• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.6
• /
• pp.1-7
• /
• 2011

The depth camera measures range information of the scene in real time using Time-of-Flight (TOF) technology. Measured depth data is then regularized and provided as a depth image. This depth image is utilized with the stereo or multi-view image to generate high-resolution depth map of the scene. However, it is required to correct noise and distortion of TOF depth image due to the technical limitation of the TOF depth camera. The corrected depth image is combined with the color image in various methods, and then we obtain the high-resolution depth of the scene. In this paper, we introduce the principal and various techniques of sensor fusion for high-quality depth generation that uses multiple camera with depth cameras.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.3
• /
• pp.1302-1309
• /
• 2012

In this paper, we propose applying differential techniques for 2D/3D video conversion to the objects grouped by depth information. One of the problems converting 2D images to 3D images using the technique tracking the motion of pixels is that objects not moving between adjacent frames do not give any depth information. This problem can be solved by applying relative height cue only to the objects which have no moving information between frames, after the process of splitting the background and objects and extracting depth information using motion vectors between objects. Using this technique all the background and object can have their own depth information. This proposed method is used to generate depth map to generate 3D images using DIBR(Depth Image Based Rendering) and verified that the objects which have no movement between frames also had depth information.

• The KIPS Transactions:PartD
• /
• v.12D no.6 s.102
• /
• pp.807-816
• /
• 2005

Sequential pattern mining finds all of the frequent sequences satisfying a minimum support threshold in a large database. However, when mining long frequent sequences, or when using very low support thresholds, the performance of currently reported algorithms often degrades dramatically. In this paper, we propose a novel sequential pattern algorithm using only closed frequent sequences which are small subset of very large frequent sequences. Our algorithm generates the candidate sequences by depth-first search strategy in order to effectively prune. using bitmap representation of underlying databases, we can effectively calculate supports in terms of bit operations and prune sequences in much less time. Performance study shows that our algorithm outperforms the previous algorithms.

• Journal of Broadcast Engineering
• /
• v.15 no.6
• /
• pp.723-730
• /
• 2010

Recently, diverse 3D image processing technologies have been applied in industries. Among them, stereoscopic conversion is a technology to generate a stereoscopic image from a conventional 2D image. The technology can be applied to movie and broadcasting contents and the viewer can watch 3D stereoscopic contents. Further the stereoscopic conversion is required to be applied to other fields. Following such trend, the aim of this paper is to apply the stereoscopic conversion to medical fields. The medical images can deliver more detailed 3D information with a stereoscopic image compared with a 2D plane image. This paper presents a novel methodology for converting a 2D medical image into a 3D stereoscopic image. For this, mean shift segmentation, edge detection, intensity analysis, etc are utilized to generate a final depth map. From an image and the depth map, left and right images are constructed. In the experiment, the proposed method is performed on a medical image such as CT (Computed Tomograpy). The stereoscopic image displayed on a 3D monitor shows a satisfactory performance.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.1-8
• /
• 2008

The belief propagation method that has been studied recently yields good performance in disparity extraction. The method in which a target function is modeled as an energy function based on Markov random field(MRF), solves the stereo matching problem by finding the disparity to minimize the energy function. MRF models provide robust and unified framework for vision problem such as stereo and image restoration. the belief propagation method produces quite correct results, but it has difficulty in real time implementation because of higher computational complexity than other stereo methods. To relieve this problem, in this paper, we propose a fast algorithm of the belief propagation method. Energy function consists of a data term and a smoothness tern. The data term usually corresponds to the difference in brightness between correspondences, and smoothness term indicates the continuity of adjacent pixels. Smoothness information is created from messages, which are assigned using four different message arrays for the pixel positions adjacent in four directions. The processing time for four message arrays dominates 80 percent of the whole program execution time. In the proposed method, we propose an algorithm that dramatically reduces the processing time require in message calculation, since the message.; are not produced in four arrays but in a single array. Tn the last step of disparity extraction process, the messages are called in the single integrated array and this algorithm requires 1/4 computational complexity of the conventional method. Our method is evaluated by comparing the disparity error rates of our method and the conventional method. Experimental results show that the proposed method remarkably reduces the execution time while it rarely increases disparity error.

