Postdoctoral Research Associate with Steve Seitz

neeraj@cs.washington.edu

PhD 2011 from Columbia University

Co-advised by Peter Belhumeur & Shree Nayar

Computer Science and Engineering

University of Washington

Box 352350, Paul G. Allen Center 282

Seattle, WA 98195-2350

Research Overview

I work in the areas of computer vision and machine learning, developing techniques that leverage large collections of real-world images for a variety of applications. I am particularly interested in the use of intermediate representations such as parts and attributes, both for improving performance on classical vision tasks such as search and recognition, as well as for creating novel applications such as automatic face replacement and exploration of image collections. Much of my work is directed toward human faces.


Watch a recent talk I gave at UNC-Chapel Hill describing my PhD work.


neeraj headshot

News

April 13, 2012: CVPR 2012 paper on Multi-Attribute Spaces now available
Our CVPR 2012 paper on Multi-Attribute Spaces is now available.
April 13, 2012: DMASM 2012 slides on Leafsnap released
I've added some slides on Leafsnap that I presented at the Digital Media Analysis, Search and Management (DMASM 2012) Workshop in San Diego, a month ago.
November 11, 2011: Two Faces are Better than One poster released
I've added our poster for our IJCB 2011 work on using multiple faces in an image to improve face recognition.
October 02, 2011: Face parts localization slides released
I've added some slides for our CVPR 2011 work on localizing parts of faces.
September 30, 2011: First public release of the Attribute Face Service
I have released the first version of the Attribute Face Service for automatically labeling your own images with our face attribute classifiers.

Curriculum Vitae

CV (pdf)
Citations to my work

Teaching

CSE 590V: Computer vision seminar (Fall 2011)

Selected Publications (see all)

  1. "Multi-Attribute Spaces: Calibration for Attribute Fusion and Similarity Search,"
    Walter Scheirer, Neeraj Kumar, Peter N. Belhumeur, Terrance E. Boult,
    Proceedings of the 25th IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
    June 2012.
  2. "Fusing with Context: a Bayesian Approach to Combining Descriptive Attributes," (oral presentation)
    Walter Scheirer, Neeraj Kumar, Karl Ricanek, Terrance E. Boult, Peter N. Belhumeur,
    Proceedings of the IEEE International Joint Conference on Biometrics (IJCB),
    October 2011.
  3. "Two faces are better than one: Face recognition in group photographs,"
    Ohil K. Manyam, Neeraj Kumar, Peter N. Belhumeur, David J. Kriegman,
    Proceedings of the IEEE International Joint Conference on Biometrics (IJCB),
    October 2011.
  4. "Describable Visual Attributes for Face Verification and Image Search,"
    Neeraj Kumar, Alexander C. Berg, Peter N. Belhumeur, Shree K. Nayar,
    IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI),
    October 2011.
  5. "Describable Visual Attributes for Face Images," (PhD Thesis)
    Neeraj Kumar,
    Technical Report CUCS-035-11, Department of Computer Science, Columbia University,
    August 2011.
  6. "Localizing Parts of Faces Using a Consensus of Exemplars,"
    Peter N. Belhumeur, David W. Jacobs, David J. Kriegman, Neeraj Kumar,
    Proceedings of the 24th IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
    June 2011.
  7. "Attribute and Simile Classifiers for Face Verification," (oral presentation)
    Neeraj Kumar, Alexander C. Berg, Peter N. Belhumeur, Shree K. Nayar,
    Proceedings of the 12th IEEE International Conference on Computer Vision (ICCV),
    October 2009.
  8. "FaceTracer: A Search Engine for Large Collections of Images with Faces,"
    Neeraj Kumar, Peter N. Belhumeur, Shree K. Nayar,
    Proceedings of the 10th European Conference on Computer Vision (ECCV),
    pp. 340--353, October 2008.

Selected Projects (see all)

Localizing Parts of Faces Using a Consensus of Exemplars

Localizing Parts of Faces Using a Consensus of Exemplars

Fall 2009 - Summer 2011

Robustly localizing face parts using local detectors and non-parametric, exemplar-based global models

Keywords: face parts, localization, fiducial points, exemplars, non-parametric, real-world
Leafsnap: An Electronic Field Guide

Leafsnap: An Electronic Field Guide

Summer 2009 - Present

Automatically identifying plant species using photos of leaves taken in the field by mobile devices

Keywords: species identification, electronic field guide, curvature histograms, education, leafsnap, smithsonian, botany
Attribute and Simile Classifiers for Face Verification

Attribute and Simile Classifiers for Face Verification

Spring 2009 - Present

Identifying faces using describable visual attributes of the face, automatically detected using attribute and simile classifiers

Keywords: face verification, attributes, similes, attribute classifiers
FaceTracer: A Search Engine for Large Collections of Images with Faces

FaceTracer: A Search Engine for Large Collections of Images with Faces

Spring 2008 - Present

Searching face images using automatically-trained classifiers of descriptions of face and image attributes

Keywords: face search, attributes, attribute classifiers, content-based image retrieval
Face Swapping: Automatically Replacing Faces in Photographs

Face Swapping: Automatically Replacing Faces in Photographs

Fall 2007 - Fall 2008

Using large collections of images to automatically replace faces in photos

Keywords: face swapping, large image collections, face replacement, deidentification, privacy protection

Software

Attribute Face Service

Attribute Face Service

Fall 2011 - Present

An online web service to apply our attribute classifiers to your own images. Upload images and get automatically-labeled attributes on all faces found. For research purposes only right now.

Keywords: real-world faces, attributes, attribute classifiers
Leafsnap

Leafsnap

Spring 2011 - Present

The Leafsnap homepage, with links to download the iPhone and iPad apps, as well as browse the list of species and their images.

Keywords: species identification, electronic field guide, education, leafsnap, smithsonian, botany
MugHunt Online Face Search Engine

MugHunt Online Face Search Engine

Spring 2010 - Present

An online face search engine built using this research. This search engine is a joint collaboration with Securics, Inc..

Keywords: face search, attributes, attribute classifiers, content-based image retrieval

Databases

Labeled Face Parts in the Wild (LFPW)

Labeled Face Parts in the Wild (LFPW)

Summer 2011

1,432 Images with 29 points labeled on each image

Keywords: real-world faces, labeled face parts, fiducial points
PubFig: Public Figures Face Database

PubFig: Public Figures Face Database

Fall 2009

60,000 face images with several face verification benchmarks & 65 automatically computed attribute labels for 42,000 images

Keywords: real-world faces, public figures, face verification, attributes
FaceTracer Database

FaceTracer Database

Fall 2008

15,000 face images with detailed pose and fiducial information & 5,000 manual attribute labels

Keywords: real-world faces, attributes, attribute classifiers