This paper introduces a new fast motion estimation based on a maximum matching error constraint that can eliminate an impossible candidate block much earlier than a conventional partial distortion elimination (PDE) scheme. The main contributions of the proposed scheme are that 1) it can reduce the computational cost considerably for the matching error calculation; and 2) the proposed matching ideas can be applied to the conventional PDE algorithms without significant changes. In order to evaluate the proposed scheme, we compare the results of the proposed algorithm with several baseline approaches. The experimental results show that the proposed algorithm can reduce the computations more than 62% for motion estimation at the cost of 0.005 dB quality degradation versus the general PDE algorithm.

Undisturbed interaction is essential to provide immersive AR environments. There have been a lot of approaches to interact with VEs (virtual environments) so far, especially in hand metaphor. When the user’s hands are being used for hand-based work such as maintenance and repair, necessity of alternative interaction technique has arisen. In recent research, hands-free gaze information is adopted to AR to perform original actions in concurrence with interaction. [3, 4]. There has been little progress on that research, still at a pilot study in a laboratory setting. In this paper, we introduce such a simple WARS(wearable augmented reality system) equipped with an HMD, scene camera, eye tracker. We propose ‘Aging’ technique improving traditional dwell-time selection, demonstrate AR gallery — dynamic exhibition space with wearable system.

In this paper, we propose a new method using only three vanishing points to recover the dimensions of object and its pose from a single image with a camera of unknown focal length. Our approach is to compute the dimensions of objects represented by the unit vector of objects from an image. The dimension vector v can be solved by the standard nonlinear optimization techniques with a multistart method which generates multiple starting points for the optimizer by sampling the parameter space uniformly. This method allows model-based vision to be computed for the dimensions of object for a 3D model from matches to a single 2D image. Experimental results show the actual dimensions of object from an image agree well with the calculated results.

In this paper, we present a content-based image retrieval method that is less sensitive to some rotations and translations of an image by using the fuzzy region segmentation. The algorithm uses the two features of color and spatial information to retrieve similar images from a database. That is to say, we use both color and average coordinate of color distribution to obtain the spatial information of each segmented region. Furthermore, we also propose the alternative to the ripple phenomenon, which is occurred in the conventional fuzzy region segmentation algorithm.

This paper proposes a new key frame extraction algorithm and also provides a novel video-indexing scheme for fast content-based browsing and retrieval in a video database. The algorithm is designed to function with a compressed video stream. We first extract key frames by matching AHIM (Accumulated Histogram Intersection Measure) of DC image sequence constructed from the MPEG video sequence. Then we use the region segmentation-based projective histogram and its moments as database indices for video retrieval. From the simulation results, the proposed method clearly shows the validity and superiority in respect of computation time and memory space, and that in conjunction with other techniques for indexing, such as color, can provide a powerful framework for image indexing and retrieval.

This paper presents a new volumetric approach to 3D object recognition by using PBSM (Part-Based Superquadric Model). The assembly part object usually can be constructed with the set of volumetric primitives and the relationships of them. We describe volumetric properties of the model object with superquadric parameters. In addition, our model base has the relationships of volumetric primitives as well as the surface information: the surface type, the junction type between neighboring surfaces. These surface attributes and relationships of volumetrics are effectively used in recognition process. Our integrated method is robust to recognition of the identity, position, and orientation of randomly oriented assembly parts. Furthermore, we can reduce the effects of self-occlusion and non-linear shape changes according to the viewpoint.

We propose a method to extract depth information from a two-dimensional (2D) image. We project a sinusoidal pattern on to the object and compute the pattern's frequency as it varies with the shape of the object. The extracted frequency enables us to measure depth information. The main advantage of the proposed method is the use of only one active projected image. We expect that the proposed algorithm will be applied to vision systems, virtual environments etc. In an experiment, the algorithm showed a mean error of about 2.44 percent in measuring depth compared with ground truth.

The main interest of previous researches for shape from shading was only on a Lambertian surface which consists of a diffuse reflectance surface. In practice, natural objects have hybrid reflectance, which limits the recovery of shape from intensity distribution. In this paper, we propose the method of determining shape using a neural network and diffuse illumination. The segmented region of sphere surface is used for a training set, which can be determined by shadow line and edge of surface. Diffuse illumination is used to avoid specular spike and highlight which usually cause many problems such as intensity disparities. The diffuse illumination method using flat paper allows us to reduce these phenomena with a simple scheme. The neural network and diffuse illumination method are useful for shape from shading, because it can be applied to objects of unknown reflectance properties, but it is applied only to Torrance-Sparrow style reflectance.

This paper presents a new algorithm that obtains the surface orientation of the texture image using structural approaches. The proposed method showed that structural approaches can be effectively used in 3-D analysis of textures as well as description and segmentation. By examining the Fourier power spectrum of the texture image, we determine the tilt of the textured surface. Then, 1-D projection information of the texture in the obtained tilt direction is used to compute the slant. Using the obtained information, we can compute the vanishing point, and rearrange the textured surface with lines converging to the vanishing point and lines perpendicular to the tilt. In the experimental results, we have ascertained the proposed method can make a precise 3-D analysis of structural textures.

We propose the new method for analyzing the properties of surface reflectance and reconstructing the shape of an object using estimated reflectance parameters. We examine the hybrid reflectance surface which has a specular lobe and a diffuse component, which can be explained by the Torrance-Sparrow (1967) model. A sample sphere made of one material is used to estimate the reflectance properties by using the LMS algorithm. We can obtain the reference image which consists of the surface normal and brightness value using estimated reflectance parameters. The photometric matching method proposed is robust because it matches the object image with the reference image by considering its neighbour brightness distribution. It is expected that the proposed algorithm can be applied in vision inspection systems.

To recover 3-D shapes, a new algorithm which is based on the hybrid reflectance model and the indirect diffuse illumination method is presented. The photometric stereo method (PSM) is modified with the indirect diffuse illumination method (IDIM). We derive the surface reflectance function for a disk-type light source and make the 3-D reference table generated by applying the derived surface reflectance function to the hemisphere. The slant and tilt angle at each point of the intensity images which are the inputs for the algorithm is determined from the 3-D reference table. Finally, the 3-D shape of objects is recovered from the needle map.

On the premise that the texture image is planar surface, we present a new morphological approach to obtain a surface orientation using the variation of texture image caused by projective distortions. Centroids of texels can be obtained from the farthest texel to the nearest texel sequentially by recursive erosions, so an entire texture image is segmented into several sub-regions. Using the major axes and the size of structuring elements in segmented regions, we can compute the vanishing point and obtain the 3-D surface orientation. In the experimental results, we have ascertained the proposed algorithm was more effective than the aggregation transform.>

In this paper, we propose a new visual positioning and inspecting algorithm. The positioning algorithm is the process of computing the center position and the rotated angle of the surface mount device. The inspecting algorithm is capable of detecting the location of broken or bent leads. The morphological opening is applied to obtain a leadless image of the quad flat package or the small outline package, and the leadless image is able to be treated like an image of the square-shaped passive devices. The center position and the rotated orientation are found by four corner points which are obtained from the morphological skeleton subsets of the leadless image. After finding them, the hit and miss transform is utilized to extract the feature of the corner identification (ID). The ultimate orientation is determined correctly using results detected by hit structuring element and miss structuring element which satisfy the feature of the corner ID. The morphological inspection is executed extending over two steps, i.e., a rough inspection and a detailed inspection. The former is practiced before finding the center and the orientation, and the latter is performed after finding them.

A subband coder which can be applied to various kinds of images such as HDTV (high-definition television), digital TV, and videophone is proposed using human visual characteristics. Data compression is achieved by the procedure based on human visual characteristics, where motion vectors in lower band are utilized to process components in higher band according to the degree of movement. Since the compression of the signal in higher band is accomplished by motion information, the proposed method shows good performance regardless of the degree of movement. Compared with the existing methods, which uniformly process all subblocks in the same subband, the quality of the image reconstructed by the proposed method is shown to be superior.>

The new and accurate methods estimating ultrasonic velocity and attenuation are presented. The ultrasonic velocity is measured using the dip points obtained from echo spectrum. We have also developed the measurement method of attenuation considering the dispersion. To verify the algorithms, puls-reflection simulation and experiment are performed. The techniques are shown to be accurate.