• The Journal of the Korea Contents Association
• /
• v.18 no.11
• /
• pp.447-454
• /
• 2018

This paper proposes a depth image generation algorithm of stereo images using a deep learning model composed of a CNN (convolutional neural network). The proposed algorithm consists of a feature extraction unit which extracts the main features of each parallax image and a depth learning unit which learns the parallax information using extracted features. First, the feature extraction unit extracts a feature map for each parallax image through the Xception module and the ASPP(Atrous spatial pyramid pooling) module, which are composed of 2D CNN layers. Then, the feature map for each parallax is accumulated in 3D form according to the time difference and the depth image is estimated after passing through the depth learning unit for learning the depth estimation weight through 3D CNN. The proposed algorithm estimates the depth of object region more accurately than other algorithms.

• Journal of Advanced Navigation Technology
• /
• v.27 no.2
• /
• pp.182-189
• /
• 2023

Depth estimation is a key technology in 3D map generation for autonomous driving of vehicles, robots, and drones. The existing sensor-based method has high accuracy but is expensive and has low resolution, while the camera-based method is more affordable with higher resolution. In this study, we propose self-attention-based unsupervised monocular depth estimation for UAV camera system. Self-Attention operation is applied to the network to improve the global feature extraction performance. In addition, we reduce the weight size of the self-attention operation for a low computational amount. The estimated depth and camera pose are transformed into point cloud. The point cloud is mapped into 3D map using the occupancy grid of Octree structure. The proposed network is evaluated using synthesized images and depth sequences from the Mid-Air dataset. Our network demonstrates a 7.69% reduction in error compared to prior studies.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.12-13
• /
• 2010

본 논문에서는 USB(Universal Serial Bus) 인터페이스를 이용하여 PC와 하드웨어 사이의 영상을 송수신 하는 시스템을 제안한다. 구현된 시스템은 PC에서 스테레오 캠으로부터 획득된 좌안, 우안 영상을 USB 인터페이스를 이용하여 고속으로 하드웨어에 전송해주고, 하드웨어에서 생성된 깊이 맵을 고속으로 전송받도록 구성되어 있다. Cypress사의 USB2.0 컨트롤러 칩인 CY68013A를 사용하여 구현하였으며, USB칩과 FPGA와의 인터페이스는 GPIF(General Programmable Interface)를 이용하여 병목현상이 없이, 고속의 데이터 전송을 달성하도록 하였다.

• Journal of Broadcast Engineering
• /
• v.16 no.1
• /
• pp.33-43
• /
• 2011

Recently, 3D displays are supplied in the 3D markets so that the demand for 3D stereoscopic contents increases. In general, a simple method is to use a stereoscopic camera. As well, the production of 3D from 2D materials is regarded as an important technology. Such conversion works have gained much interest in the field of 3D converting. However, the stereoscopic image generation from a single 2D image is limited to simple 2D to 3D conversion so that the better realistic perception is difficult to deliver to the users. This paper presents a new stereoscopic content production method where foreground objects undergo alive action events. Further stereoscopic animation is viewed on 3D displays. Given a 2D image, the production is composed of background image generation, foreground object extraction, object/background depth maps and stereoscopic image generation The alive objects are made using the geometric transformation (e.g., translation, rotation, scaling, etc). The proposed method is performed on a Korean traditional painting, Danopungjung as well as Pixar's Up. The animated video showed that through the utilization of simple object transformations, more realistic perception can be delivered to the viewers.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10c
• /
• pp.510-515
• /
• 2007

2차원 영상으로부터 3차원 정보를 획득하기 위해서는 disparity map의 정확한 계산이 요구된다. Disparity map을 구하기 위한 기존의 알고리즘은 크게 상관도 기반 방법과 특징 기반 방법으로 분류되는데, 본 논문에서는 이들 각 방법에 대한 분석을 통해서 좀 더 정확한 disparity map을 구하는 방법을 모색한다. 이를 위해 스테레오 카메라로부터 획득된 2차원 영상에서 건물에 대한 깊이 정보 추출을 위해 SIFT 기법을 이용한 disparity map 생성 알고리즘을 제안한다. 제안된 기법은 수정된 MAD인 MMAD(Modified Mean of Absolute Differences) 알고리즘을 새로 제안하여 영역 기반의 유사도 측정을 기반으로 하면서 특징 기반 방법의 하나인 SIFT를 적용하여 거짓 정합(false matching)에 의한 에러를 줄이고 폐색(occlusion) 영역에 대한 오류를 보정한 disparity map을 생성하는데 초점을 둔다.