Takeshi Shakunaga

Decomposed eigenface for face recognition under various lighting conditions

Face recognition under various lighting conditions is discussed to cover cases when too few images are available for registration. This paper proposes decomposition of an eigenface into two orthogonal eigenspaces for realizing robust face recognition under such conditions. The decomposed eigenfaces consisting of two eigenspaces are constructed for each person even if only one image is available. A universal eigenspace called the canonical space (CS) plays an important role in creating the eigenspaces by way of decomposition, where CS is constructed a priori by principal component analysis (PCA) over face images of many people under many lighting conditions. In the registration stage, an input face image is decomposed to a projection image in CS and the residual of the projection. Then two eigenspaces are created independently in CS and in the orthogonal complement CS/sup /spl perp//. Some refinements of the two eigenspaces are also discussed. By combining the two eigenspaces, we can easily realize face identification that is robust to illumination change, even when too few images are registered. Through experiments, we show the effectiveness of the decomposed eigenfaces as compared with conventional methods.

Analysis of photometric factors based on photometric linearization.

We propose a method for analyzing photometric factors, such as diffuse reflection, specular reflection, attached shadow, and cast shadow. For analyzing real images, we utilize the photometric linearization method, which was originally proposed for image synthesis. First, we show that each pixel can be photometrically classified by a simple comparison of the pixel intensity. Our classification algorithm requires neither 3D shape information nor color information of the scene. Then, we show that the accuracy of the photometric linearization can be improved by introducing a new classification-based criterion to the linearization process. Experimental results show that photometric factors can be correctly classified without any special devices. A further experiment shows that the proposed method is effective for photometric stereo.

Invariant histograms and deformable template matching for SAR target recognition

Recognizing a target in synthetic-aperture radar (SAR) images is an important, yet challenging, application of the model-based vision technique. This paper describes a model-based SAR recognition system based on invariant histograms and deformable template matching techniques. An invariant histogram is a histogram of invariant values defined by geometric features such as points and lines in SAR images. Although a few invariants are sufficient to recognize a target, we use a histogram of all invariant values given by all possible target feature pairs. This redundant histogram enables robust recognition under severe occlusions typical in SAR recognition scenarios. Multi-step deformable template matching examines the existence of an object by superimposing templates over potential energy field generated from images or primitive features. It determines the template configuration which has the minimum deformation and the best alignment of the template with features. The deformability of the template absorbs the instability of SAR features. We have implemented the system and evaluated the system performance using hybrid SAR images, generated from synthesized model signatures and real SAR background signatures.

Photometric image-based rendering for image generation in arbitrary illumination

A Photometric Image-Based Rendering (PIER) concept is proposed that facilitates the generation of an image with an arbitrary illumination. Based on this concept, we aim to generate both diffuse and specular reflections. It is not necessary to explicitly recover 3D shape and reflection properties of the scene. In order to control appearance changes caused by modifications in the lighting conditions, we utilize a set of real images taken, in a variety of lighting conditions. Since the diffuse and specular reflection components have different characteristics, we separate these components and apply different methods to each. A photometric linearization is introduced to control diffuse reflections as well as for separating the other components. This also facilitates the treatment of attached shadows as a part of the diffuse reflection. A morphing technique is utilized to generate specular reflections. This is an effective technique for dealing with glossy objects, even when the light shape is clearly observed in the specular reflection. Experimental results show that realistic images can be successfully generated using this technique.

Integration of eigentemplate and structure matching for automatic facial feature detection

An algorithm is proposed for facial feature detection from a facial image. The algorithm consists of the bottom-up and the top-down interpretation processes, which work with the feature matching module and the structure matching module. Experimental results show that the proposed algorithm can detect no less than five features in 99.3% of the frontal views and can work even if the face orientation is unknown.

