Geometry vs appearance for discriminating between posed and spontaneous emotions

Spontaneous facial expressions differ from posed ones in appearance, timing and accompanying head movements. Still images cannot provide timing or head movement information directly. However, indirectly the distances between key points on a face extracted from a still image using active shape models can capture some movement and pose changes. This information is superposed on information about non-rigid facial movement that is also part of the expression. Does geometric information improve the discrimination between spontaneous and posed facial expressions arising from discrete emotions? We investigate the performance of a machine vision system for discrimination between posed and spontaneous versions of six basic emotions that uses SIFT appearance based features and FAP geometric features. Experimental results on the NVIE database demonstrate that fusion of geometric information leads only to marginal improvement over appearance features. Using fusion features, surprise is the easiest emotion (83.4% accuracy) to be distinguished, while disgust is the most difficult (76.1%). Our results find different important facial regions between discriminating posed versus spontaneous version of one emotion and classifying the same emotion versus other emotions. The distribution of the selected SIFT features shows that mouth is more important for sadness, while nose is more important for surprise, however, both the nose and mouth are important for disgust, fear, and happiness. Eyebrows, eyes, nose and mouth are important for anger.

Graph rigidity for near-coplanar structure from motion

Recent algorithms for monocular motion capture (MoCap) estimate weak-perspective camera matrices between images using a small subset of approximately-rigid points on the human body (i.e. the torso and hip). A problem with this approach, however, is that these points are often close to coplanar, causing canonical linear factorisation algorithms for rigid structure from motion (SFM) to become extremely sensitive to noise. In this paper, we propose an alternative solution to weak-perspective SFM based on a convex relaxation of graph rigidity. We demonstrate the success of our algorithm on both synthetic and real world data, allowing for much improved solutions to marker less MoCap problems on human bodies. Finally, we propose an approach to solve the two-fold ambiguity over bone direction using a k-nearest neighbour kernel density estimator.

Fourier Lucas-Kanade Algorithm

In this paper we propose a framework for both gradient descent image and object alignment in the Fourier domain. Our method centers upon the classical Lucas & Kanade (LK) algorithm where we represent the source and template/model in the complex 2D Fourier domain rather than in the spatial 2D domain. We refer to our approach as the Fourier LK (FLK) algorithm. The FLK formulation is advantageous when one pre-processes the source image and template/model with a bank of filters (e.g. oriented edges, Gabor, etc.) as: (i) it can handle substantial illumination variations, (ii) the inefficient pre-processing filter bank step can be subsumed within the FLK algorithm as a sparse diagonal weighting matrix, (iii) unlike traditional LK the computational cost is invariant to the number of filters and as a result far more efficient, and (iv) this approach can be extended to the inverse compositional form of the LK algorithm where nearly all steps (including Fourier transform and filter bank pre-processing) can be pre-computed leading to an extremely efficient and robust approach to gradient descent image matching. Further, these computational savings translate to non-rigid object alignment tasks that are considered extensions of the LK algorithm such as those found in Active Appearance Models (AAMs).

Social signal processing for pain monitoring using a hidden conditional random field

utomatic pain monitoring has the potential to greatly improve patient diagnosis and outcomes by providing a continuous objective measure. One of the most promising methods is to do this via automatically detecting facial expressions. However, current approaches have failed due to their inability to: 1) integrate the rigid and non-rigid head motion into a single feature representation, and 2) incorporate the salient temporal patterns into the classification stage. In this paper, we tackle the first problem by developing a “histogram of facial action units” representation using Active Appearance Model (AAM) face features, and then utilize a Hidden Conditional Random Field (HCRF) to overcome the second issue. We show that both of these methods improve the performance on the task of pain detection in sequence level compared to current state-of-the-art-methods on the UNBC-McMaster Shoulder Pain Archive.

4D deformation modeling of cortical disease progression in Alzheimer's dementia

This work describes the development of a model of cerebral atrophic changes associated with the progression of Alzheimer's disease (AD). Linear registration, region-of-interest analysis, and voxel-based morphometry methods have all been employed to elucidate the changes observed at discrete intervals during a disease process. In addition to describing the nature of the changes, modeling disease-related changes via deformations can also provide information on temporal characteristics. In order to continuously model changes associated with AD, deformation maps from 21 patients were averaged across a novel z-score disease progression dimension based on Mini Mental State Examination (MMSE) scores. The resulting deformation maps are presented via three metrics: local volume loss (atrophy), volume (CSF) increase, and translation (interpreted as representing collapse of cortical structures). Inspection of the maps revealed significant perturbations in the deformation fields corresponding to the entorhinal cortex (EC) and hippocampus, orbitofrontal and parietal cortex, and regions surrounding the sulci and ventricular spaces, with earlier changes predominantly lateralized to the left hemisphere. These changes are consistent with results from post-mortem studies of AD.

