• Journal of the Korea Computer Graphics Society
• /
• v.23 no.3
• /
• pp.31-38
• /
• 2017

In this paper, we propose a system which estimates Manhattan coordinate systems for urban scene images using a convolutional neural network (CNN). Estimating the Manhattan coordinate system from an image under the Manhattan world assumption is the basis for solving computer graphics and vision problems such as image adjustment and 3D scene reconstruction. We construct a CNN that estimates Manhattan coordinate systems based on GoogLeNet [1]. To train the CNN, we collect about 155,000 images under the Manhattan world assumption by using the Google Street View APIs and calculate Manhattan coordinate systems using existing calibration methods to generate dataset. In contrast to PoseNet [2] that trains per-scene CNNs, our method learns from images under the Manhattan world assumption and thus estimates Manhattan coordinate systems for new images that have not been learned. Experimental results show that our method estimates Manhattan coordinate systems with the median error of $3.157^{\circ}$ for the Google Street View images of non-trained scenes, as test set. In addition, compared to an existing calibration method [3], the proposed method shows lower intermediate errors for the test set.

• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.567-575
• /
• 2021

In order to correct the provided RPC and DEM generated from the high-resolution satellite images, the acquisition of the ground control point (GCP) must be preceded. This task is a very complicate that requires field surveys, GPS surveying, and image coordinate reading corresponding to GCPs. In addition, since it is difficult to set up and measure a GCP in areas where access is difficult or impossible (tidal flats, polar regions, volcanic regions, etc.), an alternative method is needed. In this paper, we propose a 3D surface matching technique using only the established ground coordinate package, avoiding the ground-image-location survey of the GCP to correct the DEM produced from WorldView-2 satellite images and the provided RPCs. The location data of the public control points were obtained from the National Geographic Information Institute website, and the DEM was corrected by performing 3D surface matching with this package. The accuracy of 3-axis translation and rotation obtained by the matching was evaluated using pre-measured GPS checkpoints. As a result, it was possible to obtain results within 2 m in the plane location and 1 m in height.

• Korean Journal of Remote Sensing
• /
• v.31 no.2
• /
• pp.111-125
• /
• 2015

In this study, the target detection using both high-resolution satellite SAR and Elecro-Optical (EO) images such as TerraSAR-X and WorldView-2 is performed, considering the characteristics of targets. The targets of our interest are featured by being stationary and appearing as cluster targets. After the target detection of SAR image by using Constant False Alarm Rate (CFAR) algorithm, a series of processes is performed in order to reduce false alarms, including pixel clustering, network clustering and coherence analysis. We extend further our algorithm by adopting the fast and effective ellipse detection in EO image using randomized hough transform, which is significantly reducing the number of false alarms. The performance of proposed algorithm has been tested and analyzed on TerraSAR-X SAR and WordView-2 EO images. As a result, the average false alarm for group of targets is 1.8 groups/$64km^2$ and the false alarms of single target range from 0.03 to 0.3 targets/$km^2$. The results show that groups of targets are successfully identified with very low false alarms.

• Journal of International Society for Simulation Surgery
• /
• v.3 no.2
• /
• pp.39-45
• /
• 2016

The augmented reality is the environment which consists of real-world view and information drawn by computer. Since the image which user can see through augmented reality device is a synthetic image composed by real-view and virtual image, it is important to make the virtual image generated by computer well harmonized with real-view image. In this paper, we present reviews of several works about computer vision and graphics methods which give user realistic augmented reality experience. To generate visually harmonized synthetic image which consists of a real and a virtual image, 3D geometry and environmental information such as lighting or material surface reflectivity should be known by the computer. There are lots of computer vision methods which aim to estimate those. We introduce some of the approaches related to acquiring geometric information, lighting environment and material surface properties using monocular or multi-view images. We expect that this paper gives reader's intuition of the computer vision methods for providing a realistic augmented reality experience.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.34 no.2
• /
• pp.121-132
• /
• 2016

This study proposes an approach for simulating high spatial resolution satellite images acquired under arbitrary sun-sensor geometry using existing images and 3D (three-dimensional) data. First, satellite images, having significant differences in spectral regions compared with those in the simulated image were transformed to the same spectral regions as those in simulated image by using the UPDM (Universal Pattern Decomposition Method). Simultaneously, shadows cast by buildings or high features under the new sun position were modeled. Then, pixels that changed from shadow into non-shadow areas and vice versa were simulated on the basis of existing images. Finally, buildings that were viewed under the new sensor position were modeled on the basis of open library-based 3D reconstruction program. An experiment was conducted to simulate WV-3 (WorldView-3) images acquired under two different sun-sensor geometries based on a Pleiades 1A image, an additional WV-3 image, a Landsat image, and 3D building models. The results show that the shapes of the buildings were modeled effectively, although some problems were noted in the simulation of pixels changing from shadows cast by buildings into non-shadow. Additionally, the mean reflectance of the simulated image was quite similar to that of actual images in vegetation and water areas. However, significant gaps between the mean reflectance of simulated and actual images in soil and road areas were noted, which could be attributed to differences in the moisture content.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2008.10a
• /
• pp.51-54
• /
• 2008

