東華大學圖書館 |

Facial Expression Recognition and Editing with Limited Data.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Facial Expression Recognition and Editing with Limited Data./
作者:	Ding, Hui.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	135 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-04, Section: B.
Contained By:	Dissertations Abstracts International82-04B.
標題:	Electrical engineering. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27998301
ISBN:	9798678188373

Facial Expression Recognition and Editing with Limited Data.
Ding, Hui.

Facial Expression Recognition and Editing with Limited Data. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 135 p.

Source: Dissertations Abstracts International, Volume: 82-04, Section: B.

Thesis (Ph.D.)--University of Maryland, College Park, 2020.

This item must not be sold to any third party vendors.

Over the past five years, methods based on deep features have taken over the computer vision field. While dramatic performance improvements have been achieved for tasks such as face detection and verification, these methods usually need large amounts of annotated data. In practice, not all computer vision tasks have access to large amounts of annotated data. Facial expression analysis is such a task. In this dissertation, we focus on facial expression recognition and editing problems with small datasets. In addition, to cope with challenging conditions like pose and occlusion, we also study unaligned facial attribute detection and occluded expression recognition problems. This dissertation has been divided into four parts. In the first part, we present FaceNet2ExpNet, a novel idea to train a light-weight and high accuracy classification model for expression recognition with small datasets. We first propose a new distribution function to model the high-level neurons of the expression network. Based on this, a two-stage training algorithm is carefully designed. In the pre-training stage, we train the convolutional layers of the expression net, regularized by the face net; In the refining stage, we append fully-connected layers to the pre-trained convolutional layers and train the whole network jointly. Visualization shows that the model trained with our method captures improved high-level expression semantics. Evaluations on four public expression databases demonstrate that our method achieves better results than state-of-the-art. In the second part, we focus on robust facial expression recognition under occlusion and propose a landmark-guided attention branch to find and discard corrupted feature elements from recognition. An attention map is first generated to indicate if a specific facial part is occluded and guide our model to attend to the non-occluded regions. To further increase robustness, we propose a facial region branch to partition the feature maps into non-overlapping facial blocks and enforce each block to predict the expression independently. Depending on the synergistic effect of the two branches, our occlusion adaptive deep network significantly outperforms state-of-the-art methods on two challenging in-the-wild benchmark datasets and three real-world occluded expression datasets. In the third part, we propose a cascade network that simultaneously learns to localize face regions specific to attributes and performs attribute classification without alignment. First, a weakly-supervised face region localization network is designed to automatically detect regions (or parts) specific to attributes. Then multiple part-based networks and a whole-image-based network are separately constructed and combined together by the region switch layer and attribute relation layer for final attribute classification. A multi-net learning method and hint-based model compression are further proposed to get an effective localization model and a compact classification model, respectively. Our approach achieves significantly better performance than state-of-the-art methods on unaligned CelebA dataset, reducing the classification error by 30.9% In the final part of this dissertation, we propose an Expression Generative Adversarial Network (ExprGAN) for photo-realistic facial expression editing with controllable expression intensity. An expression controller module is specially designed to learn an expressive and compact expression code in addition to the encoder-decoder network. This novel architecture enables the expression intensity to be continuously adjusted from low to high. We further show that our ExprGAN can be applied for other tasks, such as expression transfer, image retrieval, and data augmentation for training improved face expression recognition models. To tackle the small size of the training database, an effective incremental learning scheme is proposed. Quantitative and qualitative evaluations on the widely used Oulu-CASIA dataset demonstrate the effectiveness of ExprGAN.

