Pools of AAMs: Towards Automatically Fitting any Face Image

Peyras, Julien; Bartoli, Adrien; Khoualed, Samir

doi:10.5244/c.22.44

Cited by 5 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An example is proposed by Kanaujia et al2006 [6] where multiple 2D pose-specific Active Shape Models (ASMs) are coupled with a switching mechanism using SIFT descriptors. Peyras et al2008 [10] used a pool of Active Appearance Models (AAMs) that were specialized at different poses and expressions but not robust to changes in illumination. However, in order to train a reliable ASM or AAM model, a large amount of training data in the form of labeled shapes was needed.…”

Section: Introductionmentioning

confidence: 99%

Online learning of robust facial feature trackers

Sheerman-Chase

Ong

Bowden

2009

2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops

View full text Add to dashboard Cite

This paper presents a head pose and facial feature estimation technique that works over a wide range of pose variations without a priori knowledge of the appearance of the face. Using simple LK trackers, head pose is estimated by Levenberg-Marquardt (LM) pose estimation using the feature tracking as constraints. Factored sampling and RANSAC are employed to both provide a robust pose estimate and identify tracker drift by constraining outliers in the estimation process. The system provides both a head pose estimate and the position of facial features and is capable of tracking over a wide range of head poses

show abstract

Section: Introductionmentioning

confidence: 99%

Online learning of robust facial feature trackers

Sheerman-Chase

Ong

Bowden

2009

2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops

View full text Add to dashboard Cite

show abstract

“…Cootes et al [105] proposed an active appearance model for matching statistical models of appearance to images, by employing interactive algorithms. Peyras et al [106] presented a method of fitting active appearance models for unseen faces. The method allows variations in poses and expressions solved by active appearance models.…”

Section: D Face Modelingmentioning

confidence: 99%

Human emotional state recognition using 3D facial expression features

Tie¹

2021

Preprint

View full text Add to dashboard Cite

In recent years there has been a growing interest in improving all aspects of the interaction between human and computers. Emotion recognition is a new research direction in human-computer interaction (HCI) which is based on affective computing that is expected to significantly improve the quality of HCI system and communications. Most existing works address this problem using 2D features, but they are sensitive to head pose, clutter, and variations in lighting conditions. In light of such problems, two 3D visual feature based approaches are presented in this dissertation. First, we present a recognition method based on the Gabor library for real 3D visual features extraction and an improved kernel canonical correlation analysis (IKCCA) algorithm for emotion classification. Second, to reduce the computation cost and provide a more general approach, we propose using a fiducial points' controlled 3D face model to recognize human emotion from video sequences. An Elastic body spine (EBS) technique is applied for deformation feature extraction and a discriminative Isomap (D-Isomap) based classification is used for the final decision. The most significant contributions of this work are detecting and tracking fiducial points automatically from video sequences to construct a generic 3D face model, and the introduction of EBS deformation features for emotion recognition. The experimental results show the robustness and effectiveness of the proposed methods.

show abstract

“…Previous approaches can be divided into three categories: view (2D) based, 3D based and combined 2D+3D based. View based methods (Cootes et al, 2000), (Zhou et al, 2005), (Faggian et al, 2005), (Peyras et al, 2008), train a set of 2D models, each of which is designed to cope with shape or texture variation within a small range of viewpoints. We have found for some applications that switching between 2D views can cause notable artifacts (e.g.…”

Section: Related Workmentioning

confidence: 99%

Face Detection and Tracking With 3d Pga CLM

Meng¹,

Tiddeman²

2010

Proceedings of the International Conference on Computer Vision Theory and Applications

View full text Add to dashboard Cite

In this paper we describe a system for facial feature detection and tracking using a 3D extension of the Constrained Local Model (CLM) (Cristinacce and Cootes, 2006) algorithm. The use of a 3D shape model allows improved tracking through large head rotations. CLM uses a shape and texture appearance model to generate a set of region template detectors. A search is then performed in the global pose / shape space using these detectors. The proposed extension uses multiple appearance models from different viewpoints and a single 3D shape model built using Principal Geodesic Analysis (PGA) (Fletcher et al., 2004) instead of direct Principal Components Analysis (PCA). During fitting or tracking the current estimate of pose is used to select the appropriate appearance model. We demonstrate our results by fitting the model to image sequences with large head rotations. The results show that the proposed multi-view 3D CLM algorithm using PGA improves the performance of the algorithm using PCA for tracking faces in videos with large out-of-plane head rotations.Peer reviewe

show abstract

Pools of AAMs: Towards Automatically Fitting any Face Image

Cited by 5 publications

References 10 publications

Online learning of robust facial feature trackers

Online learning of robust facial feature trackers

Human emotional state recognition using 3D facial expression features

Face Detection and Tracking With 3d Pga CLM

Contact Info

Product

Resources

About