Water-ethanol mixtures by molecular dynamics and x-ray Compton scattering

Water-ethanol mixtures are commonly used in industry and house holds. However, quite surprisingly their molecular-level structure is still not completely understood. In particular, there is evidence that the local intermolecular geometries depend significantly on the concentration. The aim of this study was to gain information on the molecular-level structures of water-ethanol mixtures by two computational methods. The methods are classical molecular dynamics (MD), where the movement of molecules can be studied, and x-ray Compton scattering, in which the scattering cross section is sensitive to the electron momentum density. Firstly, the water-ethanol mixtures were studied with MD simulations, with the mixture concentration ranging from 0 to 100%. For the simulations well-established force fields were used for the water and ethanol molecules (TIP4P and OPLS-AA, respectively). Moreover, two models were used for ethanol, rigid and non-rigid. In the rigid model the intramolecular bond lengths are fixed, whereas in the non-rigid model the lengths are determined by harmonic potentials. Secondly, mixtures with three different concentrations employing both ethanol models were studied by calculating the experimentally observable x-ray quantity, the Compton profile. In the MD simulations a slight underestimation in the density was observed as compared to experiment. Furthermore, a positive excess of hydrogen bonding with water molecules and a negative one with ethanol was quantified. Also, the mixture was found more structured when the ethanol concentration was higher. Negligible differences in the results were found between the two ethanol models. In contrast, in the Compton scattering results a notable difference between the ethanol models was observed. For the rigid model the Compton profiles were similar for all the concentrations, but for the non-rigid model they were distinct. This leads to two possibilities of how the mixing occurs. Either the mixing is similar in all concentrations (as suggested by the rigid model) or the mixing changes for different concentrations (as suggested by the non-rigid model). Either way, this study shows that the choice of the force field is essential in the microscopic structure formation in the MD simulations. When the sources of uncertainty in the calculated Compton profiles were analyzed, it was found that more statistics needs to be collected to reduce the statistical uncertainty in the final results. The obtained Compton scattering results can be considered somewhat preliminary, but clearly indicative of the behaviour of the water-ethanol mixtures when the force field is modified. The next step is to collect more statistics and compare the results with experimental data to decide which ethanol model describes the mixture better. This way, valuable information on the microscopic structure of water-ethanol mixtures can be found. In addition, information on the force fields in the MD simulations and on the ability of the MD simulations to reproduce the microscopic structure of binary liquids is obtained.

Learning based automatic face annotation for arbitrary poses and expressions from frontal images only

Statistical approaches for building non-rigid deformable models, such as the Active Appearance Model (AAM), have enjoyed great popularity in recent years, but typically require tedious manual annotation of training images. In this paper, a learning based approach for the automatic annotation of visually deformable objects from a single annotated frontal image is presented and demonstrated on the example of automatically annotating face images that can be used for building AAMs for fitting and tracking. This approach employs the idea of initially learning the correspondences between landmarks in a frontal image and a set of training images with a face in arbitrary poses. Using this learner, virtual images of unseen faces at any arbitrary pose for which the learner was trained can be reconstructed by predicting the new landmark locations and warping the texture from the frontal image. View-based AAMs are then built from the virtual images and used for automatically annotating unseen images, including images of different facial expressions, at any random pose within the maximum range spanned by the virtually reconstructed images. The approach is experimentally validated by automatically annotating face images from three different databases. © 2009 IEEE.

Curved Inertia Frames: Visual Attention and Perceptual Organization Using Convexity and Symmetry

In this paper we present an approach to perceptual organization and attention based on Curved Inertia Frames (C.I.F.), a novel definition of "curved axis of inertia'' tolerant to noisy and spurious data. The definition is useful because it can find frames that correspond to large, smooth, convex, symmetric and central parts. It is novel because it is global and can detect curved axes. We discuss briefly the relation to human perception, the recognition of non-rigid objects, shape description, and extensions to finding "features", inside/outside relations, and long- smooth ridges in arbitrary surfaces.

Recognition by Linear Combinations of Models

Visual object recognition requires the matching of an image with a set of models stored in memory. In this paper we propose an approach to recognition in which a 3-D object is represented by the linear combination of 2-D images of the object. If M = {M1,...Mk} is the set of pictures representing a given object, and P is the 2-D image of an object to be recognized, then P is considered an instance of M if P = Eki=aiMi for some constants ai. We show that this approach handles correctly rigid 3-D transformations of objects with sharp as well as smooth boundaries, and can also handle non-rigid transformations. The paper is divided into two parts. In the first part we show that the variety of views depicting the same object under different transformations can often be expressed as the linear combinations of a small number of views. In the second part we suggest how this linear combinatino property may be used in the recognition process.

On The Uniqueness of Correspondence Under Orthographic and Perspective Projections

The task of shape recovery from a motion sequence requires the establishment of correspondence between image points. The two processes, the matching process and the shape recovery one, are traditionally viewed as independent. Yet, information obtained during the process of shape recovery can be used to guide the matching process. This paper discusses the mutual relationship between the two processes. The paper is divided into two parts. In the first part we review the constraints imposed on the correspondence by rigid transformations and extend them to objects that undergo general affine (non rigid) transformation (including stretch and shear), as well as to rigid objects with smooth surfaces. In all these cases corresponding points lie along epipolar lines, and these lines can be recovered from a small set of corresponding points. In the second part of the paper we discuss the potential use of epipolar lines in the matching process. We present an algorithm that recovers the correspondence from three contour images. The algorithm was implemented and used to construct object models for recognition. In addition we discuss how epipolar lines can be used to solve the aperture problem.