Perspective angle transform: Principle of shape from angles

This paper introduces a new relation, called the perspective angle transform (PAT), to deal with shape-from-angle problems, together with its application to 3D configuration recovery from an image. Three main aspects of PAT are presented in this paper. The first is the derivation of PAT which holds between the apparent and real angles under perspective projection. A concept is proposed of a virtual image plane and a new coordinate system, named the first perspective moving coordinate (FPMC) system, for analysis of the shape-from-range problems. Characteristics of FPMC are discussed briefly. The second point is the analysis of PAT properties, for which the general PAT form is introduced on another new coordinate system, named the second perspective moving coordinate (SPMC) system. Using this general form, the gradient of the plane including the real angle is constrained on a curve (PAT curve) of the fourth degree in the virtual image plane. The characteristics of PAT curves and relations between the general PAT form and skewed symmetry are summarized briefly. The last point concerns an application of PAT. As an example, we treat 3D configuration recovery from three arbitrary line segments in the image plane. This recovery corresponds to a generalization of the right-angled interpretation problem proposed and discussed by S.T. Barnard. A solution to the problem is shown using the general PAT form in conjunction with the concept of a virtual crossing point. The right-angled interpretation problem is ascribed to a quadratic equation. A simplified solution is also provided for the case where three line segments have a common crossing point in the image plane. This solution is based on the primary PAT form and is applicable to interpretation of a trihedral vertex with at least two right angles.

One-Dimensional search for reliable epipole estimation

Given a set of point correspondences in an uncalibrated image pair, we can estimate the fundamental matrix, which can be used in calculating several geometric properties of the images. Among the several existing estimation methods, nonlinear methods can yield accurate results if an approximation to the true solution is given, whereas linear methods are inaccurate but no prior knowledge about the solution is required. Usually a linear method is employed to initialize a nonlinear method, but this sometimes results in failure when the linear approximation is far from the true solution. We herein describe an alternative, or complementary, method for the initialization. The proposed method minimizes the algebraic error, making sure that the results have the rank-2 property, which is neglected in the conventional linear method. Although an approximation is still required in order to obtain a feasible algorithm, the method still outperforms the conventional linear 8-point method, and is even comparable to Sampson error minimization.

Pose estimation of jointed structures

A framework is proposed for model-based monocular vision covering not only conventional 3-D rigid models, but also flexible structures made up of 3-D rigid bodies connected by rotational joints. Pose-estimation problems from a single view are defined and discussed according to this object model composed of rigid bodies and rotational axes, respectively represented by sets of unit vectors and by single unit vectors. The authors define primitive problems as those which are solvable, but which would be unsolvable if any vector in the problem were invisible. A theorem is derived to extract a primitive problem family, members of which correspond to models containing rigid bodies and invisible rotational axes. Two generic rotation-estimation algorithms applicable to this problem family are also constructed. Experimental results from several primitive problems show the effectiveness of the proposed framework.<<ETX>>

Shape From Angles Under Perspective Projection

This paper provides a mathematical framework for shape-from-angle problems, and discusses four primitive problems based on it. For three of them, algebraic equations are provided from a geometric analysis. A generic lD search algorithm is proposed for designing a 1-D search algorithm, usable for any shape-from-angle problems. By designing the problem-dependent parts of the generic algorithm, specific l-D search algorithms are constructed for each problem. The generic algorithm finds all the feasible configurations by detecting the zero-crossing points of a problem-dependent test function on a constraint curve. Experimental results show the effectiveness of each specific algorithm and the applicability of the generic algorithm to several types of shape-from-angle problems.

An Object Pose Estimation System Using A Single Camera

This paper presents the framework for a CAD-based monocular vi- sion system that estimates an object pose. Estimation of pose from a single view is discussed according to an a priori 3-d ob- ject model that includes key primitive ele- ments and the relations among them. Es- timation strategy is discussed briefly, and several algorithms used in a pose estima- tion are presented. The proposed sys- tem - implemented in four subsystems for ob ject-image correspondence, feasible pose enumeration, pose selection, and estima- tion refinement - integrates sensor infor- mation and a priori knowledge about the treated objects. Experimental results for synthesized and real images show the effec- tiveness of this system for pose estimation and 3-d pose tracking.