ISBN: 9798678188373Subjects--Topical Terms:

649834
Electrical engineering.
Subjects--Index Terms:

Computer Vision

Facial Expression Recognition and Editing with Limited Data.
LDR:05266nmm a2200385 4500 001 2279566
005 20210823080246.5
008 220723s2020 ||||||||||||||||| ||eng d
020 $a 9798678188373
035 $a (MiAaPQ)AAI27998301
035 $a AAI27998301
040 $a MiAaPQ $c MiAaPQ
100 1 $a Ding, Hui. $3 3558026
245 1 0 $a Facial Expression Recognition and Editing with Limited Data.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 135 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-04, Section: B.
500 $a Advisor: Chellappa, Rama.
502 $a Thesis (Ph.D.)--University of Maryland, College Park, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Over the past five years, methods based on deep features have taken over the computer vision field. While dramatic performance improvements have been achieved for tasks such as face detection and verification, these methods usually need large amounts of annotated data. In practice, not all computer vision tasks have access to large amounts of annotated data. Facial expression analysis is such a task. In this dissertation, we focus on facial expression recognition and editing problems with small datasets. In addition, to cope with challenging conditions like pose and occlusion, we also study unaligned facial attribute detection and occluded expression recognition problems. This dissertation has been divided into four parts. In the first part, we present FaceNet2ExpNet, a novel idea to train a light-weight and high accuracy classification model for expression recognition with small datasets. We first propose a new distribution function to model the high-level neurons of the expression network. Based on this, a two-stage training algorithm is carefully designed. In the pre-training stage, we train the convolutional layers of the expression net, regularized by the face net; In the refining stage, we append fully-connected layers to the pre-trained convolutional layers and train the whole network jointly. Visualization shows that the model trained with our method captures improved high-level expression semantics. Evaluations on four public expression databases demonstrate that our method achieves better results than state-of-the-art. In the second part, we focus on robust facial expression recognition under occlusion and propose a landmark-guided attention branch to find and discard corrupted feature elements from recognition. An attention map is first generated to indicate if a specific facial part is occluded and guide our model to attend to the non-occluded regions. To further increase robustness, we propose a facial region branch to partition the feature maps into non-overlapping facial blocks and enforce each block to predict the expression independently. Depending on the synergistic effect of the two branches, our occlusion adaptive deep network significantly outperforms state-of-the-art methods on two challenging in-the-wild benchmark datasets and three real-world occluded expression datasets. In the third part, we propose a cascade network that simultaneously learns to localize face regions specific to attributes and performs attribute classification without alignment. First, a weakly-supervised face region localization network is designed to automatically detect regions (or parts) specific to attributes. Then multiple part-based networks and a whole-image-based network are separately constructed and combined together by the region switch layer and attribute relation layer for final attribute classification. A multi-net learning method and hint-based model compression are further proposed to get an effective localization model and a compact classification model, respectively. Our approach achieves significantly better performance than state-of-the-art methods on unaligned CelebA dataset, reducing the classification error by 30.9% In the final part of this dissertation, we propose an Expression Generative Adversarial Network (ExprGAN) for photo-realistic facial expression editing with controllable expression intensity. An expression controller module is specially designed to learn an expressive and compact expression code in addition to the encoder-decoder network. This novel architecture enables the expression intensity to be continuously adjusted from low to high. We further show that our ExprGAN can be applied for other tasks, such as expression transfer, image retrieval, and data augmentation for training improved face expression recognition models. To tackle the small size of the training database, an effective incremental learning scheme is proposed. Quantitative and qualitative evaluations on the widely used Oulu-CASIA dataset demonstrate the effectiveness of ExprGAN.
590 $a School code: 0117.
650 4 $a Electrical engineering. $3 649834
650 4 $a Artificial intelligence. $3 516317
650 4 $a Computer science. $3 523869
650 4 $a Deep learning. $3 3554982
653 $a Computer Vision
653 $a Deep Learning
653 $a Facial Attributes
653 $a Facial Expression Editing
653 $a Facial Expression Recognition
653 $a Transfer Learning
690 $a 0544
690 $a 0800
690 $a 0984
710 2 $a University of Maryland, College Park. $b Electrical Engineering. $3 1018746
773 0 $t Dissertations Abstracts International $g 82-04B.
790 $a 0117
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27998301