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.

Isolation of a novel cyanophage infectious to the filamentous cyanobacterium Planktothrix agardhii (Cyanophyceae) from Lake Donghu, China

A previously unknown cyanophage, PaV-LD (Planktothrix agardhii Virus isolated from Lake Donghu), which causes lysis of the bloom-forming filamentous cyanobacterium P. agardhii, was isolated from Lake Donghu, Wuhan, China. PaV-LD only lysed P. agardhii strains isolated from Lake Donghu and not those isolated from other lakes. The PaV-LD particle has an icosahedral, non-tailed structure, ca. 70 to 85 nm (mean +/- SD = 76 +/- 6 nm) in diameter. PaV-LD was stable at freezing temperature, but lost its infectivity at temperatures >50 degrees C. Lysis of host cells was delayed about 3 d after the PaV-LD treatment with chloroform, and the virus was inactivated by exposure to low pH (<= 4). The latent period and burst size of the PaV-LD were estimated to be 48 to 72 h and about 340 infectious units per cell, respectively. The regrowth cultures of surviving host filaments were not lysed by the PaV-LD suspension. To our knowledge, this is the first isolation and cultivation of a virus infectious to the filamentous bloom-forming cyanobacterium Planktothrix from a freshwater lake.

Principle of the moving-mirror-pair interferometer and the tilt tolerance of the double moving mirror

A novel type of interferometer, the moving-mirror-pair interferometer, is presented, and its principle and properties are studied. The new interferometer is built with three flat mirrors, which include two flat moving mirrors fixed as a single moving part by a rigid structure and one flat fixed mirror. The optical path difference (OPD) is obtained by the straight reciprocating motion of the double moving mirror, and the OPD value is four times the physical shift value of the double moving mirror. The tilt tolerance of the double moving mirror of the novel interferometer is systematically analyzed by means of modulation depth and phase error. Where the square aperture is concerned, the formulas of the tilt tolerance were derived. Due to the novel interferometer's large OPD value and low cost, it is very applicable to the high-spectral-resolution Fourier-transform spectrometers for any wavenumber region from the far infrared to the ultraviolet. (C) 2008 Optical Society of America.

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.

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.

The structure of helokinestatin-5 and its biosynthetic precursor from Gila monster (Heloderma suspectum) venom: Evidence for helokinestatin antagonism of bradykinin-induced relaxation of rat tail artery smooth muscle

Here we report the primary structure of a novel peptide, named helokinestatin-5 (VPPPLQMPLIPR), from the venom of the Gila monster (Heloderma suspectum). Helokinestatin-5 differs in structure from helokinestatin-3 by deletion of a single prolyl residue in the N-terminally located polyproline region. Two different biosynthetic precursors were consistently cloned from a venom-derived cDNA library. The first encoded helokinestatins 1–4 and a single copy of C-type natriuretic peptide, as previously described, whereas the second was virtually identical, lacking only a single prolyl codon as found in the mature attenuated helokinestatin-5 peptide. Helokinestatins 1–3 and 5 were synthesized by solid-phase fmoc chemistry and each synthetic replicate was found to antagonize the relaxation effect induced by bradykinin on rat tail artery smooth muscle. Helokinestatins thus represent a novel family of vasoactive peptides from the venom of helodermatid lizards

Statistical-based monitoring of multivariate non-Gaussian systems

The monitoring of multivariate systems that exhibit non-Gaussian behavior is addressed. Existing work advocates the use of independent component analysis (ICA) to extract the underlying non-Gaussian data structure. Since some of the source signals may be Gaussian, the use of principal component analysis (PCA) is proposed to capture the Gaussian and non-Gaussian source signals. A subsequent application of ICA then allows the extraction of non-Gaussian components from the retained principal components (PCs). A further contribution is the utilization of a support vector data description to determine a confidence limit for the non-Gaussian components. Finally, a statistical test is developed for determining how many non-Gaussian components are encapsulated within the retained PCs, and associated monitoring statistics are defined. The utility of the proposed scheme is demonstrated by a simulation example, and the analysis of recorded data from an industrial melter.

The synthesis and electronic structure of a novel [NiS4Fe2(CO)(6)] radical cluster: Implications for the active site of the [NiFe] hydrogenases

