Clothing
Cosegmentation for Recognizing People
|
|
|
|
It is
extremely difficult even for humans to determine how many different individuals
are shown and which images are of the same individuals from only the faces
(top). However, when the faces are embedded in the context of clothing, it is
much easier to distinguish the three individuals (bottom).
In this paper, we explore the relationship between accurate clothing
segmentation and person recognition. These tasks are inter-related. Improved
clothing segmentation leads to improvements in person recognition and
vice-versa.
|
|
|
|
Abstract
Reseachers have verified that clothing provides
information about the identity of the individual. To extract features from
the clothing, the clothing region first must be localized or segmented in
the image. At the same time, given multiple images of the same person
wearing the same clothing, we expect to improve the effectiveness of
clothing segmentation. Therefore, the identity recognition and clothing
segmentation problems are inter-twined; a good solution for one aides in
the solution for the other.
We build on this idea by analyzing the
mutual information between pixel locations near the face and the identity
of the person to learn a global clothing mask. We segment the clothing
region in each image using graph cuts based on a clothing model learned
from one or multiple images believed to be the same person wearing the
same clothing. We use facial features and clothing features to recognize
individuals in other images. The results show that clothing segmentation
provides a significant improvement in recognition accuracy for large image
collections, and useful clothing masks are simultaneously produced.
A
further significant contribution is that we introduce the Gallagher Collection
Person Dataset where each individual is identified. We hope this dataset
allows the vision community to more easily compare results for tasks
related to recognizing people in consumer image collections.
|
|
|
Citation
A. Gallagher, T. Chen, "Clothing
Cosegmentation for Recognizing People
," IEEE Conference on Computer Vision and
Pattern Recognition 2008.
Bibtex
@inproceedings{gallagher_cvpr_08_clothing,
author = {A. Gallagher and T. Chen},
title = {Clothing Cosegmentation for Recognizing
People},
booktitle = {Proc. CVPR},
year = {2008},
}
|
|
The Gallagher Collection
Person Dataset
We introduce a publically available dataset for researchers
interested in recognizing people in consumer images.
This dataset is for academic research only (non-commercial). I
believe this is the first publically available
dataset of labeled faces in a consumer image collection.
Please cite my paper paper if you use these images.
|