• Journal of the Semiconductor & Display Technology
• /
• v.15 no.3
• /
• pp.67-71
• /
• 2016

This paper addresses a problem of defocus map recovery from single image. We describe a simple effective approach to estimate the spatial value of defocus blur at the edge location of the image. At first, we perform a re-blurring process using Gaussian function with input image, and calculate a gradient magnitude ratio with blurring amount between input image and re-blurred image. Then we get a full defocus map by propagating the blur amount at the edge location. Experimental result reveals that our method outperforms a reliable estimation of depth map, and shows that our algorithm is robust to noise, inaccurate edge location and interferences of neighboring edges within input image.

• Journal of IKEEE
• /
• v.28 no.1
• /
• pp.65-71
• /
• 2024

This paper proposes a neural network that can reduce the number of parameters while reducing matching errors in occluded regions to increase the accuracy of depth maps in stereo matching. Stereo matching-based object recognition is utilized in many fields to more accurately recognize situations using images. When there are many objects in a complex image, an occluded area is generated due to overlap between objects and occlusion by background, thereby lowering the accuracy of the depth map. To solve this problem, existing research methods that create context information and combine it with the cost volume or RoIselect in the occluded area increase the complexity of neural networks, making it difficult to learn and expensive to implement. In this paper, we create a depthwise seperable neural network that enhances regional feature extraction before cost volume generation, reducing the number of parameters and proposing a neural network that is robust to occlusion errors. Compared to PSMNet, the proposed neural network reduced the number of parameters by 30%, improving 5.3% in color error and 3.6% in test loss.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.3_4
• /
• pp.138-148
• /
• 2003

Orthogonal Cross Cylinder mapping and segmentation based modeling methods have been implemented for constructing the image-based navigation system in this paper. The Orthogonal Cross Cylinder (OCC) is the object expressed by the intersection area that occurs when a cylinder is orthogonal with another. OCC mapping method eliminates the singularity effect caused in the environment maps and shows an almost even amount of area for the environment occupied by a single texel. A full-view image from a fixed point-of-view can be obtained with OCC mapping although it becomes difficult to express another image when the point-of-view has been changed. The OCC map is segmented according to the objects that form the environment and the depth value is set by the characteristics of the classified objects for the segmentation based modeling. This method can easily be implemented on an environment map and makes the environment modeling easier through extracting the depth value by the image segmentation. An environment navigation system with a full-view can be developed with these methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.154-156
• /
• 2022

본 논문은 카메라 어레이기반 실사 다시점 입체영상을 획득·생성하기 위한 워크플로우를 제시하고 이를 검증하기 위한 실험 결과를 소개한다. 구체적으로, 액션 캠 기반 수렴형 리그 구조, 획득 동기화, 카메라 캘리브레이션, 깊이 맵 추출을 포함하는 일련의 과정 및 이에 대한 검증으로 실내외 2종의 콘텐츠의 획득 실험 결과를 기술한다.

• Journal of Broadcast Engineering
• /
• v.13 no.6
• /
• pp.859-874
• /
• 2008

In this paper, we develop a fast view synthesis method that generates multiple intermediate views in real-time for the 3D display system when the camera geometry and depth map of reference views are given in advance. The proposed method achieves faster view synthesis than previous approaches in GPU by processing in parallel the entire computations required for the view synthesis. Specifically, we use $CUDA^{TM}$ (by NVIDIA) to control GPU device. For increasing the processing speed, we adapted all the processes for the view synthesis to single instruction multiple data (SIMD) structure that is a main feature of CUDA, maximized the use of the high-speed memories on GPU device, and optimized the implementation. As a result, we could synthesize 9 intermediate view images with the size of 720 by 480 pixels within 0.128 second.

