RI | Research | Projects | Neural Network-Based Face Detection

Text only version of this site

Neural Network-Based Face Detection This project is no longer active. Head: Takeo Kanade Contact: Takeo Kanade (tk@cs.cmu.edu) Mailing address: Carnegie Mellon University Robotics Institute 5000 Forbes Avenue Pittsburgh, PA 15213 Associated center: VASC Associated lab/group: Face Group For more information, see this project's homepage.

Jump to: Project Description | Personnel | Publications

Project Description

A retinally connected neural network examines small windows of an image, and decides whether each window contains a face. The system arbitrates between multiple networks to improve performance over a single network. We use a bootstrap algorithm for training the networks, which adds false detections into the training set as training progresses. This eliminates the difficult task of manually selecting non-face training examples, which must be chosen to span the entire space of non-face images. Comparisons with other state-of-the-art face detection systems are presented; our system has better performance in terms of detection and false-positive rates.

Past members

	Name	Title	Email Address
	Shumeet Baluja	Adjunct Faculty (Adjunct)	baluja@cs.cmu.edu
	Takeo Kanade	U.A. and Helen Whitaker University Prof., RI/CS	tk@cs.cmu.edu
	Henry Rowley	PhD Student, CS

Recent publications [View all 11 publications]

• Applying Machine Learning for High Performance Named-Entity Extraction
  S. Baluja, V. Mittal, and R. Sukthankar
  Pacific Association for Computtational Linguistics, 1999.
  Download: pdf [329 KB], ps.gz [252 KB] copyrighted

• Making Templates Rotationally Invariant: An Application to Rotated Digit Recognition
  S. Baluja
  Advances in Neural Information Processing Systems NIPS '99, 1999.
  Download: pdf [537 KB], ps.gz [747 KB] copyrighted

• Rotation Invariant Neural Network-Based Face Detection
  H. Rowley, S. Baluja, and T. Kanade
  Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, June, 1998. [Abstract]
  Download: pdf [1478 KB], ps.gz [6370 KB] copyrighted

• Neural Network-Based Face Detection
  H. Rowley, S. Baluja, and T. Kanade
  IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 20, No. 1, January, 1998, pp. 23-38. [Abstract]
  Download: pdf [1411 KB], ps.gz [4621 KB] copyrighted

• Probabilistic Modeling for Face Orientation Discrimination: Learning from Labeled and Unlabeled Data
  S. Baluja
  Neural Information Procesing systems (NIPS '98), 1998.
  Download: pdf [82 KB], ps.gz [154 KB] copyrighted

• Rotation Invariant Neural Network-Based Face Detection
  H. Rowley, S. Baluja, and T. Kanade
  tech. report CMU-CS-97-201, Computer Science Department, Carnegie Mellon University, December, 1997. [Abstract]
  Download: pdf [1574 KB], ps.gz [6372 KB] copyrighted

• Genetic Algorithms and Explicit Search Statistics
  S. Baluja
  Advances in Neural ndformation Processing Systems (NIPS '96), December, 1996.
  Download: pdf [29 KB], ps.gz [45 KB] copyrighted

• Neural Network-Based Face Detection
  H. Rowley, S. Baluja, and T. Kanade
  Computer Vision and Pattern Recognition '96, June, 1996. [Abstract]
  Download: pdf [452 KB], ps.gz [1535 KB] copyrighted

• Neural Network-Based Face Detection
  H. Rowley, S. Baluja, and T. Kanade
  Proceedings of the 1996 DARPA Image Understanding Workshop, February, 1996. [Abstract]
  Download: pdf [773 KB], ps.gz [2475 KB], ps.z [2660 KB] copyrighted

• Human Face Detection in Visual Scenes
  H. Rowley, S. Baluja, and T. Kanade
  Advances in Neural Information Processing Systems 8, 1996, pp. 875 - 881. [Abstract]
  Download: pdf [698 KB], ps.gz [2434 KB] copyrighted