Image analysis – Applications – Personnel identification
Reexamination Certificate
2007-06-26
2007-06-26
Mancuso, Joseph (Department: 2624)
Image analysis
Applications
Personnel identification
C382S159000, C382S173000, C358S403000
Reexamination Certificate
active
11095984
ABSTRACT:
A method for human face detection that detects faces independently of their particular poses and simultaneously estimates those poses. Our method exhibits an immunity to variations in skin color, eyeglasses, facial hair, lighting, scale and facial expressions, and others. In operation, we train a convolutional neural network to map face images to points on a face manifold, and non-face images to points far away from that manifold, wherein that manifold is parameterized by facial pose. Conceptually, we view a pose parameter as a latent variable, which may be inferred through an energy-minimization process. To train systems based upon our inventive method, we derive a new type of discriminative loss function that is tailored to such detection tasks. Our method enables a multi-view detector that can detect faces in a variety of poses, for example, looking left or right (yaw axis), up or down (pitch axis), or tilting left or right (roll axis). Systems employing our method are highly-reliable, run at near real time (5 frames per second on conventional hardware), and is robust against variations in yaw (±90°), roll(±45°), and pitch(±60°).
REFERENCES:
patent: 5649068 (1997-07-01), Boser et al.
patent: 6959109 (2005-10-01), Moustafa
patent: 2004/0264745 (2004-12-01), Gu et al.
patent: 2005/0180626 (2005-08-01), Moon et al.
patent: 2006/0034495 (2006-02-01), Miller et al.
patent: 2006/0078172 (2006-04-01), Zhang et al.
Y.Li, S.Gong, and H. Liddell in an article entitled “Support vector regression and classification based on multi-view face detection and recognition”, which was presented at FG2000.
Estimating Facial Pose from a sparse representation ( Hankyu Moon and Matt Miller NEC Laboratories America; 2004 International conference on image processing ; ICP).
Face Alignment Under Various Poses and Expressions; Shengjun Xin and Haizhou Ai; Computer Science and Technology Department, Tsinghua University, Beijing 100084, China; ahz@mail.tsinghua.edu.cn.
Learning a Sparse representation from multiple still images for on-line face recognition in an unconstrained environment, Shouten et al; IEEE 2006.
View Based and modular eigenspaces for face recognition; CVPR, 1994; A pentland B. Moghaddam and T. Starner.
Fast Multi View Face detection; M.Jones and P. Viola ; 2003-96, Mitsubishi Electric Research Laboratories, 2003.
LeCun Yann
Miller Matthew L.
Osadchy Margarita
Bitar Nancy
Brosemer, Kolefas & Associates LLC
Mancuso Joseph
NEC Laboratories America, Inc.
LandOfFree
Synergistic face detection and pose estimation with... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Synergistic face detection and pose estimation with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Synergistic face detection and pose estimation with... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3826156