• Journal of Korea Multimedia Society
• /
• v.16 no.1
• /
• pp.11-18
• /
• 2013

This paper presents a SFF(shape from focus) algorithm using a new focus measure based on higher order statistics for the exact depth estimation. Since conventional SFF-based 3D depth reconstruction algorithms used SML(sum of modified Laplacian) as the focus measure, their performance is strongly depended on the image characteristics. These are efficient only for the rich texture and well focused images. Therefore, this paper adopts a new focus measure using HOS(higher order statistics), in order to extract the focus value for relatively poor texture and focused images. The initial best focus area map is generated by the measure. Thereafter, the area refinement, thinning, and corner detection methods are successively applied for the extraction of the locally best focus points. Finally, a 3D model from the carefully selected points is reconstructed by Delaunay triangulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.819-824
• /
• 2021

As technologies related to image processing such as autonomous driving and CCTV develop, fog removal algorithms using a single image are being studied to improve the problem of image distortion. As a method of predicting fog density, there is a method of estimating the depth of an image by generating a depth map, and various fog features may be used as training data of the depth map. In addition, it is essential to implement a hardware capable of processing high-definition images in real time in order to apply the fog removal algorithm to actual technologies. In this paper, we implement NLCV (Normalize Local Coefficient of Variation), a feature of fog based on coefficient of variation, in hardware. The proposed hardware is an FPGA implementation of Xilinx's xczu7ev-2ffvc1156 as a target device. As a result of synthesis through the Vivado program, it has a maximum operating frequency of 479.616MHz and shows that real-time processing is possible in 4K UHD environment.

• Journal of Broadcast Engineering
• /
• v.17 no.3
• /
• pp.437-446
• /
• 2012

A face detection algorithms using two-dimensional (2-D) intensity or color images have been studied for decades. Recently, with the development of low-cost range sensor, three-dimensional (3-D) information (i.e., depth image that represents the distance between a camera and objects) can be easily used to reliably extract facial features. Most people have a similar pattern of 3-D facial structure. This paper proposes a face detection method using intensity and depth images. At first, adaboost algorithm using intensity image classifies face and nonface candidate regions. Each candidate region is divided into $5{\times}5$ blocks and depth values are averaged in each block. Then, $5{\times}5$ block rank pattern is constructed by sorting block averages of depth values. Finally, candidate regions are classified as face and nonface regions by matching the constructed depth map based block rank patterns and a template pattern that is generated from training data set. For template matching, the $5{\times}5$ template block rank pattern is prior constructed by averaging block ranks using training data set. The proposed algorithm is tested on real images obtained by Kinect range sensor. Experimental results show that the proposed algorithm effectively eliminates most false positives with true positives well preserved.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.911-921
• /
• 2020

Recently, MPEG-I (Immersive) has been exploring compression performance through standardization projects for immersive video. The MPEG Immersion Video (MIV) standard technology is intended to provide limited 6DoF based on depth map-based image rendering (DIBR). MIV is a model that processes the Basic View and the residual information into an Additional View, which is a collection of patches. Atlases have the unique characteristics depending on the kind of the view they are included, requiring consideration of the compression efficiency. In this paper, the performance comparison analysis of screen content coding tools such as intra block copy (IBC) is conducted, based on the pattern of various views and patches repetition. It is demonstrated that the proposed method improves coding performance around -15.74% BD-rate reduction in the MIV.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.9
• /
• pp.1214-1219
• /
• 2018

A variety of VR contents are being introduced as of 2017. VR contents are concentrated in the genre of games and interactive because of the difficulty of $360^{\circ}$ shooting production environment. Live action $360^{\circ}$ VR content has many problems due to the difficulty of the production environment. In this paper, a three - dimensional information value is generated in binocular disparity of a real image by using a re-light technique based on real image data. The generated 3D information values are combined with a technique of converting the depth information into a depth map and a re-light technique by installing virtual lighting on the surface formed in the 3D space. In order to solve the problem of lighting exposure, we apply the technique of re-lighting to the VR production pipeline by comparing and analyzing the result image of actual image and virtual image data.