• Journal of the Korea Computer Graphics Society
• /
• v.27 no.3
• /
• pp.31-42
• /
• 2021

We propose a deep learning-based method that can estimate an appropriate lighting of both indoor and outdoor images. The method consists of two networks: Crop-to-PanoLDR network and LDR-to-HDR network. The Crop-to-PanoLDR network predicts a low dynamic range (LDR) environment map from a single partially observed normal field of view image, and the LDR-to-HDR network transforms the predicted LDR image into a high dynamic range (HDR) environment map which includes the high intensity light information. The HDR environment map generated through this process is applied when rendering virtual objects in the given image. The direction of the estimated light along with ambient light illuminating the virtual object is examined to verify the effectiveness of the proposed method. For this, the results from our method are compared with those from the methods that consider either indoor images or outdoor images only. In addition, the effect of the loss function, which plays the role of classifying images into indoor or outdoor was tested and verified. Finally, a user test was conducted to compare the quality of the environment map created in this study with those created by existing research.

• Journal of the Korea Computer Graphics Society
• /
• v.14 no.3
• /
• pp.11-17
• /
• 2008

Many methods which represent complex surfaces using height map without a number of vertex have been researched. However, a single layer height map cannot present more complex objects because it has only one height value on each position. In this paper, we introduce the new approach to render more complex objects, which are not generated by single layer height map, using multi layer height map. We store height values of the scene to each texture channel by the ascending order. A pair of ordered height values composes a geometry block and we use this property. For accurate ray search, we store the highest value in odd channels and the lowest value in even channels to generate quad tree height map. Our ray search algorithm shows accurate intersections between viewing ray and height values using quad tree height map. We solve aliasing problems on grazing angles occurred in previous methods and render the result scene on real-time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.10
• /
• pp.887-895
• /
• 2013

Metastatic bone cancer, the cancer which occurred in the various organs and progressively spread to bone, is one of the complications in cancer patients. This cancer is divided into the osteoblast and osteolytic metastasis. Although Computer Tomography(CT) could be an useful tool in diagnosis of bone metastasis, lesions are often missed by the visual inspection and it makes clinicians difficult to detect metastasis earlier. Therefore, in this study, we construct a three-dimensional(3D) volume rendering data from tomography images of the chest CT, and apply a 3D based image processing algorithm to them for detection bone metastasis lesions. Then we perform a three-dimensional visualization of the detected lesions.From our test using 10 clinical cases, we confirmed 94.1% of average sensitivity for osteoblast, and 90.0% of average sensitivity, respectively. Consequently, our findings showed a promising possibility and potential usefulness in diagnosis of metastastic bone cancer.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.2
• /
• pp.37-44
• /
• 2020

In this paper, we proposed that a fully automatic multi-class whole heart segmentation algorithm using deep learning. The proposed method is based on U-Net architecture which consist of recurrent convolutional block, residual multi-dilated convolutional block. The evaluation was accomplished by comparing automated analysis results of the test dataset to the manual assessment. We obtained the average DSC of 96.88%, precision of 95.60%, and recall of 97.00% with CT images. We were able to observe and analyze after visualizing segmented images using three-dimensional volume rendering method. Our experiment results show that proposed method effectively performed to segment in various heart structures. We expected that our method can help doctors and radiologist to make image reading and clinical decision.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.3
• /
• pp.101-109
• /
• 2018

Recently, the improvement of virtual reality and augmented reality technologies leverages many new technologies like the virtual study room, virtual architecture room. Such virtual worlds require free handed drawing technology such as writing descriptions of formula or drawing blue print of buildings. In nature, lots of view point modifications occur when we walk around inside the virtual world. Especially, we often look some objects from near to far distance in the virtual world. Traditional drawing methods like using fixed size image for drawing unit is not produce acceptable result because they generate blurred and jaggy result as view distance varying. We propose a novel method which robust to the environment that produce lots of magnifications and minimizations like the virtual reality world. We implemented our algorithm both two dimensional and three dimensional devices. Our algorithm does not produce any artifacts, jaggy or blurred result regardless of scaling factor.

