• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.2
• /
• pp.97-101
• /
• 2013

Bit-depth expansion is a process of enhancing the image quality by increasing the number of intensity levels. To solve this problem, a hybrid method is proposed, where the pixels are categorized into smooth and complex regions, and are processed using different methods. The pixels in the smooth region are reconstructed with a smooth prior, and a Bayesian estimator is used for the pixels in the complex region. The proposed method effectively removes the false contour artifacts while requiring less computation than conventional methods. In addition, the method shows good quantitative performance, and the PSNR gains over the best existing method are 1.45 dB and 0.26 dB for 4 bits and 3 bits expansion cases, respectively.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.10
• /
• pp.1005-1013
• /
• 2008

This paper describes a new sensor system for 3D range measurement using the structured infrared light. Environment and obstacle sensing is the key issue for mobile robot localization and navigation. Laser scanners and infrared scanners cover $180^{\circ}$ and are accurate but too expensive. Those sensors use rotating light beams so that the range measurements are constrained on a plane. 3D measurements are much more useful in many ways for obstacle detection, map building and localization. Stereo vision is very common way of getting the depth information of 3D environment. However, it requires that the correspondence should be clearly identified and it also heavily depends on the light condition of the environment. Instead of using stereo camera, monocular camera and the projected infrared light are used in order to reduce the effects of the ambient light while getting 3D depth map. Modeling of the projected light pattern enabled precise estimation of the range. Identification of the cells from the pattern is the key issue in the proposed method. Several methods of correctly identifying the cells are discussed and verified with experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• pp.800-816
• /
• 2018

In recent years, video surveillance research has been able to recognize various behaviors of pedestrians and analyze the overall situation of objects by combining image analysis technology and deep learning method. Human Activity Recognition (HAR), which is important issue in video surveillance research, is a field to detect abnormal behavior of pedestrians in CCTV environment. In order to recognize human behavior, it is necessary to detect the human in the image and to estimate the pose from the detected human. In this paper, we propose a novel approach for 2D Human Pose Estimation based on object detection using RGB-D information. By adding depth information to the RGB information that has some limitation in detecting object due to lack of topological information, we can improve the detecting accuracy. Subsequently, the rescaled region of the detected object is applied to ConVol.utional Pose Machines (CPM) which is a sequential prediction structure based on ConVol.utional Neural Network. We utilize CPM to generate belief maps to predict the positions of keypoint representing human body parts and to estimate human pose by detecting 14 key body points. From the experimental results, we can prove that the proposed method detects target objects robustly in occlusion. It is also possible to perform 2D human pose estimation by providing an accurately detected region as an input of the CPM. As for the future work, we will estimate the 3D human pose by mapping the 2D coordinate information on the body part onto the 3D space. Consequently, we can provide useful human behavior information in the research of HAR.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.109-111
• /
• 2021

Recently, computer vision research using light field cameras has been actively conducted. Since light field cameras have spatial information, various studies are being conducted in fields such as depth map estimation, super resolution, and 3D object detection. In this paper, we propose a method for detecting objects in blur images through a 7×7 array of images acquired through a light field camera. The blur image, which is weak in the existing camera, is detected through the light field camera. The proposed method uses the SSD algorithm to evaluate the performance using blur images acquired from light field cameras.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.7
• /
• pp.634-640
• /
• 2015

This paper proposes a novel three-dimensional mapping algorithm in Omni-Directional Vision SLAM based on a fisheye image and laser scanner data. The performance of SLAM has been improved by various estimation methods, sensors with multiple functions, or sensor fusion. Conventional 3D SLAM approaches which mainly employed RGB-D cameras to obtain depth information are not suitable for mobile robot applications because RGB-D camera system with multiple cameras have a greater size and slow processing time for the calculation of the depth information for omni-directional images. In this paper, we used a fisheye camera installed facing downwards and a two-dimensional laser scanner separate from the camera at a constant distance. We calculated fusion points from the plane coordinates of obstacles obtained by the information of the two-dimensional laser scanner and the outline of obstacles obtained by the omni-directional image sensor that can acquire surround view at the same time. The effectiveness of the proposed method is confirmed through comparison between maps obtained using the proposed algorithm and real maps.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.67-80
• /
• 2017

This paper presents a hole-filling method using extrapolated spatio-temporal background information to obtain a synthesized view. A new temporal background model using non-overlapped patch based background codebook is introduced to extrapolate temporal background information In addition, a depth-map driven spatial local background estimation is addressed to define spatial background constraints that represent the lower and upper bounds of a background candidate. Background holes are filled by comparing the similarities between the temporal background information and the spatial background constraints. Additionally, a depth map-based ghost removal filter is described to solve the problem of the non-fit between a color image and the corresponding depth map of a virtual view after 3-D warping. Finally, an inpainting is applied to fill in the remaining holes with the priority function that includes a new depth term. The experimental results demonstrated that the proposed method led to results that promised subjective and objective improvement over the state-of-the-art methods.

