• 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 Korean Institute of Communications and Information Sciences
• /
• v.25 no.4B
• /
• pp.763-770
• /
• 2000

In this paper, a new stereo image compression algorithm is suggested in which the residual image extracted from the stereo image by the disparity-compensated prediction method is compressed using the wavelet transform considering the inter & intra correlation between subbands. The compression performance of the proposed method is significantly improved by comparing with the conventional algorithm such as EPIC, EPWIC & JPEG through the computer simulation and the PSNR is also increased about 3.5dB compared with the EPIC. Finally, the stereo image having a good 3D effect can be reconstructed from the compressed image data by the proposed method.

• Journal of Korea Multimedia Society
• /
• v.19 no.6
• /
• pp.1003-1013
• /
• 2016

This paper proposes a framework of superpixel-based vehicle detection method using plane normal vector in disparity space. We utilize two common factors for detecting vehicles: Hypothesis Generation (HG) and Hypothesis Verification (HV). At the stage of HG, we set the regions of interest (ROI) by estimating the lane, and track them to reduce computational cost of the overall processes. The image is then divided into compact superpixels, each of which is viewed as a plane composed of the normal vector in disparity space. After that, the representative normal vector is computed at a superpixel-level, which alleviates the well-known problems of conventional color-based and depth-based approaches. Based on the assumption that the central-bottom of the input image is always on the navigable region, the road and obstacle candidates are simultaneously extracted by the plane normal vectors obtained from K-means algorithm. At the stage of HV, the separated obstacle candidates are verified by employing HOG and SVM as for a feature and classifying function, respectively. To achieve this, we trained SVM classifier by HOG features of KITTI training dataset. The experimental results demonstrate that the proposed vehicle detection system outperforms the conventional HOG-based methods qualitatively and quantitatively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.4A
• /
• pp.371-382
• /
• 2002

A camera model with optical axes parallel has been widely used for stereo vision applications. A pair of input ages are obtained from a convergent-looking stereo camera model based on the human visual system in this per, and each image is divided into quadrant regions with respect to the fixation point. The reasoning of quadrant partitions is based on the human visual system and is proven by a geometrical method. Image patches : constructed from the right and left stereo images. A modified cepstrum filter is applied to the patches and disparity vectors are determined by peak detection algorithm. The three-dimensional information for synthetic ages is obtained from the measured disparity and the convergent stereo camera model. It is shown that the experimental results of the proposed method for various stereo images are accurate around the fixation point like the human visual system.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.3
• /
• pp.1-10
• /
• 2008

An occupancy map is a representation methodology describing the region occupied by objects in 3D space, which can be utilized for autonomous navigation and object recognition. In this paper, we describe a technique for building an occupancy map using depth data extracted from stereo images. In addition, some techniques are proposed for utilizing the occupancy map for the segmentation of object regions. After the geometric information on the ground plane is extracted from a disparity image, the occupancy map is constructed by projecting each matched point to the ground plane-based 3D space. We explain techniques for extracting moving object regions using the occupancy map and present experimental results using real stereo images.

• Journal of Broadcast Engineering
• /
• v.19 no.2
• /
• pp.205-214
• /
• 2014

Minimization of visual fatigue is important for production of high quality stereoscopic 3D contents. Vertical disparity of stereo images occurred during contents production is considered as the main factor of visual fatigue. To ensure correct stereoscopy vertical disparity needs to be eliminated. In this paper, a method for automated image alignment was proposed for Stereoscopic 3D contents generation and post-processing steps. The proposed method consists of two parts: rectification for image alignment and camera motion detection. The proposed method showed that its rectification performance was the most superior among the existing methods tested and that camera motion detection had a success rate of 98.35%. Through these evaluations, we confirmed that the proposed method can be effectively applied to 3D contents production.

• Smart Media Journal
• /
• v.2 no.4
• /
• pp.27-33
• /
• 2013

Multi-view video system provides both realistic 3D feelings and free-view navigation. But it is hard to transmit too huge data, so we send only two or three view images and generate intermediate view image using depth information. In this paper, we propose high quality multi-view image generation method from stereoscopic images. Since the stereo matching method does not provide accurate disparity values for all the pixels, especially at the occlusion area, we propose an occlusion handling method using the background pixels at first. We also apply a joint bilateral filtering to enhance the disparity map at the object boundary since it can affect the quality of synthesized images significantly. Finally, we can generate virtual view images at intermediate view positions using confidence map to reduce bad pixel and hole's error. Experimental results show the proposed method performs better than the conventional method.

• Proceedings of the IEEK Conference
• /
• 2000.06d
• /
• pp.210-213
• /
• 2000

In the closed range space, the parallel two CCD cameras are used to acquire a pair of stereo image. The acquired stereo image are computed with Wavelet Transform repeatedly and including the low frequency component, the image size of those are reduced. It is the pyramid structure. The optimum matching point is searched to the pixel. Then appling the optimum matching point to DLT, it extract the three - dimensional surface coordinate from a stereo image. The direct linear transformation(DLT) method is used to calibrate the stereo camera compute the coordinate on a three dimensional space. To find the parameters for the DLT method, 30 control points which marked on the cylinder type object are used. To improve the matching algorithm, the paper select the pyramid structure for Wavelet Transform. The acquired disparity information is used to represent the really three-dimensional surface coordinate.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.241-243
• /
• 2010

Previous approaches to the 2D to 3D conversion problem require heavy computation or considerable amount of user input. In this paper, we propose a rather simple method in estimating the depth map from a single image using a monocular depth cue: haze. Using the haze imaging model, we obtain the distance information and estimate a reliable depth map from a single scenery image. Using the depth map, we also suggest an algorithm that converts the single image to 3D stereoscopic images. We determine a disparity value for each pixel from the original 'left' image and generate a corresponding 'right' image. Results show that the algorithm gives well refined depth maps despite the simplicity of the approach.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.6
• /
• pp.382-387
• /
• 2014

A computational approach to depth estimations using a color over lapped integral imaging system is presented. The proposed imaging system acquires multiple color images simultaneously through a single lens with an array of multiple pinholes that are distributed around the optical axis. This paper proposes a computational model of the relationship between the real distance of an object and the disparity among different color images. The proposed model can serve as a computational basis of a single camera-based depth estimation.