M. J. Black and A. D. Jepson. Recognizing facial expressions in image sequences using local parameterized models of image motion. IJCV, 26(1), 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....method, ICondensation [9] improves the performance in cluttered scenes by using importance sampling, guided by an additional information channel. A limitation of this method is applicablity to onedimensional contours. The most successful model less approach to tracking, due to Black and Yacoob [2], is based on a model of motion which is obtained from computing optical flow. Although this method has proved to be extremely practical, we bear in mind that modeling motion is itself a challenging problem, especially for non rigid motion. It is also unclear how this method can handle the ....

Michael J. Black and Yasser Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

.... Point distribution model [41] 2 view point based models [68] Labeled graphs [86] 56] 40] Motion Extraction Holistic Methods Local Methods Dense Optical Flow Dense ow elds [52] 2] Region based ow [57] 65] 85] Motion Models 3D motion models [33] 25] Parametric motion models [9][84] 3D deformable models [21] 3D motion models [5] Feature Point Tracking Feature tracking [52] 80] 82] 64] 72] Di erence Images Holistic di . imgs [2] 22] 35] 34] Region based di erence images [12] Marker based Highlighted facial features [4] Dot markers [78] 44] Table 1: ....

....[8] Note that even though face normalization may be a reasonable approach in conjunction with some face analysis approaches, it is not mandatory, as long as extracted feature parameters are normalized prior to their classi cation. Indeed, appearance based model [51] and local motion model [9] approaches have dealt with signi cant out of plane rotations without relying on face normalization. 3.2 Facial Feature Extraction and Representation Feature extraction methods can be categorized according to whether they focus on motion or deformation of faces and facial features, respectively ....

[Article contains additional citation context not shown here]

M. J. Black and Y. Yacoob. Recognizing Facial Expressions in Image Sequences using Local Parameterized Models of Image Motion. International Journal of Computer Vision, 25(1):23-48, 1997.

....method. The best recognition is a rate of 92.7 obtained by combining Gabor wavelets and geometry features. 1. Introduction In facial feature extraction of expression analysis, there are mainly two types of approaches: geometric featurebased methods and appearance based methods [1, 2, 3, 5, 6, 7, 10, 11, 12, 13, 15, 17, 16, 18, 19]. The geometric facial features present the shape and locations of facial components (including mouth, eyes, brows, nose etc. The facial components or facial feature points are extracted to form a feature vector that represents the face geometry. In appearance based methods, image filters, such ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

....of variability in shape, texture, pose, and imaging conditions. Detecting faces in images has received much attention. A comprehensive survey on face detection methods can be found in [1] A huge research effort has been devoted to detecting and tracking of facial features in 2D and 3D (e.g. [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]) Recovering the face position and the facial expression automatically from a video is a difficult problem. This problem can be made easier by using markers on the face such as heavy makeup or a set of colored dots stuck onto the face [14] Although the use of markers reduces the difficulty of ....

M.J. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parametrized models of image motion," International Journal of Computer Vision, vol. 25, no. 1, pp. 23--48, 1997.

....[15] uses anthropometric data and inspired deformations to generate faces. A learning based statistical model can help tracking of face models [6] Eigen based approaches can successfully track, fit, and even recognize objects [33, 5, 34] In [9] the head is modeled as a cylinder, and in [3] as a plane, and in [34] tracking is used for animation, to mention just a few. Cue integration is not a new topic. In [14] a two cue integration algorithm is presented based on the use of constraints, in which optical flow is defined to be the constraining (i.e. most important) cue, and edges ....

M. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. Int. J. of Comp. Vision, 25(1):23--48, 1997. 32

....conducted we show the gain in performance and compare different procedures to select the set of points used in tracking. 1. INTRODUCTION Tracking planar patches is a subject of interest in computer vision, with applications in augmented reality [1] mobile robot navigation [2] face tracking [3], or the generation of super resolution images [4] Traditional approaches to tracking are based on nding correspondences in successive images. This can be achieved by computing optical ow [5] or by matching a sparse collection of features [6] In ow based methods, a velocity vector is computed ....

M. J. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parameterized models of image motion," Int. Journal of Computer Vision, vol. 25, no. 1, pp. 23--48, 1997.

