• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.1
• /
• pp.69-75
• /
• 2023

In recent years, there has been a growing interest in deep learning-based animal pose estimation, especially in the areas of animal behavior analysis and healthcare. However, existing animal pose estimation techniques do not perform well when body parts are occluded or not present. In particular, the occlusion of dog tail or ear might lead to a significant degradation of performance in pet behavior and emotion recognition. In this paper, to solve this intractable problem, we propose a simple yet novel framework for pet pose estimation where pet pose is predicted on an outpainted image where some body parts hidden outside the input image are reconstructed by the image inpainting network preceding the pose estimation network, and we performed a preliminary study to test the feasibility of the proposed approach. We assessed CE-GAN and BAT-Fill for image outpainting, and evaluated SimpleBaseline for pet pose estimation. Our experimental results show that pet pose estimation on outpainted images generated using BAT-Fill outperforms the existing methods of pose estimation on outpainting-less input image.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.189-201
• /
• 2021

The art of dynamic images has experienced three development stages, including experimental films, recording art, and new media image. By introducing all kinds of new materials, new media to the art, and the art of dynamic images has created more freedom for art creation. With the development of digital information technology, dynamic image works have put forward an increasingly high requirement of visual art. The combination of dynamic images and visual illusion can give rise to different forms and expression methods, thus endowing artworks with more vigor. This paper provides an overview by sorting out the lineage and development of dynamic images in the background, as well as understanding the application and performance of contrasted visual illusion. Based on the understanding of the characteristics of visual illusion, we discuss the new characteristics of applying the theory of visual illusion to new media dynamic images in relation to the technical approach of dynamic images. Through the analysis of specific works of Telematic Vision, we search for its reasonable combination and find the appropriate technical means of implementation. We discuss how to use digital multimedia technology and spatial optical illusion to make the design more novel and impactful, and consider how the combination of digital dynamic image technology and visual illusion should be interpreted and applied.

• The Journal of the Korea Contents Association
• /
• v.13 no.6
• /
• pp.40-47
• /
• 2013

In case of wild fire, early detection of wild fire is the most important factor in minimizing the damages. In this paper, we suggest an effective system that detects wild fire using a panoramic image from a single camera with PAN/TILT head. This enables the system to detect the size and the location of the fire in the early stages. After converting RGB image input to color YCrCb image, the differential image is used to detect changes in movement of the smoke to determine the regions which may be prone to forest fire. Histogram analysis of fire flame is used to determine the possibility of fire in the predetermined regions. In addition, image matching and SURF were used to create the panoramic image. There are many advantages in this system. First of all, it is very economical because this system needs only a single camera and a monitor. Second, it shows the live image of wide view through panoramic image. Third, this system can reduce the quantity of saved data by storing panoramic images.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.257-269
• /
• 2023

There is currently a burgeoning demand for image synthesis from photos and videos using deep learning models. Existing video synthesis models solely extract motion information from the provided video to generate animation effects on photos. However, these synthesis models encounter challenges in achieving accurate lip synchronization with the audio and maintaining the image quality of the synthesized output. To tackle these issues, this paper introduces a novel framework based on an image animation approach. Within this framework, upon receiving a photo, a video, and audio input, it produces an output that not only retains the unique characteristics of the individuals in the photo but also synchronizes their movements with the provided video, achieving lip synchronization with the audio. Furthermore, a super-resolution model is employed to enhance the quality and resolution of the synthesized output.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.91-96
• /
• 2010

A laser speckle is a random pattern that has a granular appearance produced by reflected light when a coherent laser illuminates an irregular course surface. Laser speckle system has many advantages. It can detect some animals functional parts. Moreover, it relatively consists of simple and in-expensive system. It is very important that detecting micro-vessels through image processed image. Current study is to improve image quality through variable image processing method. But this paper made laser speckle endoscope for miniaturization and commercialization laser speckle system. We had endoscope test through goldfish's tail. We will compare the processed speckle image and halogen image.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.178-182
• /
• 2007

In this paper, we design and implement the UCC services prototype system for image and video. The proposed system consists of the two components such as the multimedia processing subsystem and the metadata management subsystem, and provides the API to UCC service developers. The multimedia processing subsystem supports the media management and editing of image and video, and the streaming services of video. The metadata management subsystem supports the metadata management and retrieval of image and video, and the reply management and script processing of UCC.

• Journal of Internet of Things and Convergence
• /
• v.5 no.2
• /
• pp.15-20
• /
• 2019

As the machine learning research has developed, the field of translation and image analysis such as optical character recognition has made great progress. However, video translation that combines these two is slower than previous developments. In this paper, we develop an image translator that combines existing OCR technology and translation technology and verify its effectiveness. Before developing, we presented what functions are needed to implement this system and how to implement them, and then tested their performance. With the application program developed through this paper, users can access translation more conveniently, and also can contribute to ensuring the convenience provided in any environment.

• Journal of IKEEE
• /
• v.18 no.1
• /
• pp.64-76
• /
• 2014

In the image acquisition tests of image sensors and modules, how to compare pixel values, how to digitize "defect", and how to determine "fault" are extremely varied from test equipments. Therefore, setup of test equipments for image acquisition characteristics should be customized for target devices under test. In general, test procedures are delivered as technical documents in not standardized but arbitrary formats. So test equipment should be manually programmed fitting individual setup condition. This costs huge time and effort and has high risk of human error. In this paper, the standardized description method is proposed to apply to almost all existing tests where various tests are described in a single framework.

• Journal of Digital Contents Society
• /
• v.9 no.3
• /
• pp.461-470
• /
• 2008

There are two methods to express natural phenomena using image processing techniques. One is the presentation of simulations for natural phenomena and the other is the display of images based on script. A lot of people used to get the display using vast data and complicated math formulas. When we get the output images in this way, we will have some problems in time and cost. In this paper, we use fluid images excluding using the complicated math formulas, programming, and taking pictures to present the natural phenomena. We construct the library to express the natural phenomena effectively using 2D images and simulation of fog for the background of oriental painting in 3D space.

• The Journal of the Korea Contents Association
• /
• v.8 no.2
• /
• pp.48-56
• /
• 2008

In this paper, we design and implement the UCC services prototype system for image and video. The proposed system consists of the two components such as the multimedia processing subsystem and the metadata management subsystem. It provides the API to UCC service developers. The multimedia processing subsystem supports the media management and editing of image and video, and the streaming services of video. The metadata management subsystem supports the metadata management and retrieval of image and video. The reply management and script processing of UCC are given in the subsystem.