018, Stockholm, Sweden, 7/10/18.

Quickshift++: Provably good initializations for sample-based mean Shift. / Jiang, Heinrich; Jang, Jennifer; Kpotufe, Samory.
35th International Conference on Machine Learning, ICML 2018. ed. / Jennifer Dy; Andreas Krause. International Machine Learning Society (IMLS), 2018. p. 3591-3600 (35th International Conference on Machine Learning, ICML 2018; Vol. 5).

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

TY - GEN

T1 - Quickshift++

T2 - 35th International Conference on Machine Learning, ICML 2018

AU - Jiang, Heinrich

AU - Jang, Jennifer

AU - Kpotufe, Samory

PY - 2018

Y1 - 2018

N2 - We provide initial seedings to the Quick Shift clustering algorithm, which approximate the locally high-density regions of the data. Such seedings act as more stable and expressive cluster-cores than the singleton modes found by Quick Shift. We establish statistical consistency guarantees for this modification. We then show strong clustering performance on real datasets as well as promising applications to image segmentation.

AB - We provide initial seedings to the Quick Shift clustering algorithm, which approximate the locally high-density regions of the data. Such seedings act as more stable and expressive cluster-cores than the singleton modes found by Quick Shift. We establish statistical consistency guarantees for this modification. We then show strong clustering performance on real datasets as well as promising applications to image segmentation.

UR - http://www.scopus.com/inward/record.url?scp=85057216302&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057216302&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85057216302

T3 - 35th International Conference on Machine Learning, ICML 2018

SP - 3591

EP - 3600

BT - 35th International Conference on Machine Learning, ICML 2018

A2 - Dy, Jennifer

A2 - Krause, Andreas

PB - International Machine Learning Society (IMLS)

Y2 - 10 July 2018 through 15 July 2018

ER -

Quickshift++: Provably good initializations for sample-based mean Shift

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this