A graphical model based solution to the facial feature point tracking problem

Coşar, Serhan; Çetin, Müjdat

doi:10.1016/j.imavis.2010.12.001

Cited by 10 publications

(7 citation statements)

References 24 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the ratio of distance between co-linear points is fixed under affine transformations, line segments between facial points are also used to model the face shape such as [36]. Liang et al [37] use a condensation algorithm modified with spatial constraints.…”

Section: B Facial Point Detectionmentioning

confidence: 99%

Privileged Information-Based Conditional Structured Output Regression Forest for Facial Point Detection

Yang

Patras

2015

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

This paper introduces a regression method, called Privileged Information-based Conditional Structured Output Regression Forest (PI-CSORF) for facial point detection. In order to train Regression Forest more efficiently, the method utilizes both privileged information, that is available only during training such as head pose or gender, and shape constraints on the location of the facial points. We propose to select the test functions at some randomly chosen internal tree nodes according to the information gain calculated on the privileged information. In this way the training patches arrive at leaves tend to have low variance both in terms of their displacements in relation to the facial points and in terms of the privileged information. At each leaf node, we learn three models: first, a probabilistic model of the pdf of the privileged information; second, a probabilistic regression model for the locations of the facial points; and third, shape models that model the interdependencies of the locations of neighbouring facial points in a predefined structure graph. Both of the latter two are conditioned on the privileged information. During testing, the marginal probability of the privileged information is estimated and the facial point locations are localized using the appropriate conditional regression and shape models. The proposed method is validated and compared with very recent methods, especially that use Regression Forests, on datasets recorded in controlled and uncontrolled environments, namely the BioID, the Labelled Faces in the Wild, the Labelled Face Parts in the Wild and the Annotated Facial Landmarks in the Wild.

show abstract

Section: B Facial Point Detectionmentioning

confidence: 99%

Privileged Information-Based Conditional Structured Output Regression Forest for Facial Point Detection

Yang

Patras

2015

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…For example, Cosar and Cetin [4] use the fact that the ratio of distances between co-linear points is fixed under affine transformations to restrict the inter-frame change in facial point tracking. Liang et al [21] use a condensation algorithm modified with spatial constraints.…”

Section: Related Workmentioning

confidence: 99%

Local Evidence Aggregation for Regression-Based Facial Point Detection

Martínez

Valstar

Binefa

et al. 2013

IEEE Trans. Pattern Anal. Mach. Intell.

119

103

View full text Add to dashboard Cite

Abstract-We propose a new algorithm to detect facial points in frontal and near-frontal face images. It combines a regressionbased approach with a probabilistic graphical model-based face shape model, that restricts the search to anthropomorphically consistent regions. While most regression-based approaches perform a sequential approximation of the target location, our algorithm detects the target location by aggregating the estimates obtained from stochastically selected local appearance information into a single robust prediction. The underlying assumption is that by aggregating the different estimates, their errors will cancel out as long as the regressor inputs are uncorrelated. Once this new perspective is adopted, the problem is reformulated as how to optimally select the test locations over which the regressors are evaluated. We propose to extend the regressionbased model to provide a quality measure of each prediction, and use the shape model to restrict and correct the sampling region. Our approach combines the low computational cost typical of regression-based approaches with the robustness of exhaustive-search approaches. The proposed algorithm was tested on over 7,500 images from 5 databases. Results showed significant improvement over the current state of the art.

show abstract

“…There are two methods, feature-based method and optical-flow method, which are used in solving these problems. Feature-based method [5][6][7] requires the relationship between a set of features extracted from one image and those extracted from the next image. There have been various features, including points, lines, planes, conics and combinations of these features.…”

Section: Introductionmentioning

confidence: 99%

“…The perspective projection of a circle in arbitrary orientations is always an ellipse. 5. It has been shown that circular feature has the property of high location accuracy.…”

Section: Introductionmentioning

confidence: 99%

A novel image analysis method for rotational motion of circular feature based on perspective projection

Liu

Zhao

2015

The Imaging Science Journal

View full text Add to dashboard Cite

Circular feature has been widely applied in recognising moving object and estimating its position and orientation. Based on the principle of perspective projection, the rotational motion of circular feature was analysed. According to whether the rotational motion occurred in the plane of circular feature, the rotational motion could be classified into two types: in-plane rotation and non-in-plane rotation. For the convenience of computation, we set the camera's initial position and orientation. The optical axis of camera was set perpendicular to the plane of circular feature. We analysed the perspective distortion of the circular feature. Additionally, based on the geometrical reasoning, we also proposed the method of computing the rotation angles. Finally, some experiments were carried to verify correctness and feasibility of the computing method. The results showed that the measurement of rotation angle could achieve better performance though the error existed. The maximum value of errors was only 0?4u. The average errors and the standard deviations of errors were also small.

show abstract

A graphical model based solution to the facial feature point tracking problem

Cited by 10 publications

References 24 publications

Privileged Information-Based Conditional Structured Output Regression Forest for Facial Point Detection

Privileged Information-Based Conditional Structured Output Regression Forest for Facial Point Detection

Local Evidence Aggregation for Regression-Based Facial Point Detection

A novel image analysis method for rotational motion of circular feature based on perspective projection

Contact Info

Product

Resources

About