....is through the use of multiple cameras yielding multiple views of the face and its features. To date, nonetheless, the works on automatic facial gestures analysis have avoided dealing with facial views other than a frontal one: portraits (e.g. 5, 7] or nearly frontal views of faces (e.g. [8, 9]) constitute the input data processed by the existing systems. For an exhaustive review on the past attempts to address the problems of automatic facial gesture recognition in frontal and nearly frontal views of faces, readers are referred to [3] From several methods for recognition of facial ....

....of automatic AU coding from face profile image sequences. It was undertaken with two motivations: 1. In a frontal view of the face, facial gestures such as showing the tongue (AU19) or pushing the jaw forwards (AU29) represent out plane non rigid facial movements which are difficult to detect [7, 8, 9]. Such facial gestures are clearly observable in a profile view of the face. 2. A basic understanding of how to achieve automatic facial gesture analysis from human face profiles is necessary for the 0 7803 7304 9 02 17.00 2002 IEEE establishment of a technological framework for automatic ....

M. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parameterized models of image motion", Computer Vision, vol. 25, no. 1, pp. 23-48, 1997.

....is through the use of multiple cameras yielding multiple views of the face and its features. To date, nonetheless, the works on automatic facial gestures analysis have avoided dealing with facial views other than a frontal one: portraits (e.g. 6, 8] or nearly frontal views of faces (e.g. [9, 10]) constitute the input data processed by the existing systems. For exhaustive reviews on the past attempts to address the problems of automatic facial gesture recognition in frontal and nearly frontal views of faces, readers are referred to [4, 6] From several methods for recognition of facial ....

....of automatic AU coding from face profile image sequences. It was undertaken with two motivations: 1. In a frontal view of the face, facial gestures such as showing the tongue (AU19) or pushing the jaw forwards (AU29) represent out plane non rigid facial movements which are difficult to detect [8, 9, 10]. Such facial gestures are clearly observable in a profile view of the face. 2. A basic understanding of how to achieve automatic facial gesture analysis from human face profiles is necessary for the establishment of a technological framework for automatic facial gestures analysis from multiple ....

M. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parameterized models of image motion", Computer Vision, vol. 25, no. 1, pp. 23-48, 1997.

....not only planar objects, but also non planar objects with limited out of plane rotations, as is the case of face tracking. 1. Introduction Tracking planar patches is a subject of interest in computer vision, with applications in augmented reality [5] mobile robot navigation [7] face tracking [1], or the generation of super resolution images [2] Traditional approaches to tracking are based on finding correspondences in successive images. This can be achieved by computing optical flow [4] or by matching a sparse collection of features [6] In flow based methods, a velocity vector is ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. Int. Journal of Computer Vision, 25(1):23--48, 1997.

....how well the features represent the facial actions, but also on how well these features can be extracted. Most researchers have focused on video or images of the frontal face [1, 13, 18, 31, 3] The features that are usually extracted for the purpose of face analysis range from optical ow elds [4, 18, 3] to gabor coecients [37] A lot of di erent approaches [4, 13, 18, 31, 3, 37]have been used to classify the features to recognize facial actions. This work is divided into three parts. Figure 3 1 gives you an overview of the system. The rst two parts are concerned with robust extraction of the ....

....how well these features can be extracted. Most researchers have focused on video or images of the frontal face [1, 13, 18, 31, 3] The features that are usually extracted for the purpose of face analysis range from optical ow elds [4, 18, 3] to gabor coecients [37] A lot of di erent approaches [4, 13, 18, 31, 3, 37]have been used to classify the features to recognize facial actions. This work is divided into three parts. Figure 3 1 gives you an overview of the system. The rst two parts are concerned with robust extraction of the features that are highly correlated with the facial actions. I use more than ....

M. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. Int. Journal of Computer Vision, 25(1):23-48, 1997.

....transformations. Much of the previous work to capture facial deformations has relied heavily on two forms of parameterizable models, namely, geometry of face musculature and head shape (Essa and Pentland [10] Terzopoulos and Waters [18] Yuille et al. 1] and motion estimation (Black and Yacoob [4], Ezzat [11] While the former needs to be hand crafted, the latter relies on repeated estimation of optical flow which can be computationally intensive. An important aspect of our work is that the models for all of the sub tasks needed to achieve the final goal is automatically learned from ....

