Learning Robust Objective Functions for Model Fitting in Image Understanding Applications

Wimmer, Matthias; Stulp, Freek; Tschechne, Stephan; Radig, Bernd

doi:10.5244/c.20.118

Cited by 23 publications

(23 citation statements)

References 5 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model is fitted to the face image using objective function approach [4]. After fitting the model to the example face image, we use the projections of the 3D landmarks in 2D for texture mapping.…”

Section: Our Approachmentioning

confidence: 99%

“…In this work, image features are obtained by approximating the pixel values in a region around a pixel of interest The learning algorithm use to map images features to objective values is a k-Nearest-Neighbor classifier (kNN) learned from the data. We used similar methodology developed by Wimmer et al [4] which combines multitude of qualitatively different features [19], determines the most relevant features using machine learning and learns objective functions from annotated images [18]. To extract discriptive features from the image, Michel et al [14] extracted the location of 22 feature points within the face and determine their motion between an image that shows the neutral state of the face and an image that represents a facial expression.…”

Section: Related Workmentioning

confidence: 99%

“…Then, model parameters are estimated by applying the approach of Wimmer et al [4] because it is able to robustly determine model parameters in real-time.…”

Section: Determining High-level Informationmentioning

confidence: 99%

See 2 more Smart Citations

Model Based Analysis of Face Images for Facial Feature Extraction

Riaz

Mayer

Beetz

et al. 2009

Computer Analysis of Images and Patterns

View full text Add to dashboard Cite

Abstract. This paper describes a comprehensive approach to extract a common feature set from the image sequences. We use simple features which are easily extracted from a 3D wireframe model and efficiently used for different applications on a benchmark database. Features verstality is experimented on facial expressions recognition, face reognition and gender classification. We experiment different combinations of the features and find reasonable results with a combined features approach which contain structural, textural and temporal variations. The idea follows in fitting a model to human face images and extracting shape and texture information. We parametrize these extracted information from the image sequences using active appearance model (AAM) approach. We further compute temporal parameters using optical flow to consider local feature variations. Finally we combine these parameters to form a feature vector for all the images in our database. These features are then experimented with binary decision tree (BDT) and Bayesian Network (BN) for classification. We evaluated our results on image sequences of Cohn Kanade Facial Expression Database (CKFED). The proposed system produced very promising recognition rates for our applications with same set of features and classifiers. The system is also realtime capable and automatic.

show abstract

Section: Our Approachmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Model Based Analysis of Face Images for Facial Feature Extraction

Riaz

Mayer

Beetz

et al. 2009

Computer Analysis of Images and Patterns

View full text Add to dashboard Cite

show abstract

“…A recent approach to using local regression models is described by Wimmer et al [18], who train model trees to regress from local haar wavelet features to a objective function designed to peak at the true feature location. At run-time this allows the best matching location to be predicted for each feature.…”

Section: Introductionmentioning

confidence: 99%

Boosted Regression Active Shape Models

Cristinacce

Cootes²

2007

Procedings of the British Machine Vision Conference 2007

137

140

View full text Add to dashboard Cite

We present an efficient method of fitting a set of local feature models to an image within the popular Active Shape Model (ASM) framework [3]. We compare two different types of non-linear boosted feature models trained using GentleBoost [9]. The first type is a conventional feature detector classifier, which learns a discrimination function between the appearance of a feature and the local neighbourhood. The second local model type is a boosted regression predictor which learns the relationship between the local neighbourhood appearance and the displacement from the true feature location. At run-time the second regression model is much more efficient as only the current feature patch needs to be processed. We show that within the local iterative search of the ASM the local feature regression provides improved localisation on two publicly available human face test sets as well as increasing the search speed by a factor of eight.

show abstract

“…We use models which consist of distribution of points in 2D and 3D spaces. A model used in 2D consists of 134 anatomical landmarks corresponding to different face features [4]. Whereas a 3D model is a wireframe model called Candide-III [5].…”

Section: Introductionmentioning

confidence: 99%

Image normalization for face recognition using 3D model

Riaz

Beetz

Radig

2009

2009 International Conference on Information and Communication Technologies

View full text Add to dashboard Cite

Abstract-This paper describes an image segmentation and normalization technique using 3D point distribution model and its counterpart in 2D space. This segmentation is efficient to work for holistic image recognition algorithm. The results have been tested with face recognition application using Cohn Kanade Facial Expressions Database (CKFED). The approach follows by fitting a model to face image and registering it to a standard template. The models consist of distribution of points in 2D and 3D. We extract a set of feature vectors from normalized images using principal components analysis and using them for a binary decision tree for classification. A promising recognition rate of up to 98.75% has been achieved using 3D model and 92.93% using 2D model emphasizing the goodness of our normalization. The experiments have been performed on more than 3500 face images of the database. This algorithm is capable to work in real time in the presence of facial expressions.

show abstract

Learning Robust Objective Functions for Model Fitting in Image Understanding Applications

Cited by 23 publications

References 5 publications

Model Based Analysis of Face Images for Facial Feature Extraction

Model Based Analysis of Face Images for Facial Feature Extraction

Boosted Regression Active Shape Models

Image normalization for face recognition using 3D model

Contact Info

Product

Resources

About