Identifying nude pictures

Research output: Contribution to conferencePaper

Abstract

This paper demonstrates an automatic system for telling whether there are naked people present in an image. The approach combines color and texture properties to obtain a mask for skin regions, which is shown to be effective for a wide range of shades and colors of skin. These skin regions are then fed to a specialized grouper, which attempts to group a human figure using geometric constraints on human structure. This approach introduces a new view of object recognition, where an object model is an organized collection of grouping hints obtained from a combination of constraints on color and texture and constraints on geometric properties such as the structure of individual parts and the relationships between parts. The system demonstrates excellent performance on a test set of 565 uncontrolled images of naked people, mostly obtained from the Internet, and 4289 assorted control images, drawn from a wide collection of sources.

Original languageEnglish (US)
Pages103-107
Number of pages5
StatePublished - Dec 1 1996
Externally publishedYes
EventProceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96 - Sarasota, FL, USA
Duration: Dec 2 1996Dec 4 1996

Other

OtherProceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96
CitySarasota, FL, USA
Period12/2/9612/4/96

Fingerprint

Skin
Color
Textures
Object recognition
Masks
Internet

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Forsyth, D. A., & Fleck, M. M. (1996). Identifying nude pictures. 103-107. Paper presented at Proceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96, Sarasota, FL, USA, .

Identifying nude pictures. / Forsyth, D. A.; Fleck, M. M.

1996. 103-107 Paper presented at Proceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96, Sarasota, FL, USA, .

Research output: Contribution to conferencePaper

Forsyth, DA & Fleck, MM 1996, 'Identifying nude pictures', Paper presented at Proceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96, Sarasota, FL, USA, 12/2/96 - 12/4/96 pp. 103-107.
Forsyth DA, Fleck MM. Identifying nude pictures. 1996. Paper presented at Proceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96, Sarasota, FL, USA, .
Forsyth, D. A. ; Fleck, M. M. / Identifying nude pictures. Paper presented at Proceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96, Sarasota, FL, USA, .5 p.
@conference{e08802946c9f4fbc9918997ab99a75cb,
title = "Identifying nude pictures",
abstract = "This paper demonstrates an automatic system for telling whether there are naked people present in an image. The approach combines color and texture properties to obtain a mask for skin regions, which is shown to be effective for a wide range of shades and colors of skin. These skin regions are then fed to a specialized grouper, which attempts to group a human figure using geometric constraints on human structure. This approach introduces a new view of object recognition, where an object model is an organized collection of grouping hints obtained from a combination of constraints on color and texture and constraints on geometric properties such as the structure of individual parts and the relationships between parts. The system demonstrates excellent performance on a test set of 565 uncontrolled images of naked people, mostly obtained from the Internet, and 4289 assorted control images, drawn from a wide collection of sources.",
author = "Forsyth, {D. A.} and Fleck, {M. M.}",
year = "1996",
month = "12",
day = "1",
language = "English (US)",
pages = "103--107",
note = "Proceedings of the 1996 3rd IEEE Workshop on Applications of Computer Vision, WACV'96 ; Conference date: 02-12-1996 Through 04-12-1996",

}

TY - CONF

T1 - Identifying nude pictures

AU - Forsyth, D. A.

AU - Fleck, M. M.

PY - 1996/12/1

Y1 - 1996/12/1

N2 - This paper demonstrates an automatic system for telling whether there are naked people present in an image. The approach combines color and texture properties to obtain a mask for skin regions, which is shown to be effective for a wide range of shades and colors of skin. These skin regions are then fed to a specialized grouper, which attempts to group a human figure using geometric constraints on human structure. This approach introduces a new view of object recognition, where an object model is an organized collection of grouping hints obtained from a combination of constraints on color and texture and constraints on geometric properties such as the structure of individual parts and the relationships between parts. The system demonstrates excellent performance on a test set of 565 uncontrolled images of naked people, mostly obtained from the Internet, and 4289 assorted control images, drawn from a wide collection of sources.

AB - This paper demonstrates an automatic system for telling whether there are naked people present in an image. The approach combines color and texture properties to obtain a mask for skin regions, which is shown to be effective for a wide range of shades and colors of skin. These skin regions are then fed to a specialized grouper, which attempts to group a human figure using geometric constraints on human structure. This approach introduces a new view of object recognition, where an object model is an organized collection of grouping hints obtained from a combination of constraints on color and texture and constraints on geometric properties such as the structure of individual parts and the relationships between parts. The system demonstrates excellent performance on a test set of 565 uncontrolled images of naked people, mostly obtained from the Internet, and 4289 assorted control images, drawn from a wide collection of sources.

UR - http://www.scopus.com/inward/record.url?scp=0030379490&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030379490&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030379490

SP - 103

EP - 107

ER -