....ideal localization for such a widely varying pattern. All of the above phenomena can have adverse e#ects on the performance of the next stage of mouth analysis. Some mouth localizers exist in literature they either rely on color information (Oliver et al. 16] or on optical flow (Black and Yacoob [4]) but our approach was to ensure that the mouth region remains stationary with respect to other more stable landmarks on the face such as eyes and nostrils. In this as well as the later stage of mouth analysis, the ability of the wavelet transform to encode localized variations plays a crucial ....

M.J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 1996. Revised preprint.

....more reliable. One can use an anatomically based model for head motion recovery. However, such a method tends to require precise or manual initialization for it to work well. Use of a much simpler geometric model of a head is often effective. Various planar model based methods have been presented [8,10]. They treat the face as a plane and use a single face texture (static template) to recover the head motion. They work well when the head orientation is not far from the frontal view. In [1,5] an ellipsoidal model was used with good results on 3D head or body tracking. Cascia et al. 3] developed ....

M. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. In IJCV, vol. 25, no. 1, pp. 23-48, 1997.

....Computer Vision Laboratory at the University of Maryland has been investigating problems related to detection, tracking and analysis of human activities for almost ten years. Our earliest work focused on tracking of facial features in the context of recognizing human facial expressions from motion [11, 2]. The system described in [2] which involved robust flow estimation, image stabilization, and tracking of several facial features, required more than one minute of what then passed for CPU time per frame. Clearly, the system was far from real time, which limited both experimentation during ....

....University of Maryland has been investigating problems related to detection, tracking and analysis of human activities for almost ten years. Our earliest work focused on tracking of facial features in the context of recognizing human facial expressions from motion [11, 2] The system described in [2], which involved robust flow estimation, image stabilization, and tracking of several facial features, required more than one minute of what then passed for CPU time per frame. Clearly, the system was far from real time, which limited both experimentation during development as well as ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, 1997.

....more efficient and more effective. Automatic Recognition of Facial Actions Recent advances in computer vision and pattern analysis facilitated automatic analysis of facial expressions from images. Different approaches have been taken in tackling the problem: analysis of facial motion [6] [1], 12] greylevel pattern analysis [20] analysis of facial features and their spatial arrangements [2] 8] 13] 10] holistic spatial pattern analysis [7] 17] The image analysis techniques in these systems are relevant to the goal of automatic facial expression data extraction, but the ....

....Yet, for investigations of facial behaviour itself, such as studying of the difference between genuine and simulated affective state, an objective and detailed measure of facial activity such as FACS is needed. Explicit attempts to automate facial action coding in images are few [3] Black et al. [1] use local parameterised models of image motion and few mid level predicates that are derived from the estimated motion parameters and describe the encountered facial change. Here the specificity of optical flow to action unit discrimination has not been described. Essa et al. 6] use ....

[Article contains additional citation context not shown here]

Black, M.J.; Yacoob, Y. 1998. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal on Computer Vision 25(1): 23-48.

....standard methods [10, 9] What makes this problem difficult is the confounding of motion, structure, and imaging geometry such that the appearances of different objects are often indistinguishable from one another. Similar problems are inherent in recognizing particular motions, e.g. gestures [8, 5, 7, 4], where the problem is often made tractable by the dominance of the motion component of the optical flow field. In fact, this is precisely what we wish to avoid in the object recognition context, hence some control of imaging parameters is required to ensure that a reasonable component of the flow ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parametrized models of image motion. Int. Journal of Computer Vision, 25(1):23--48, 1997. Also found in Xerox PARC, Techinical Report SPL-95-020.

....Our strategy will be to factor out of the optical flow field a signature related to shape by exploiting a priori knowledge of motion and imaging parameters learned through training. Others have demonstrated that this factoring problem can be solved in the presence of suitable a priori constraints [2, 19, 5], usually for the case of a stationary observer where the task is to identify the motion of a moving object. Here, we are attempting to recognize the object itself by driving the camera through a sequence of trajectories with the purpose of extracting a component of the flow related to shape (i.e. ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parametrized models of image motion. Int. Journal of Computer Vision, 25(1):23--48, 1997. Also found in Xerox PARC, Techinical Report SPL-95-020.

