• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1250-1260
• /
• 2020

This study proposes a novel deep neural network model that can accurately detect objects and their relationships in an image and represent them as a scene graph. The proposed model utilizes several multimodal features, including linguistic features and visual context features, to accurately detect objects and relationships. In addition, in the proposed model, context features are embedded using graph neural networks to depict the dependencies between two related objects in the context feature vector. This study demonstrates the effectiveness of the proposed model through comparative experiments using the Visual Genome benchmark dataset.

• 한국BIM학회 논문집
• /
• 제6권1호
• /
• pp.1-8
• /
• 2016

This paper provides a method for generating flight path of Unmanned Aerial Vehicle (UAV) that is intended to be used in visual inspection of dome-type nuclear containment building. The method basically employs 3-D model to extract accurate location coordinates. Two basic route patterns that provide guide lines in defining moving locations were defined for each side wall and dome section of the containment. The route patterns support sequential capturing of images as well. In addition, several simple equations and an algorithm for calculation of the moving location on the route were developed on the basis of 3-D geometric characteristics of the containment building. A prototype computer program has been implemented to validate the proposed method, and a case study shows the method can visualize covering area in 3-D model as well.

• 대한기계학회논문집
• /
• 제19권5호
• /
• pp.1220-1228
• /
• 1995

In this paper, we consider the problems of tracking an object in a real image. In evaluating these problems, we explore a new technique based on an active contour model commonly called a snake model, and propose the active bar models to represent target. Using this model, we simplified the target welection problems, reduced the search space of energy surface, and obtained the better performances than those of snake model. This approach improves the numerical stability and the tendency for points to bunch up and speed up the computational efficiency. Representing the object by active bar, we can easily obtain the zeroth, the first, and the second moment and it facilitates the target tracking. Finally, we present the good result for the visual tracking problem.

• 한국국방경영분석학회지
• /
• 제26권2호
• /
• pp.28-38
• /
• 2000

This research investigates a kill probability model for the performance evaluation of guided missile system, and also develops graphical user interface for the input and output of the model based on the visual object-oriented programming application. The major simulation events used in this research are missile guidance homing point, burst points, and kill mechanism(direct kill, blast kill and fragment kill). For the user interface, we also design and implement the visualization system that can show the graphic style of the kill probability attained by the model. The results of sample run are shown, but these could be improved to be better with visual simulation which can visulaize all the simulation process of the model.

• 한국멀티미디어학회논문지
• /
• 제23권2호
• /
• pp.368-375
• /
• 2020

In this study, the area of interest(AOI) of Soju ad poster was tracked for analysis the time to frist fixation, the average of fixation duration and count by the study indexes. As a result of the analysis, Visual attention was higher the face than the body shape of the ad model. This means "when we look at printed ads, we see picture elements first, not language one" but language elements can't be overlooked either. Also, the importance of the model role could be verified by measuring the visual attention on the Soju ad poster. Based on the results of this study, if further research on ad posters is carried out and scientific and quantitative interpretation methods are presented, it can be used as product marketing data that can be reflected in ad model selection and poster design.

• 정보처리학회논문지B
• /
• 제10B권6호
• /
• pp.639-646
• /
• 2003

본 논문은 일련의 불연속적인 도트 자극으로부터 시각적 착시현상을 추출하는 신경회로망 모델을 제시한다. 제안된 모델은 시각 정보처리 경로에서 발견되는 시각 세포들의 특성을 근거로 한다. 본 연구는 일련의 이산 도트 자극들이 개별적인 도트들로 인식하지 않고 연속적인 가상의 윤곽으로 인식하는 시각적 착시 현상을 나타내는 생리심리학 실험을 기초로 하여 도트 자극의 시각적 착시를 구현한 것으로서 실험에서는 가상 다각형 형태로 배치된 6에서 10개의 도트자극들을 사용한다. 이 실험 데이터는 Smith & Vos가 생리심리학적 실험에서 다룬 데이터와 유사하다. 제안된 모델은 이산 도트자극으로부터 연속적인 착시 윤곽을 성공적으로 추출한다.

