Bum-Jong Lee

Vision-based real-time camera match moving using a known marker

We propose a fast and stable camera match-moving method aimed for real-time augmented reality applications. A known marker is used for fast and robust detection of marker points and for stable camera pose estimation. From the detected image points of a known marker on a single frame, we estimate the camera position and translation parameters. The entire pose estimation process is linear and initial estimates are not required. The most suitable reference point for the pose estimation is automatically chosen by investigating the depth variation. A planar marker is also allowed, since our linear method handles rank deficient cases. As an application of the proposed method, we implement a video augmentation system that replaces the marker in image frames with a virtual graphical object during the marker tracking. Experimental results show that the proposed camera tracking method is robust and fast enough for interactive video-based applications.

Vision-Based real-time camera matchmoving with a known marker

A primary requirement for practical augmented reality systems is a method of accurate and reliable camera tracking. In this paper, we propose a fast and stable camera matchmoving method aimed for real-time augmented reality application. A known marker is used for the fast detection and tracking of feature points. From the feature tracking of one of three different types of markers on a single frame, we estimate the camera position and translation parameters. The entire pose estimation process is linear and initial estimates are not required. As an application of the proposed method, we implemented a video augmentation system that replaces the marker in the image frames with a virtual 3D graphical object during the marker tracking. Experimental results showed that the proposed camera tracking method is robust and fast enough to interactive video-based applications.

View Synthesis from Two Unrectified Images

Interpolating views of real objects allows rendering realistic virtual views created without explicit 3D models. This paper proposes a method for rendering arbitrary views from a pair of uncalibrated images. The proposed method utilizes the matching of sparse edge points along epipolar lines instead of computing the entire dense disparity map for the view interpolation. Unavoidably, morphing image patches out of the field-of-view introduces holes in the virtual view. We fills such holes in a synthesized virtual view by blending neighboring colors. The experimental results show the viability of the proposed method for the wide baseline images on which a dense disparity is difficult to be obtained.

Hierarchical contribution culling for fast rendering of complex scenes

An intelligent culling technique is necessary in rendering 3D scenes to allow the real-time navigation in complex virtual environments. When rendering a large scene, a huge number of objects can reside in a viewing frustum. Virtual urban environments present challenges to interactive visualization systems, because of the huge complexity. This paper describes a LOD management (or contribution culling) method for the view-dependent real-time rendering of complex huge urban scenes. From the experimental results, we found the proposed LOD management method effectively limits the amount of objects to put into the rendering pipeline without notable degradation of rendering quality.

View-dependent simplification of complex urban scenes using weighted quadtrees

This article describes a new contribution culling method for the view-dependent real-time rendering of complex huge urban scenes. As a preprocessing step, the view frustum culling technique is used to cull away invisible objects that are outside the view frustum. For the management of the levels-of-detail, we subdivide the image regions and construct a weighted quadtree. The weight of each quadtree node is defined as the sum of weights of all objects contained in the node or its child nodes. The weight of an object is proportional to the view space area of the projected object as well as the distance from the viewpoint. Hence, large buildings in the far distance are not always culled out since their contributions to the rendering quality can be larger than those of near small buildings. We tested the proposed method by applying it to render a huge number of structures in our metropolitan section which is currently under development. Experimental results showed that the proposed rendering method guarantees real-time rendering of complex huge scenes.

Computation of Stereo Dense Disparity Maps Using Region Segmentation

Stereo vision is a fundamental method for measuring 3D structures by observing them from two cameras placed on different positions. In order to reconstruct 3D structures, it is necessary to create a disparity map from a pair of stereo images. To create a disparity map we compute the matching cost for each point correspondence and compute the disparity that minimizes the sum of the whole matching costs. In this paper, we propose a method to estimate a dense disparity map using region segmentation. We segment each scanline using region homogeneity properties. Using the segmented regions, we prohibit false matches in the stereo matching process. Disparities for pixels that failed in matching are filled by interpolating neighborhood disparities. We applied the proposed method to various stereo images of real environments. Experimental results showed that the proposed method is stable and potentially viable in practical applications.

Fast Vision-Based Camera Tracking for Augmented Environments

This article describes a fast and stable camera tracking method aimed for real-time augmented reality applications. From the feature tracking of a known marker on a single frame, we estimate the camera position and translation parameters. The entire pose estimation process is linear and initial estimates are not required. As an experimental setup, we implemented a video augmentation system to replace detected markers with virtual 3D graphical objects. Experimental results showed that the proposed camera tracking method is robust and fast applicable to interactive augmented reality applications.