.... of the left ventricle in both 2D and 3D [17,15] In addition to tracking rigid objects, previous work focused on arbitrary non rigid motion and gave little attention to tracking objects moving in specific motion patterns, without the incorporation of statistical prior knowledge in both 2D and time [2]. In this paper, we present a new method for locating spatio temporal shapes (STshapes) in image sequences. We extend ASM [6] to include knowledge of temporal shape variations and present a ST shape modelling and segmentation technique. The method is well suited to model and segment objects with ....

Black M., Yacoob Y., Recognizing Facial Expressions in Image Sequences using Local Parametrized Models of Image Motion. IJCV, 1997, 25(1), 23-48.

....and immediate means for human beings to communicate emotions and intentions. Often emotions are expressed through the face before they are verbalized. In the past decade, much progress has been made in building computer systems to understand and use this natural form of human communication [1, 3, 2, 5, 7, 8, 9, 11, 13, 15, 18, 22, 19, 23, 24, 25]. Two procedures are necessary for an automatic expression analysis system: facial feature extraction and facial expression recognition. In facial feature extraction, there are mainly two types of approaches: geometric feature based methods and appearance based methods. In geometric feature based ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

....used to track the non rigid facial motion while estimating muscle actuator controls. In [13] a control theoretic approach was employed, based on normalized correlation between the incoming data and templates. Finally, global head motion can be tracked using a plane under perspective projection [7]. Recovered global planar motion is used to stabilize incoming images. Facial expression recognition is accomplished by tracking deforming image patches in the stabilized images. Most of the above mentioned techniques are not able to track the face in presence of large rotations and some require ....

....system automatically. Simple models, like a cylinder, require the estimation of fewer parameters in automatic placement schemes. As will be confirmed in experiments described in Sec. 8, tracking with the cylinder model is relatively robust to slight perturbations in initialization. A planar model [7] also offers these advantages; however, the experiments indicate that this model is not powerful enough to cope with the self occlusions generated by large head rotations. On the other hand, we have also experimented with a complex rigid head model generated averaging the Cyberware scans of ....

M.J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 1997.

....focused on classification of the universal expressions defined by Ekman [7] These expressions are sadness, anger, fear, disgust, surprise, happiness and contempt. Thus, the algorithms were tailored towards buildingmodels to recognize the universal expressions from static images or video sequences [4, 8, 14]. Recently, some work is being done towards recognition of individual action units that measure muscle action, proposed by Ekman as the basis for Facial Action Coding (FACS) 1, 5, 6] All the experiments done and models built for facial actions or expressions require precise image registration ....

M. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. Int. Journal on Computer Vision,25(1):23--48, 1997.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. IJCV, 25(1):23--48, 1997. 910, 920, 921

No context found.

M.J. Black, Y. Yacoob, Recognizing facial expressions in image sequences using local parameterized models of image motion, Internat. J. Comput. Vision 25 (1) (1997) 23--48.

No context found.

M.J. Black, Y. Yacoob, Recognizing facial expressions in image sequences using local parameterized models of image motion, Internat. J. Comput. Vision 25 (1) (1997) 23--48.

No context found.

M. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. IJCV, 25(1):23--48, 1997.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. IJCV, 25(1):23-48, 1997.

....first frame (no appearance model is previously learned) and after that the method is fully automatic. In this paper we focus on the application of face modeling. Most of the previous work on face tracking and modeling is focused on generic trackers, which are independent of the person s identity [5,9,10,23,24]. In particular, appearance based face trackers [10,26, 34] make use of PCA in order to construct a linear model of the face s subspace (variation across people) rather than the intra person variations due to changes in expression. When working with person specific models [15,17,20,23] PCA will ....

....k l appearance bases for the l layer. dk l will be equal to B for all pixels where# p =1(i.e. belonging to the l t mask) and otherwise can take an arbitrary value. 3. 2 Motion If the face to be tracked can be considered to be far away from the camera, it can be approximated by a plane [5]. The motion of planar surfaces, under orthographic or perspective projection, can be recovered with a parametric model of 6 or8 parameters. The rigid motion of the face will be parameterized by an affine model: x p y p y c# (3) where a . ....