A novel [Ni'S-4'Fe-2(CO)(6)] cluster (1: 'S-4'=(CH3C6H3S2)(2)(CH2)(3)) has been synthesised, structurally characterised and has been shown to undergo a chemically reversible reduction process at -1.31 V versus Fc(+)/Fc to generate the EPR-active monoanion 1(-). Multifrequency Q-, X- and S-band EPR spectra of Ni-61-enriched 1(-) show a well-resolved quartet hyperfine splitting in the low-field region due to the interaction with a single Ni-61 (I = 3/2) nucleus. Simulations of the EPR spectra require the introduction of a single angle of non-coincidence between g, and A(1), and g(3) and A(3) to reproduce all of the features in the S- and X-band spectra. This behaviour provides a rare example of the detection and measurement of non-coincidence effects from frozen-solution EPR spectra without the need for single-crystal measurements, and in which the S-band experiment is sensitive to the non-coincidence. An analysis of the EPR spectra of 1(-) reveals a 24% Ni contribution to the SOMO in 1(-), supporting a delocalisation of the spin-density across the NiFe2 cluster. This observation is supported by IR spectroscopic results which show that the CO stretching frequencies, v(CO), shift to lower frequency by about 70 cm(-1) when 1 is reduced to 1(-). Density functional calculations provide a framework for the interpretation of the spectroscopic properties of 1(-) and suggest that the SOMO is delocalised over the whole cluster, but with little S-centre participation. This electronic structure contrasts with that of the Ni-A, -B, -C and -L forms of [NiFe] hydrogenase in which there is considerable S participation in the SOMO.

Soft tissue artifact distribution on lower limbs during treadmill gait: Influence of skin markers' location on cluster design.

Segment poses and joint kinematics estimated from skin markers are highly affected by soft tissue artifact (STA) and its rigid motion component (STARM). While four marker-clusters could decrease the STA non-rigid motion during gait activity, other data, such as marker location or STARM patterns, would be crucial to compensate for STA in clinical gait analysis. The present study proposed 1) to devise a comprehensive average map illustrating the spatial distribution of STA for the lower limb during treadmill gait and 2) to analyze STARM from four marker-clusters assigned to areas extracted from spatial distribution. All experiments were realized using a stereophotogrammetric system to track the skin markers and a bi-plane fluoroscopic system to track the knee prosthesis. Computation of the spatial distribution of STA was realized on 19 subjects using 80 markers apposed on the lower limb. Three different areas were extracted from the distribution map of the thigh. The marker displacement reached a maximum of 24.9mm and 15.3mm in the proximal areas of thigh and shank, respectively. STARM was larger on thigh than the shank with RMS error in cluster orientations between 1.2° and 8.1°. The translation RMS errors were also large (3.0mm to 16.2mm). No marker-cluster correctly compensated for STARM. However, the coefficient of multiple correlations exhibited excellent scores between skin and bone kinematics, as well as for STARM between subjects. These correlations highlight dependencies between STARM and the kinematic components. This study provides new insights for modeling STARM for gait activity.

Should a non-rival public good always be provided centrally?

This paper discusses the problem of optimal design of a jurisdiction structure from the view point of a utilitarian social planner when individuals with identical utility functions for a non-rival public good and private consumption have private information about their contributive capacities. It shows that the superiority of a centralized provision of a non-rival public good over a federal one does not always hold. Specifically, when differences in individuals’ contributive capacities are large, it is better to provide the public good in several distinct jurisdictions rather than to pool these jurisdictions into a single one. In the specific situation where individuals have logarithmic utilities, the paper provides a complete characterization of the optimal jurisdiction structure in the two-type case.

Three-Dimensional Motion Estimation Using Shading Information in Multiple Frames

A new formulation for recovering the structure and motion parameters of a moving patch using both motion and shading information is presented. It is based on a new differential constraint equation (FICE) that links the spatiotemporal gradients of irradiance to the motion and structure parameters and the temporal variations of the surface shading. The FICE separates the contribution to the irradiance spatiotemporal gradients of the gradients due to texture from those due to shading and allows the FICE to be used for textured and textureless surface. The new approach, combining motion and shading information, leads directly to two different contributions: it can compensate for the effects of shading variations in recovering the shape and motion; and it can exploit the shading/illumination effects to recover motion and shape when they cannot be recovered without it. The FICE formulation is also extended to multiple frames.

Image-Based View Synthesis

We present a new method for rendering novel images of flexible 3D objects from a small number of example images in correspondence. The strength of the method is the ability to synthesize images whose viewing position is significantly far away from the viewing cone of the example images ("view extrapolation"), yet without ever modeling the 3D structure of the scene. The method relies on synthesizing a chain of "trilinear tensors" that governs the warping function from the example images to the novel image, together with a multi-dimensional interpolation function that synthesizes the non-rigid motions of the viewed object from the virtual camera position. We show that two closely spaced example images alone are sufficient in practice to synthesize a significant viewing cone, thus demonstrating the ability of representing an object by a relatively small number of model images --- for the purpose of cheap and fast viewers that can run on standard hardware.

Geometric and Algebraic Aspects of 3D Affine and Projective Structures from Perspective 2D Views

We investigate the differences --- conceptually and algorithmically --- between affine and projective frameworks for the tasks of visual recognition and reconstruction from perspective views. It is shown that an affine invariant exists between any view and a fixed view chosen as a reference view. This implies that for tasks for which a reference view can be chosen, such as in alignment schemes for visual recognition, projective invariants are not really necessary. We then use the affine invariant to derive new algebraic connections between perspective views. It is shown that three perspective views of an object are connected by certain algebraic functions of image coordinates alone (no structure or camera geometry needs to be involved).