DMCA
Dynamic Cascades with Bidirectional Bootstrapping for Action Unit Detection in Spontaneous Facial Behavior
Cached
Download Links
| Venue: | IEEE TRANSACTIONS ON AFFECTIVE COMPUTING |
| Citations: | 17 - 2 self |
BibTeX
@MISC{Zhu_dynamiccascades,
author = {Yunfeng Zhu and Fernando De La Torre and Jeffrey F. Cohn and Yu-jin Zhang},
title = { Dynamic Cascades with Bidirectional Bootstrapping for Action Unit Detection in Spontaneous Facial Behavior},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Automatic facial action unit detection from video is a long standing problem in facial expression analysis. Research has focused on registration, choice of features, and classifiers. A relatively neglected problem is the choice of training images. Nearly all previous work uses one or the other of two standard approaches. One approach assigns peak frames to the positive class and frames associated with other actions to the negative class. This approach maximizes differences between positive and negative classes, but results in a large imbalance between them, especially for infrequent AUs. The other approach reduces imbalance in class membership by including all target frames from onsets to offsets in the positive class. However, because frames near onsets and offsets often differ little from those that precede them, this approach can dramatically increase false positives. We propose a novel alternative, dynamic cascades with bidirectional bootstrapping (DCBB) to select training samples. Using an iterative approach, DCBB optimally selects positive and negative samples in the training data. Using Cascade Adaboost as basic classifier, DCBB exploits the advantages of feature selection, efficiency, and robustness of Cascade Adaboost. To provide a real-world test, we used the RU-FACS (a.k.a. M3) database of non-posed behavior recorded during interviews. For most tested action units, DCBB improved AU detection relative to alternative approaches.
Keyphrases
dynamic cascade bidirectional bootstrapping spontaneous facial behavior action unit detection negative class positive class cascade adaboost iterative approach automatic facial action unit detection training image training sample non-posed behavior approach reduces imbalance class membership previous work large imbalance real-world test infrequent au training data standard approach au detection relative tested action unit peak frame long standing problem target frame false positive facial expression analysis negative sample neglected problem basic classifier novel alternative alternative approach feature selection
