성과동영상

실감교류인체감응솔루션연구단의 주요 성과 동영상을 소개합니다.

ARWand for an Augmented World Builde

ARWand is a smartphone-based 3D user interface for manipulating 3D virtual objects in a wearable augmented reality (AR) environment without the use of a cumbersome external tracking system. In this paper, we apply ARWand to an augmented world builder that enables users to model the real world and add virtual content to it. First, the user, wearing a video, see-through, head-mounted display (HMD), selects an object of interest (OoI) in a physical space using ARWand and assigns a local reference coordinate to the OoI. This enables the program to construct a corresponding mirror space on the fly. Then, through control menus, ARWand lets the user load, select, and manipulate (3-DoF translate, 3-DoF rotate, scale, and texture) 3D virtual objects at various distances while the user moves within a wearable AR environment. We believe our proposed augmented world builder system using ARWand can help users easily construct augmented worlds anywhere and thus experience a more interesting and better informed reality. The proposed augmented world builder can be applied to various spaces such as galleries, museums, tour sites, or industrial locations and can create richly augmented reality ecosystems.
Generation of 3D Mirror World via 3D RGB-D SLAM

This is a preliminary result on building an object-oriented 3D RGB-Depth map for a large indoor space with high visual quality.The result is different from kinect based 3D RGB-Depth SLAM in two points that- It uses of 3D RGB-Depth sensor of which range is far larger (30m) than Kinect and can be of useful even in outdoor. Also, this sensor provides 360 deg full view as well as 180 deg depth coverage of which redeems the difficulty coming out of simultaneity (such as SLAM and SfM).- Its aim is not just RGB-Depth but implement that map in an object-oriented manner. In other words, the proposed technique automatically segments object and structure, so that each RGB-D map can be composed of object by object. This will enable the object-based semantic mapping which will make human interaction in this mirror world to be feasible.
Place Recognition using Straight Lines for Vision-based SLAM

Most visual simultaneous localization and mapping systems use point features as their landmarks and adopt point-based feature descriptors to recognize them.Compared to point landmarks, however, lines have strength in conveying the structural information of the environment.Despite the benefit, they have not been widely used because lines are more difficult in detecting, tracking, and recognizing, and this delayed the use of lines as landmarks.In this paper, we propose a place recognition algorithm using straight line features, which enables reliable detection of loop closures in a large complex environment under significant illumination changes.A vocabulary tree trained with mean standard-deviation line descriptor is used in finding the candidate matches between keyframes, and a Bayesian filtering framework enables reliable keyframe matching for large-scale loop closures.The proposed algorithm is compared with state-of-the-art point-based methods using scale-invariant feature transform or speeded up robust features.The experimental results show that the proposed method outperforms the others in challenging indoor environments
Robust-PCA based Hierarchical Plane Extraction for the Application of Geometric 3D SLAM

The result of plane extraction and 3D SLAM for a Multi-floor Indoor Building using the method
IMRC Gangnam Style (Kinect 3D Modeling & Character Animation) - 강남스타일

The 3D models of two toys(avatar and buzz) and a real human were created by IMRC 3D Modeler, which is described in the paper "Putting Real-World Objects into Virtual World: Fast Automatic Creation of Animatable 3D Models with a Consumer Depth Camera", ISUVR 2012. The motion data was taken from https://www.youtube.com/watch?v=1ySwO....Imaging Media Research Center, Korea Institute of Science and Technology (KIST)
kinect fusion + texture mapping + animation

This is a demonstration of fully automatic mesh/texture generation and animation after kinectfusion.
Miniature Alive: Augmented Reality-based Interactive DigiLog Experience in Miniature Exhibition

In this paper, we present Miniature Alive, a next-generation interactive miniature exhibition that provides a DigiLog experience that combines aesthetic/spatial feelings with an analog miniature and dynamic interaction with digitalized 3D content by exploiting augmented reality (AR) technology. Using our Miniature Alive, exhibition visitors can enjoy virtual storytelling in the physical miniature by turning a page of an e-book, interacting with augmented 3D objects through their mobile phones, and even change the original story. Our work is useful in guiding the design and implementation of new miniature exhibitions.
Loop Closure Through Vanishing Points in a Line-based Monocular SLAM

Vanishing points can provide absolute directional landmarks for mobile robots. This guiding ability is important in that the observation of the vanishing points is invariant with respect to the robot pose. In our system, loop closure is performed in two steps: first, the accumulated heading error is reduced using an observation of previously registered vanishing points, and second, the observation of known floor lines allows for further pose correction.

이전 1 2 3 4 5