• Journal of KIISE
• /
• v.42 no.8
• /
• pp.990-997
• /
• 2015

Panoramic image creation has been extensively studied. Existing methods use customized hardware, or apply post-processing methods to seamlessly stitch images. These result in an increase in either cost or complexity. In addition, images can only be stitched under certain conditions such as existence of characteristic points of the images. This paper proposes a low cost and easy-to-use system that produces realtime panoramic video. We use an off-the-shelf embedded platform to capture multiple images, and these are then transmitted to a server in a compressed format to be merged into a single panoramic video. Finally, we analyze the performance of the implemented system by measuring time to successfully create the panoramic image.

• Journal of the Korean Society of Costume
• /
• v.53 no.5
• /
• pp.1-12
• /
• 2003

This paper proposes a system for measuring the 3D human bodies using the multiple 2D images. The system establishes the multiple image input circumstance from the digital camera for image measurement. The algorithm considering perspective projection leads us to estimate the 3D human bodies from the multiple 2D images such as frontal. side and rear views. The results of the image measurement is compared those of the direct measurement and the 3D scanner for the total 40 items (12 heights, 15 widths and 13 depths). Three persons measure the 40 items using the three measurement methods. In comparison of the results obtained among the measurement methods and the persons, the results between the image measurement and the 3D scanner are very similar. However, the errors for the direct measurement are relatively larger than those between the image measurement and the 3D scanner. For example, the maximum errors between the image measurement and the 3D scanner are 0.41cm in height, 0.39cm in width and 0.95cm in depth. The errors are acceptable in body measurement. Performance of the image measurement is superior to the direct. because the algorithm estimates the 3D positions using the perspective projection. In above comparison, the image measurement is expected as a new method for measuring the 3D body, since it has the various advantages of the direct measurement and 3D scanner in performance for measurement as well as in the devices, cost, Portability and man power.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.233-237
• /
• 2009

This paper represents a new approach which addresses quality degradation of a synthesized view, when a virtual camera moves forward. Generally, interpolation technique using only two neighboring views is used when a virtual view is synthesized. Because a size of the object increases when the virtual camera moves forward, most methods solved this by interpolation in order to synthesize a virtual view. However, as it generates a degraded view such as blurred images, we prevent a synthesized view from being blurred by using more cameras in multiview camera configuration. That is, we solve this by applying super-resolution concept which reconstructs a high resolution image from several low resolution images. Therefore, data fusion is executed by geometric warping using a disparity of the multiple images followed by deblur operation. Experimental results show that the image quality can further be improved by reducing blur in comparison with interpolation method.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.533-538
• /
• 2001

In this paper, we extend FARGs single mode attribute to multiple attributes for real image application and present a new CBIR using FMARG(Fuzzy Multiple Attribute Relational Graph), which can handle queries involving multiple attributes, not only object label, but also color, texture and spatial relation. In the experiment using the synthetic image database of 1,024 images and the natural image database of 1.026 images built from NETRA database and Corel Draw, the proposed approach shows 6~30% recall increase in the synthetic image database and a good performance, at the displacements and the retrieved number of similar images in the natural image database, compared with the single attribute approach.

• International Journal of Contents
• /
• v.5 no.4
• /
• pp.30-34
• /
• 2009

Many studies have been done on the integral imaging pickup whose objective is to get efficiently elemental images from a lens array with respect to three-dimensional (3D) objects. In the integral imaging pickup process, it is necessary to render an elemental image from each elemental lens in a lens array for 3D objects, and then to combine them into one total image. The multiple viewpoint rendering (MVR) is one of various methods for integral imaging pickup. This method, however, has the computing and rendering time problem for obtaining element images from a lot of elemental lens. In order to solve the problems, in this paper, we propose a parallel MVR (PMVR) method to generate elemental images in a parallel through distribution of elemental lenses into multiple threads simultaneously. As a result, the computation time of integral imaging using PMVR is reduced significantly rather than a sequential approach and then we showed that the PMVR is very useful.

• Journal of the Semiconductor & Display Technology
• /
• v.9 no.3
• /
• pp.17-21
• /
• 2010

As the compact and easily accessible handheld devices, such as cellular phones and MP3 players equipped with image acquisition functionality, are becoming widely available among common users, various applications of images are rapidly increasing. Image related services and software such as web-based image presentation and image manipulation for personal or commercial purpose enable users to view contents of remote image archive and to manipulate enormous amount of images in local or network based storage as well. It is necessary for users to identify the images efficiently so that the same images are perceived as one physical entity instead of recognizing them as different images as the trends are getting stronger. In order to support this environment, we propose a method that generates image identifiers or indexing for images within a solid and efficient manner. The proposed image identifier utilizes multiple index values. The integration of component index values creates a unique composite value that can be used as a file name, file system identifier, or database index. Our experimental results on generation of constituent index values have shown favorable results.

• Journal of Information Processing Systems
• /
• v.18 no.4
• /
• pp.575-586
• /
• 2022

Uneven light in real-world causes visual degradation for underexposed regions. For these regions, insufficient consideration during enhancement procedure will result in over-/under-exposure, loss of details and color distortion. Confronting such challenges, an unsupervised low-light image enhancement network is proposed in this paper based on the guidance of the unpaired low-/normal-light images. The key components in our network include super-resolution module (SRM), a GAN-based low-light image enhancement network (LLIEN), and denoising-scaling module (DSM). The SRM improves the resolution of the low-light input images before illumination enhancement. Such design philosophy improves the effectiveness of texture details preservation by operating in high-resolution space. Subsequently, local lightness attention module in LLIEN effectively distinguishes unevenly illuminated areas and puts emphasis on low-light areas, ensuring the spatial consistency of illumination for locally underexposed images. Then, multiple discriminators, i.e., global discriminator, local region discriminator, and color discriminator performs assessment from different perspectives to avoid over-/under-exposure and color distortion, which guides the network to generate images that in line with human aesthetic perception. Finally, the DSM performs noise removal and obtains high-quality enhanced images. Both qualitative and quantitative experiments demonstrate that our approach achieves favorable results, which indicates its superior capacity on illumination and texture details restoration.

• Journal of Korea Game Society
• /
• v.3 no.2
• /
• pp.30-34
• /
• 2003

Image-based rendering is an approach to generate realistic images in real-time without modeling explicit 3D geometry, Especially, TIP(Tour Into the Picture) is preferred for its simplicity in constructing 3D background scene. However, TP has a limitation that a viewpoint cannot go far from the origin of the TIP for the lack of geometrical information. in this paper, we propose a method to interpolating the TIP images to generate smooth and realistic navigation. We construct multiple TIP models in a wide area of the virtual environment. Then we interpolate foreground objects and background object respectively to generate smooth navigation results.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.335-336
• /
• 2009

In this paper, I propose an automatic approach to annotating images dynamically based on MBRM(Multiple Bernoulli Relevance Models) using GLCM(Grey Level Co-occurrence Matrix). MBRM is more appropriate to annotate images compare with multinomial distribution. The model is used in limited test set, MSRC-v2 (Microsoft Research Cambridge Image Database). The results show that this model is significantly outperforms previously reported results on the task of image annotation and retrieval.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.377-378
• /
• 2007

This paper proposes an system to generate rolled-equivalent fingerprints by mosaicking sequential images captured by an toothless device. To capture rolled-equivalent fingerprints, previous works use multiple cameras. However, the method in this paper captures sequential fingerprint images with a single camera and mosaic the images by estimating the transform between images through optical flow.