• Progress in Medical Physics
• /
• v.17 no.3
• /
• pp.167-172
• /
• 2006

Catheter fragment and embolism are both potentially serious complications associated with the use of an intravenous (IV) catheter for contrast media bolus injection, and may be followed by serious or lethal sequelae. Though catheter fragment is a rare complication of IV catheter insertion, especially in peripheral veins, CT can be used to detect residual fragment. This study demonstrates the utility of MDCT to localize a small, subtle peripheral venous catheter, which can be easily reformatted of MDCT reformations. Various 3D techniques such as MPR and MIP, volume rendering, and shaded-surface displays are currently available for reconstructing MDCT data. Advances in MDCT technology contribute substantially to the detection and accurate localization of smaller IV catheter fragment.

• Journal of Korea Multimedia Society
• /
• v.12 no.9
• /
• pp.1262-1272
• /
• 2009

In computer graphics, there exist many studies about illumination and radiance for a realistic description of the 3D modeling and rendering. When we see a scene, the scene is lit by a source of light and the radiance of the points by a source in the scene. The radiance has direct light and glight component. The direct light gets lights directly from light source, but the global light gets lights indirectly by interreflections among complicated geometrical components. In this paper, we studied a method for increasing the accuracy of separating direct light and global light components from a scene by using high frequency illumination pattern. For experiments, we applied the separating method of Nayar's and found the best configurations for the separation through the experiments. We improved the separation accuracy of direct and global light by measuring the value of unilluminated area, which depends on the characteristics of object. Furthermore, we enhanced invisible scene of the global light by applying the image filtering technique.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.3
• /
• pp.692-698
• /
• 2016

Recently, the facility extension construction and maintenance market portion has increased instead of decreased the newly facility construction. In this context, it is important to examine the reverse engineering of MEP (Mechanical Electrical and Plumbing) facilities, which have the high operation and management cost in the architecture domains. The purpose of this study was to suggest the Large Point Cloud-based Pipe Shape Reverse Engineering Method. To conduct the study, the related researches were surveyed and the reverse engineering automation method of the pipe shapes considering large point cloud was proposed. Based on the method, the prototype was developed and the results were validated. The proposed method is suitable for large data processing considering the validation results because the rendering performance standard deviation related to the 3D point cloud massive data searching was 0.004 seconds.

• KIISE Transactions on Computing Practices
• /
• v.22 no.3
• /
• pp.157-162
• /
• 2016

The limitation of a human's visual perception is considered to be 60 frames per second, This study investigated the effects of fast update rates (above 60 fps) in terms of interaction accuracy. Initial experiments showed that the interaction accuracy increased at rates faster than 60 fps. We assumed that either or both of the following two situations would cause such an effect: the user could recognize rendering rates faster than 60 fps, or the input processing rates were significant for the high accuracy. To evaluate the significance of these events, we conducted a second and third experiment. Although the display refresh rate was also fixed at 60 fps (by disabling the vertical sync), the rendered image actually differed for 60 fps and 150 fps. This research shows that faster update rate is necessary to achieve high interaction accuracy, and its limit is far over the usually considered 60 fps.

• Journal of Broadcast Engineering
• /
• v.11 no.2 s.31
• /
• pp.164-173
• /
• 2006

The data structure of the concentric mosaic can be regarded as a video sequence with a slowly panning camera. We take a concentric mosaic with match or alignment of video sequences. Also the concentric mosaic required for huge memory. Thus, compressing is essential in order to use the concentric mosaic. Therefore we need the algorithm that compressed data structure was maintained and the scene was decoded. In this paper, we used 3D lifting transform to compress concentric mosaic. Lifting transform has a merit of wavelet transform and reduces computation quantities and memory. Because each frame has high correlation, the complexity which a scene is detected form 3D transformed bitstream is increased. Thus, in order to have higher performance and decrease the complexity of detecting of a scene we executed 3D lifting and then transformed data set was sequently compressed with each frame unit. Each frame has a flexible bit rate. Also, we proposed the algorithm that compressed data structure was maintained and the scene was decoded by using property of lifting structure.