In this paper, we propose a practical method for displaying 2D/True3D mixed contents in real-time. Many companies released their 3D display recently, but the costs of producing True3D contents are still very expensive. Since there are already a lot of 2D contents in the world and it is more effective to mix True3D objects into the 2D contents than making True3D contents directly, people became interested in mixing 2D/True3D contents. Moreover, real-time 2D/True3D mixing is helpful for 3D displays because the scenario of the contents can be easily changed on playback-time by adjusting the 3D effects and the motion of the True3D object interactively. In our system, True3D objects are rendered into multiple view-point images, which are composed with 2D contents by using depth information, and then they are multiplexed with pre-generated view masks. All the processes are performed on a graphics processor. We were still able to play a 2D/True3D mixed contents with Full HD resolution in real-time using a normal graphics processor.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.1141-1144
• /
• 2005

In this paper, we propose an automatic displacement system for testing stability of structure. Photogrammetry is a method which can measure accurate 3D data from 2D images taken from different locations and which is suitable for analyzing and measuring the displacement of structure. This paper consists of camera calibration, feature extraction using coded target & retro-reflective circle, 3D reconstruction and analyzing accuracy. Multi-view camera which is used for measuring displacement of structure is placed with different location respectively. Camera calibration calculates trifocal tensor from corresponding points in images, from which Euclidean camera is calculated. Especially, in a step of feature extraction, we utilize sub-pixel method and pattern recognition in order to measure the accurate 3D locations. Scale bar is used as reference to measure. the accurate value of world coordinate..

• The Journal of the Korea Contents Association
• /
• v.14 no.2
• /
• pp.72-83
• /
• 2014

Film is a visual art. A process of film totally depends on our sense of sight. It indicates that a way of delivering narrative in the film is "showing", not "telling". There has already been established theories about "who sees" and "who tells" in narratology. It explains who the narrator is and how the narrator delivers in literary works which is represented in terms of "Point-of View". Therefore the study contents construct internal formal elements of the narrator and point of view into 2 individual researches, and the result can be summed up as below. From a narratological perspective, the narrator has roles and deeds as a narrative mediator who mediates the story and leads the story as presenting the origin with images and voices in the text extra and intra world through the process of producing the narrative inferred. To eliminate ambiguousness the term 'point-of view', this article applies 'focalization' theory to analyze narrative structure of film. The result of analysis shows that there are three focalizers in film; director, protagonist-character and camera. And aspects of film can be varied by distance of each focalizer. These distances between focalizers limit amount of visual information.

• Journal of Astronomy and Space Sciences
• /
• v.36 no.4
• /
• pp.283-292
• /
• 2019

The purpose of this study was to develop a period analysis algorithm for detecting new variable stars in the time-series data observed by charge coupled device (CCD). We used the data from a variable star monitoring program of the CBNUO. The R filter data of some magnetic cataclysmic variables observed for more than 20 days were chosen to achieve good statistical results. World Coordinate System (WCS) Tools was used to correct the rotation of the observed images and assign the same IDs to the stars included in the analyzed areas. The developed algorithm was applied to the data of DO Dra, TT Ari, RXSJ1803, and MU Cam. In these fields, we found 13 variable stars, five of which were new variable stars not previously reported. Our period analysis algorithm were tested in the case of observation data mixed with various fields of view because the observations were carried with 2K CCD as well as 4K CCD at the CBNUO. Our results show that variable stars can be detected using our algorithm even with observational data for which the field of view has changed. Our algorithm is useful to detect new variable stars and analyze them based on existing time-series data. The developed algorithm can play an important role as a recycling technique for used data

• Journal of the Korean Society of Costume
• /
• v.54 no.3
• /
• pp.27-42
• /
• 2004

Music videos provide for many others images, alluring the audience to fall in another emotional world, while the sexual images shown in them suggest new images of man and woman. Today, men's images are being interpreted from various viewpoints. As interests in men's fashion are visualized through music video clothing, not only juveniles who want to be identified with the music video images but also adults try to imitate them, and proceed to wear the clothing, obliterating the boundary between 'reality' and 'illusion' and creating new images of men. This study is aimed at reviewing the male images shown in the music videos, particular their clothing, produced between 2000 and 2002. The results of this study could be summarized as follows : 1. Since beginning of the human history, men's image has been characterized by patriarchal system, capitalism, bourgeois class which emerged after industrial revolution and other man-dominant socio-cultural phenomena, such male image are shown in the music video as conservative and dominant image. 2. However, due to the post-modern culture, the power began to be decentralized. while feminism and men's liberation movement gain strength. As a result, women or heterosexuals began to regard men as sexual objects, and such a phenomenon is featured as sexual, bisexual or decadent images in the music videos. 3. On the threshold of the 21st century, music videos have begun to creatively describe men's life, their social conflicts, dreams and hopes and death and thereby. feature men's such images as being destroyed in view of existentialism. The numerous creative men's images interpreted in this way are featured in many music video works only to create playful, cyborg or demonic images using the senses. After all, men's images are featured in the music video costumes in diverse ways ranging from the conventional images to acquiescent images. In addition, various male images are combined with the characteristics of the music videos to be re-created anew. The young men in the our modern age tend to imitate or apply such images to create their own images or individualistic styles. All in all, men's image can be fixed no longer but diversified and fragmented in the new age.