Exploiting Phonological Constraints for Handshape Inference in ASL Video

Handshape is a key articulatory parameter in sign language, and thus handshape recognition from signing video is essential for sign recognition and retrieval. Handshape transitions within monomorphemic lexical signs (the largest class of signs in signed languages) are governed by phonological rules. For example, such transitions normally involve either closing or opening of the hand (i.e., to exclusively use either folding or unfolding of the palm and one or more fingers). Furthermore, akin to allophonic variations in spoken languages, both inter- and intra- signer variations in the production of specific handshapes are observed. We propose a Bayesian network formulation to exploit handshape co-occurrence constraints, also utilizing information about allophonic variations to aid in handshape recognition. We propose a fast non-rigid image alignment method to gain improved robustness to handshape appearance variations during computation of observation likelihoods in the Bayesian network. We evaluate our handshape recognition approach on a large dataset of monomorphemic lexical signs. We demonstrate that leveraging linguistic constraints on handshapes results in improved handshape recognition accuracy. As part of the overall project, we are collecting and preparing for dissemination a large corpus (three thousand signs from three native signers) of American Sign Language (ASL) video. The video have been annotated using SignStream® [Neidle et al.] with labels for linguistic information such as glosses, morphological properties and variations, and start/end handshapes associated with each ASL sign.

An alignment based similarity measure for hand detection in cluttered sign language video

Locating hands in sign language video is challenging due to a number of factors. Hand appearance varies widely across signers due to anthropometric variations and varying levels of signer proficiency. Video can be captured under varying illumination, camera resolutions, and levels of scene clutter, e.g., high-res video captured in a studio vs. low-res video gathered by a web cam in a user’s home. Moreover, the signers’ clothing varies, e.g., skin-toned clothing vs. contrasting clothing, short-sleeved vs. long-sleeved shirts, etc. In this work, the hand detection problem is addressed in an appearance matching framework. The Histogram of Oriented Gradient (HOG) based matching score function is reformulated to allow non-rigid alignment between pairs of images to account for hand shape variation. The resulting alignment score is used within a Support Vector Machine hand/not-hand classifier for hand detection. The new matching score function yields improved performance (in ROC area and hand detection rate) over the Vocabulary Guided Pyramid Match Kernel (VGPMK) and the traditional, rigid HOG distance on American Sign Language video gestured by expert signers. The proposed match score function is computationally less expensive (for training and testing), has fewer parameters and is less sensitive to parameter settings than VGPMK. The proposed detector works well on test sequences from an inexpert signer in a non-studio setting with cluttered background.

Active Voodoo Dolls: A Vision Based Input Device for Nonrigid Control

A vision based technique for non-rigid control is presented that can be used for animation and video game applications. The user grasps a soft, squishable object in front of a camera that can be moved and deformed in order to specify motion. Active Blobs, a non-rigid tracking technique is used to recover the position, rotation and non-rigid deformations of the object. The resulting transformations can be applied to a texture mapped mesh, thus allowing the user to control it interactively. Our use of texture mapping hardware allows us to make the system responsive enough for interactive animation and video game character control.

Improved Tracking of Multiple Humans with Trajectory Prediction and Occlusion Modeling

A combined 2D, 3D approach is presented that allows for robust tracking of moving bodies in a given environment as observed via a single, uncalibrated video camera. Tracking is robust even in the presence of occlusions. Low-level features are often insufficient for detection, segmentation, and tracking of non-rigid moving objects. Therefore, an improved mechanism is proposed that combines low-level (image processing) and mid-level (recursive trajectory estimation) information obtained during the tracking process. The resulting system can segment and maintain the tracking of moving objects before, during, and after occlusion. At each frame, the system also extracts a stabilized coordinate frame of the moving objects. This stabilized frame is used to resize and resample the moving blob so that it can be used as input to motion recognition modules. The approach enables robust tracking without constraining the system to know the shape of the objects being tracked beforehand; although, some assumptions are made about the characteristics of the shape of the objects, and how they evolve with time. Experiments in tracking moving people are described.

Trajectory Guided Tracking and Recognition of Actions

A combined 2D, 3D approach is presented that allows for robust tracking of moving people and recognition of actions. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. Low-level features are often insufficient for detection, segmentation, and tracking of non-rigid moving objects. Therefore, an improved mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. The 3D trajectory, occlusion, and segmentation information are utilized in extracting stabilized views of the moving object that are then used as input to action recognition modules. Trajectory-guided recognition (TGR) is proposed as a new and efficient method for adaptive classification of action. The TGR approach is demonstrated using "motion history images" that are then recognized via a mixture-of-Gaussians classifier. The system was tested in recognizing various dynamic human outdoor activities: running, walking, roller blading, and cycling. Experiments with real and synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise.