Search

application/pdf, This paper describes a framework for aerial imaging of high dynamic range (HDR) scenes for use in virtual reality applications, such as immersive panorama applications and photorealistic superimposition of virtual objects using image-based lighting. We propose a complete and practical system to acquire full spherical HDR images from the sky, using two omnidirectional cameras mounted above and below an unmanned aircraft. The HDR images are generated by combining multiple omnidirectional images captured with different exposures controlled automatically. Our system consists of methods for image completion, alignment, and color correction, as well as a novel approach for automatic exposure control, which selects optimal exposure so as to avoid banding artifacts. Experimental results indicated that our system generated better spherical images compared to an ordinary spherical image completion system in terms of naturalness and accuracy. In addition to proposing an imaging method, we have carried out an experiment about display methods for aerial HDR immersive panoramas utilizing spherical images acquired by the proposed system. The experiment demonstrated HDR imaging is beneficial to immersive panorama using an HMD, in addition to ordinary uses of HDR images., journal article

This paper describes a framework for aerial imaging of high dynamic range (HDR) scenes for use in virtual reality applications, such as immersive panorama applications and photorealistic superimposition of virtual objects using image-based lighting. We propose a complete and practical system to acquire full spherical HDR images from the sky, using two omnidirectional cameras mounted above and below an unmanned aircraft. The HDR images are generated by combining multiple omnidirectional images captured with different exposures controlled automatically. Our system consists of methods for image completion, alignment, and color correction, as well as a novel approach for automatic exposure control, which selects optimal exposure so as to avoid banding artifacts. Experimental results indicated that our system generated better spherical images compared to an ordinary spherical image completion system in terms of naturalness and accuracy. In addition to proposing an imaging method, we have carried out an experiment about display methods for aerial HDR immersive panoramas utilizing spherical images acquired by the proposed system. The experiment demonstrated HDR imaging is beneficial to immersive panorama using an HMD, in addition to ordinary uses of HDR images.

application/pdf, This paper proposes a method for removing image inconsistencies which occur by an existence of moving objects or a change of illumination condition when an omnidirectional image database is generated. The database is used for archiving an outdoor scene in wide areas or generating novel view images with an image-based rendering approach. In related work, it is difficult to remove moving objects in an outdoor environment where illumination condition drastically changes, and to remove inconsistencies of color tone of images which included moving objects. The proposed method iterates the two processes which are the estimation of candidate region of moving objects and the achievement of color consistency to split regions. The color consistency is achieved by estimating linear color transformation parameters which change a histogram of an input image to that of the standard image., ACCV 2010 : 10th Asian Conference on Computer Vision , Nov 8-12, 2010 , Queenstown, New Zealand, conference paper

application/pdf, This paper proposes a method for removing image inconsistencies which occur by an existence of moving objects or a change of illumination condition when an omnidirectional image database is generated. The database is used for archiving an outdoor scene in wide areas or generating novel view images with an image-based rendering approach. In related work, it is difficult to remove moving objects in an outdoor environment where illumination condition drastically changes, and to remove inconsistencies of color tone of images which included moving objects. The proposed method iterates the two processes which are the estimation of candidate region of moving objects and the achievement of color consistency to split regions. The color consistency is achieved by estimating linear color transformation parameters which change a histogram of an input image to that of the standard image.

application/pdf, This paper describes a method of stereoscopic view generation by image-based rendering in wide outdoor environments. The stereoscopic view can be generated from an omnidirectional image sequence by a light field rendering approach which generates a novel view image from a set of images. The conventional methods of novel view generation have a problem such that the generated image is distorted because the image is composed of parts of several omnidirectional images captured at different points. To overcome this problem, we have to consider the distances between the novel viewpoint and observed real objects in the rendering process. In the proposed method, in order to reduce the image distortion, stereoscopic images are generated considering depth values estimated by dynamic programming (DP) matching using the images that are observed from different points and contain the same ray information in the real world. In experiments, stereoscopic images in wide outdoor environments are generated and displayed., SCIA 2007 : 15th Scandinavian Conference on Image Analysis , Jun 10-24, 2007 , Aalborg, Denmark, conference paper

This paper describes a method of stereoscopic view generation by image-based rendering in wide outdoor environments. The stereoscopic view can be generated from an omnidirectional image sequence by a light field rendering approach which generates a novel view image from a set of images. The conventional methods of novel view generation have a problem such that the generated image is distorted because the image is composed of parts of several omnidirectional images captured at different points. To overcome this problem, we have to consider the distances between the novel viewpoint and observed real objects in the rendering process. In the proposed method, in order to reduce the image distortion, stereoscopic images are generated considering depth values estimated by dynamic programming (DP) matching using the images that are observed from different points and contain the same ray information in the real world. In experiments, stereoscopic images in wide outdoor environments are generated and displayed.

application/pdf, It is necessary to precisely measure pose (position and orientation) of a user in order to realize an augmented reality (AR) system with a wearable computer. One of major methods for measuring user’s pose in AR is visual marker-based approach which calculates them by recognizing markers pasted up on the ceilings or walls. The method needs 3D pose information of visual markers in advance. However, much cost is necessary to calibrate visual markers pasted up on the ceiling in a wide environment. In this paper, an initialization tool for installing visual markers in wearable AR is proposed. The administrator is assisted in installing visual markers in a wide environment by the proposed tool. The tool calibrates alignment of visual markers which exist in the real environment with high accuracy by recognizing them in the images captured by a high-resolution still camera. Additionally, the tool assists the administrator in repairing the incorrect pattern of marker using a wearable AR system., ICAT 2006 : The 16th International Conference on Artificial Reality and Telexistence , Nov 29-Dec 2, 2006 , Hangzhou, Zejiang, China, conference paper

It is necessary to precisely measure pose (position and orientation) of a user in order to realize an augmented reality (AR) system with a wearable computer. One of major methods for measuring user’s pose in AR is visual marker-based approach which calculates them by recognizing markers pasted up on the ceilings or walls. The method needs 3D pose information of visual markers in advance. However, much cost is necessary to calibrate visual markers pasted up on the ceiling in a wide environment. In this paper, an initialization tool for installing visual markers in wearable AR is proposed. The administrator is assisted in installing visual markers in a wide environment by the proposed tool. The tool calibrates alignment of visual markers which exist in the real environment with high accuracy by recognizing them in the images captured by a high-resolution still camera. Additionally, the tool assists the administrator in repairing the incorrect pattern of marker using a wearable AR system.

application/pdf, This paper describes a method for modeling wide area outdoor environments by integrating omnidirectional range and color images. The proposed method efficiently acquires range and color data of outdoor environments by using omnidirectional laser rangefinder and omnidirectional multi-camera system (OMS). In this paper, we also give experimental results of reconstructing our campus from data acquired at 50 points., ICCVG 2004 : International Conference on Computer Vision and Graphics , Sep 22-24, 2004 , Warsaw, Poland, conference paper

application/pdf, This paper describes a database of annotation information for augmented reality (AR) on wearable computers. With the advance of computers, AR systems using wearable computers have received a great deal of attention. To overlay annotations on the real scene image, a user’s computer needs to hold annotation information. The purpose of this paper is to construct a networked database system of annotation information for wearable AR systems. The proposed system provides users with annotation information from a server via a wireless network so that the wearable computers do not need to hold it in advance and information providers can easily update and add the database with a web browser. In experiments, the user’s position-based annotations have been proven to be shown to the user effectively., PCM 2004 : Pacific Rim Conference on Multimedia , Nov 30-Dec 3, 2004 , Tokyo, Japan, conference paper

This paper describes a method for modeling wide area outdoor environments by integrating omnidirectional range and color images. The proposed method efficiently acquires range and color data of outdoor environments by using omnidirectional laser rangefinder and omnidirectional multi-camera system (OMS). In this paper, we also give experimental results of reconstructing our campus from data acquired at 50 points.

This paper describes a database of annotation information for augmented reality (AR) on wearable computers. With the advance of computers, AR systems using wearable computers have received a great deal of attention. To overlay annotations on the real scene image, a user’s computer needs to hold annotation information. The purpose of this paper is to construct a networked database system of annotation information for wearable AR systems. The proposed system provides users with annotation information from a server via a wireless network so that the wearable computers do not need to hold it in advance and information providers can easily update and add the database with a web browser. In experiments, the user’s position-based annotations have been proven to be shown to the user effectively.

application/pdf, Three-dimensional (3-D) models of outdoor scenes can be widely used in a number of fields such as object recognition, navigation, scenic simulation, and mixed reality. Such models are often made manually with high costs, so that automatic 3-D reconstruction has been widely investigated. In related works, a dense 3-D model is generated by using a stereo method. However, such approaches cannot use several hundred images together for dense depth estimation of large constructs and urban environments because it is difficult to accurately calibrate a large number of cameras. This paper proposes a novel dense 3-D reconstruction method that uses multiple image sequences. First, our method estimates extrinsic camera parameters of each image sequence, and then reconstructs a dense 3-D model of a scene using an extended multi-baseline stereo and voxel voting techniques., SCIA 2003 : 13th Scandinavian Conference on Image Analysis , Jun 29-Jul 2, 2003 , Halmstad, Sweden, conference paper

application/pdf, Three-dimensional (3-D) models of outdoor scenes can be widely used in a number of fields such as object recognition, navigation, scenic simulation, and mixed reality. Such models are often made manually with high costs, so that automatic 3-D reconstruction has been widely investigated. In related works, a dense 3-D model is generated by using a stereo method. However, such approaches cannot use several hundred images together for dense depth estimation of large constructs and urban environments because it is difficult to accurately calibrate a large number of cameras. This paper proposes a novel dense 3-D reconstruction method that uses multiple image sequences. First, our method estimates extrinsic camera parameters of each image sequence, and then reconstructs a dense 3-D model of a scene using an extended multi-baseline stereo and voxel voting techniques.