Project-out cascaded regression with an application to face alignment

Cascaded regression approaches have been recently shown to achieve state-of-the-art performance for many computer vision tasks. Beyond its connection to boosting, cascaded regression has been interpreted as a learning-based approach to iterative optimization methods like the Newton’s method. However...

Full description

Bibliographic Details
Main Author: Tzimiropoulos, Georgios
Format: Conference or Workshop Item
Published: 2015
Online Access:https://eprints.nottingham.ac.uk/31442/
_version_ 1848794203541209088
author Tzimiropoulos, Georgios
author_facet Tzimiropoulos, Georgios
author_sort Tzimiropoulos, Georgios
building Nottingham Research Data Repository
collection Online Access
description Cascaded regression approaches have been recently shown to achieve state-of-the-art performance for many computer vision tasks. Beyond its connection to boosting, cascaded regression has been interpreted as a learning-based approach to iterative optimization methods like the Newton’s method. However, in prior work, the connection to optimization theory is limited only in learning a mapping from image features to problem parameters. In this paper, we consider the problem of facial deformable model fitting using cascaded regression and make the following contributions: (a) We propose regression to learn a sequence of averaged Jacobian and Hessian matrices from data, and from them descent directions in a fashion inspired by Gauss-Newton optimization. (b) We show that the optimization problem in hand has structure and devise a learning strategy for a cascaded regression approach that takes the problem structure into account. By doing so, the proposed method learns and employs a sequence of averaged Jacobians and descent directions in a subspace orthogonal to the facial appearance variation; hence, we call it Project-Out Cascaded Regression (PO-CR). (c) Based on the principles of PO-CR, we built a face alignment system that produces remarkably accurate results on the challenging iBUG data set outperforming previously proposed systems by a large margin. Code for our system is available from http://www.cs.nott.ac.uk/yzt/ .
first_indexed 2025-11-14T19:12:28Z
format Conference or Workshop Item
id nottingham-31442
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:12:28Z
publishDate 2015
recordtype eprints
repository_type Digital Repository
spelling nottingham-314422020-05-04T20:12:04Z https://eprints.nottingham.ac.uk/31442/ Project-out cascaded regression with an application to face alignment Tzimiropoulos, Georgios Cascaded regression approaches have been recently shown to achieve state-of-the-art performance for many computer vision tasks. Beyond its connection to boosting, cascaded regression has been interpreted as a learning-based approach to iterative optimization methods like the Newton’s method. However, in prior work, the connection to optimization theory is limited only in learning a mapping from image features to problem parameters. In this paper, we consider the problem of facial deformable model fitting using cascaded regression and make the following contributions: (a) We propose regression to learn a sequence of averaged Jacobian and Hessian matrices from data, and from them descent directions in a fashion inspired by Gauss-Newton optimization. (b) We show that the optimization problem in hand has structure and devise a learning strategy for a cascaded regression approach that takes the problem structure into account. By doing so, the proposed method learns and employs a sequence of averaged Jacobians and descent directions in a subspace orthogonal to the facial appearance variation; hence, we call it Project-Out Cascaded Regression (PO-CR). (c) Based on the principles of PO-CR, we built a face alignment system that produces remarkably accurate results on the challenging iBUG data set outperforming previously proposed systems by a large margin. Code for our system is available from http://www.cs.nott.ac.uk/yzt/ . 2015 Conference or Workshop Item PeerReviewed Tzimiropoulos, Georgios (2015) Project-out cascaded regression with an application to face alignment. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 7-12 June 2015, Boston, USA. http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7298989&filter=AND%28p_Publication_Number:7293313%29
spellingShingle Tzimiropoulos, Georgios
Project-out cascaded regression with an application to face alignment
title Project-out cascaded regression with an application to face alignment
title_full Project-out cascaded regression with an application to face alignment
title_fullStr Project-out cascaded regression with an application to face alignment
title_full_unstemmed Project-out cascaded regression with an application to face alignment
title_short Project-out cascaded regression with an application to face alignment
title_sort project-out cascaded regression with an application to face alignment
url https://eprints.nottingham.ac.uk/31442/
https://eprints.nottingham.ac.uk/31442/