• Journal of Sensor Science and Technology
• /
• v.29 no.2
• /
• pp.123-127
• /
• 2020

In this paper, the estimation of the disparity for depth extraction in monochrome complementary metal-oxide-semiconductor (CMOS) image sensors with offset pixel apertures is presented. To obtain the depth information, the disparity information between two different channel data of the offset pixel apertures is required. The disparity is caused by the difference in the response angle between the left- and right-offset pixel aperture images. A depth map is implemented by the generated disparity. Therefore, the disparity is the most important factor for realizing 3D images from the designed CMOS image sensor with offset pixel apertures. The disparity is influenced by the pixel height and offset value of the offset pixel aperture. To confirm this correlation, the offset value is set to maximum within the pixel area, and the disparity values corresponding to the difference in the heights are calculated and compared. The disparity is derived using the camera-lens formula. Two monochrome CMOS image sensors with offset pixel apertures are used in the disparity estimation.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.240-243
• /
• 2018

최근 VR(Virtual Reality)와 AR(Augmented Reality)의 발전에 따라 영상 또는 이미지에서 카메라와 물체 사이의 거리를 추정하는 기술에 대한 연구가 활발히 진행되고 있다. 본 논문에서는 카메라와 물체 사이의 거리 추정 방법 중에서 단일 카메라를 이용하여 촬영한 이미지의 흐림 정도를 분석하여 3D 거리를 추정하는 알고리즘을 연구한다. 특히 고가의 렌즈가 장착된 DSLR 카메라가 아닌 스마트폰 카메라 이미지에서 DFD를 이용한 거리 추정 방법 중 1개의 이미지를 이용한 3D 거리 추정 방법과 초점이 서로 다른 2개의 이미지를 결합하여 3D 거리를 추정하는 방법을 연구하고 최적회된 피사체 범위에 대해 연구하였다. 한 개의 이미지를 이용한 거리 추정에서는 카메라의 초점 거리를 200 mm로 설정할 때, 두 개의 이미지를 이용한 거리 추정에서는 두 이미지의 초점 거리를 각각 150 mm, 250 mm로 설정했을 때 가장 넓은 거리 추정 범위를 갖는다. 또한, 두 거리 추정 방법 모두 초점 거리가 가까울수록 가까운 물체의 거리 추정에 효율적인 것으로 나타났다.

• Journal of the HCI Society of Korea
• /
• v.1 no.1
• /
• pp.53-61
• /
• 2006

In this paper, we propose a novel method for 3D reconstruction of an indoor scene using a multi-view camera. Until now, numerous disparity estimation algorithms have been developed with their own pros and cons. Thus, we may be given various sorts of depth images. In this paper, we deal with the generation of a 3D surface using several 3D point clouds acquired from a generic multi-view camera. Firstly, a 3D point cloud is estimated based on spatio-temporal property of several 3D point clouds. Secondly, the evaluated 3D point clouds, acquired from two viewpoints, are projected onto the same image plane to find correspondences, and registration is conducted through minimizing errors. Finally, a surface is created by fine-tuning 3D coordinates of point clouds, acquired from several viewpoints. The proposed method reduces the computational complexity by searching for corresponding points in 2D image plane, and is carried out effectively even if the precision of 3D point cloud is relatively low by exploiting the correlation with the neighborhood. Furthermore, it is possible to reconstruct an indoor environment by depth and color images on several position by using the multi-view camera. The reconstructed model can be adopted for interaction with as well as navigation in a virtual environment, and Mediated Reality (MR) applications.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1321-1330
• /
• 2023

This study attempts to address the problem of 3D pose estimation for multiple human objects through a single image generated during the character development process that can be used in augmented reality. In the existing top-down method, all objects in the image are first detected, and then each is reconstructed independently. The problem is that inconsistent results may occur due to overlap or depth order mismatch between the reconstructed objects. The goal of this study is to solve these problems and develop a single network that provides consistent 3D reconstruction of all humans in a scene. Integrating a human body model based on the SMPL parametric system into a top-down framework became an important choice. Through this, two types of collision loss based on distance field and loss that considers depth order were introduced. The first loss prevents overlap between reconstructed people, and the second loss adjusts the depth ordering of people to render occlusion inference and annotated instance segmentation consistently. This method allows depth information to be provided to the network without explicit 3D annotation of the image. Experimental results show that this study's methodology performs better than existing methods on standard 3D pose benchmarks, and the proposed losses enable more consistent reconstruction from natural images.