• KIEAE Journal
• /
• 제7권5호
• /
• pp.35-40
• /
• 2007

The purpose of this study is to execute evaluation experiment to know the evaluation property of lighting environment of residential space for senior people, considering visual characteristics along aging, and finally provide basic data for the lighting plan to ensure the visual amenity. Processes of this study are as follows;1) Analyzed the variation property of visual sensibility and visual ability of senior people along aging. 2) Selected 3 types of life behavior(rest, conversation and reading) after checking life behavior in residential space for senior people based on advanced study. 3) Made the Mock-up Model that Dimming is possible, actual furnace to model. 4) Executed sensitivity evaluation experiment about lighting environment. 5) Analyzed evaluation property of lighting environment of residential space for senior people. Results of this study are as follows, 1) With lens-filter, we got comfort and amenity in bulb-color lamp which has similar color temperature with red of lens filter. 2) Lighting environment tests during conversation : With lens filters, they felt comfort on bulb color in case of higher illuminance than 850lux and daylight color in 500lux. 3) Lighting environment tests at reading : With lens filter, bulb color got better score in brightness and appropriateness than daylight color.

• 대한임베디드공학회논문지
• /
• 제16권6호
• /
• pp.323-330
• /
• 2021

Visual object tracking is a challenging area of study in the field of computer vision due to many difficult problems, including a fast variation of target shape, occlusion, and arbitrary ground truth object designation. In this paper, we focus on the reinforced feature of the dynamic search area to get better performance than conventional discriminative model prediction trackers on the condition when the accuracy deteriorates since low feature discrimination. We propose a reinforced input feature method shown like the spotlight effect on the dynamic search area of the target tracking. This method can be used to improve performances for deep learning based discriminative model prediction tracker, also various types of trackers which are used to infer the center of the target based on the visual object tracking. The proposed method shows the improved tracking performance than the baseline trackers, achieving a relative gain of 38% quantitative improvement from 0.433 to 0.601 F-score at the visual object tracking evaluation.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.354-357
• /
• 2021

Infants communicate their feelings and needs to the outside world through non-verbal methods such as crying and displaying diverse facial expressions. However, inexperienced parents tend to decode these non-verbal messages incorrectly and take inappropriate actions, which might affect the bonding they build with their babies and the cognitive development of the newborns. In this paper, we propose an aural-visual two-stream based infant cry recognition system to help parents comprehend the feelings and needs of crying babies. The proposed system first extracts the features from the pre-processed audio and video data by using the VGGish model and 3D-CNN model respectively, fuses the extracted features using a fully connected layer, and finally applies a SoftMax function to classify the fused features and recognize the corresponding type of cry. The experimental results show that the proposed system classification exceeds 0.92 in F1-score, which is 0.08 and 0.10 higher than the single-stream aural model and single-stream visual model.

• Journal of information and communication convergence engineering
• /
• 제1권2호
• /
• pp.74-81
• /
• 2003

There are many researches on visual languages, but the most of them are difficult to support various collaborative interactions on a distributed multimedia environment. So, this paper suggests a collaborative visual language for interaction between multi-users. The visual language can describe a conceptual model for collaborative interactions between multi-users. Using the visual language, generated visual sentences consist of object icons and interaction operators. An object icon represents a user who is responsible for a collaborative activity, has dynamic attributes of a user, and supports flexible interaction between multi-users. An interaction operator represents an interactive relation between multi-users and supports various collaborative interactions. Merits of the visual language are as follows: supporting of both asynchronous interaction and synchronous interaction, supporting flexible interaction between multi-users according to participation or leave of users, supporting a user oriented modeling, etc. For example, an application to a workflow system for document approval is illustrated. So we could be found that the visual language shows a collaborative interaction.