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23-- 48, 1997.

....notion of constancy to more complex types of appearance change. One motivation for this is our interest in recognizing complex nonrigid and articulated motions, such as human facial expressions. Previous work in this area has focused on analyzing the image motion of face regions such as mouths [12]. But image motion alone does not capture appearance changes such as the systematic appearance disappearance of the teeth and tongue during speech and facial expressions. For machine recognition we would like to be able to model these intensity variations. Our framework extends several previous ....

....from a training set of approximately 500 images. The training set included image sequences of different subjects performing the facial expressions joy, anger, and sadness. The faces of each subject were stabilized with respect to the first frame in the sequence using a planar motion model [12]. The mouth regions were extracted from the stabilized sequences and PCA was performed. The first 11 basis images account for 85 of the variance in the training data and the first eight of these are shown in Fig. 15. 6.2. Learned Deformations We learn a domain specific model for the deformation ....

M. J. Black and Y. Yacoob, Recognizing facial expressions in image sequences using local parameterized models of image motion, Int. J. Comput. Vision 25, 1997, 23--48.

No context found.

M. J. Black and A. D. Jepson. Recognizing facial expressions in image sequences using local parameterized models of image motion. IJCV, 26(1), 1998.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, 1997.

No context found.

M. Black and Y. Yacoob, "Recognizing Facial Expressions in Image Sequences Using Local Parameterized Models of Image Motion", Intel. J. of Computer Vision, 25(1), pp. 23-48, 1997.

No context found.

M. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. In ICCV, 1995.

No context found.

Michael J. Black and Yasser Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. Int. Journal of Computer Vision, 25(1):23--48, 1997.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. Int. Journal of Computer Vision, 25(1):23--48, 1997.

No context found.

M. J. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parameterized models of image motion," Int. Journal of Computer Vision, vol. 25, no. 1, pp. 23--48, 1997.

No context found.

M. J. Black and Y. Yacoob, "Recognizing facial expression in image sequences using local parameterized models of image motion," Int'l J. Computer Vision, vol. 25, no. 1, pp. 23--48, 1997.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997. 6

No context found.

Michael J. Black and Yasser Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

No context found.

Michael J. Black and Yasser Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

No context found.

Michael J. Black and Yasser Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

No context found.

M. J. Black and Y. Yacoob. Recognizing Facial Expressions in Image Sequences Using Local Parameterized Models of Image Motion. International Journal on Computer Vision, 25(1):23--48, October 1997.

No context found.

M.J. Black and Y. Yacoob, "Recognizing Facial Expressions in Image Sequences Using Local Parameterized Models of Image Motion," Int'l J. Computer Vision, vol. 25, no. 1, pp. 23-48, 1997.

No context found.

M. J. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parameterized models of image motion," Int. J. Comput. Vis., vol. 25, no. 1, pp. 23--48, 1997.

No context found.

M. Black and Y. Yacoob, "Recognizing facial expressions in image sequences using local parameterized models of image motion," Comput. Vis., vol. 25, no. 1, pp. 23--48, 1997.

No context found.

M. Black and Y. Yacoob, "Recognizing Facial Expressions in Image Sequences Using Local Parameterized Models of Image Motion", Intel. J. of Computer Vision, 25(1), pp. 23-48, 1997.

No context found.

Black M., Yacoob Y., Recognizing Facial Expressions in Image Sequences using Local Parametrized Models of Image Motion. IJCV, 1997, 25(1), 23-48.

No context found.

M. J. Black and Y. Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. In Procceedings of the International Conference on Computer Vision, 1995.

No context found.

M.J. Black and Y. Yacoob, Recognizing Facial Expressions in Image Sequences Using Local Parameterized Models of Image Motion, Int'l J. Computer Vision, vol. 25, no. 1, pp. 23-48, Oct. 1997.

No context found.

Michael J. Black and Yasser Yacoob. Recognizing facial expressions in image sequences using local parameterized models of image motion. International Journal of Computer Vision, 25(1):23--48, October 1997.

No context found.

Black, M. and Yacoob, Y. 1997. "Recognizing facial expressions in image sequences using local parametric models of image motion". International Journal of Computer Vision 25(1): 23-48.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC