• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.51-59
• /
• 2024

Collecting a dataset with a corresponding labeled gaze vector requires a high cost in the gaze estimation field. In this paper, we suggest a data augmentation of manipulating the gaze of an original image, which improves the accuracy of the gaze estimation model when the number of given gaze labels is restricted. By conducting multi-class gaze bin classification as an auxiliary task and adjusting the latent variable of the diffusion model, the model semantically edits the gaze from the original image. We manipulate a non-binary attribute, pitch and yaw of gaze vector to a desired range and uses the edited image as an augmented train data. The improved gaze accuracy of the gaze estimation network in the semi-supervised learning validates the effectiveness of our data augmentation, especially when the number of gaze labels is 50k or less.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.12 no.2
• /
• pp.181-192
• /
• 2008

A Web-based design automation system for concrete slab bridge has been developed to manage the resultant documents as well as to speed up the repetitive design process. Manipulation of engineering drawings in the Web page is one of the critical functions needed for Web-based design automation. Active Server Pages (ASP) is used to that collects design parameters in the database, and eXtensible Markup Language (XML) and XML-based vector graphics are expected to facilitate the representation of engineering drawings in the Web page. In this paper, we present how we used XML and Scalable Vector Graphics (SVG) to compose engineering drawings and represent them in the Web page. XML Data Island we designed to define drawing components turned out effective in manipulating the engineering drawings in the Web page.

• Journal of The Korean Association For Science Education
• /
• v.21 no.1
• /
• pp.102-113
• /
• 2001

In order to compare with the 6th and 7th science curricular for the inquiry skill elements in the elementary and secondary school, we divided skill domains into five classes which were process skill, step skill for inquiry instruction, inquiry activity skill, manipulative skill and breeding-farming skill. And then we investigated the kinds and frequencies for the inquiry skill elements of the 6th and 7th curricular in the elementary and secondary school. The results were as follows: 1. The total kinds of inquiry skill element showed 17 kinds in the 6th curriculum and 23 kinds in the 7th. Therefore, the 7th curriculum was higher 1.4 times than the 6th curriculum in the kinds of skill elements. 2. The total frequencies for the inquiry skill elements of the 6th curriculum were 408 and those of the 7th were 729. Therefore, the 7th curriculum was about 1.8 times as many as the 6th. 3. In the kinds of inquiry skill elements according to the school levels, the course of the elementary school showed 14 kinds in the 6th curriculum and 18 kinds in the 7th. The course of middle school showed 7 kinds in 6th and 16 kinds in 7th. The integrated science course of high school was 10 kinds in the 6th and 10 kinds in the 7th. The skill elements in four science curricular of the high school course showed total 11 kinds in the 6th and 21 kinds in the 7th. And then the kinds of inquiry skill elements of the 7th curriculum in the middle and high school course showed about 2 times as many as the 6th curriculum. In the school level, the increase of skill elements showed the highest in the middle school course, and then in the high school course. 4. The total skill elements from the elementary school to the high school in the 6th science curriculum showed 17 kinds and in the order from the highest to the lowest rates, such as experimenting 20%, observing 15%, interpreting and analyzing data 13%, investigating 9%, measuring 7%, drawing a conclusion and assessment 7%, discussion 6%, communicating 5%, classifying 4%, recognizing problems and formulating hypothesis 4%, predicting 3%, designing and carrying out an experiment 3%, collecting and treating data 2%, manipulating skill 1%, modeling 0.5%, breeding and farming 0.3% and inferring 0.2%. 5. The total skill elements from the elementary school to the high school in the 7th curriculum appeared 23 kinds and in the order from the highest to the lowest rates, such as drawing a conclusion and assessment 31%, investigating 14%, collecting and treating data 8%, observing 7%, experimenting 7%, recognizing problems and formulating hypothesis 6%, interpreting and analyzing data 4%, measuring 3%, discussion 3%, manipulating skill 3%, modeling 3%, classifying 2%, project 2%, educational visits 1%, controlling variables 1%, predicting 1%, inferring 1%, operational definition 1%, communicating 1%, designing and carrying out an experiment 0.3%, breeding and farming 0.3%, applicating a number 0.2% and relating with time and space 0.2%. In the conclusion, the 7th curriculum was added 6 kinds of skill elements to the 6th curriculum, such as operational definition, applicating a number, relating with time and space, controlling variables, educational visits and project.

• Journal of the Korean earth science society
• /
• v.29 no.2
• /
• pp.204-217
• /
• 2008

This study analyzed the inquiry activities appearing in the astronomy sections of elementary, middle and highschool level science textbooks according to the five essential features of inquiry in the classroom as proposed by the National Science Education Standards (NRC, 2000), and SAPA (Science-A Process Approach). On the basis of this analysis, it is clear that the science textbook inquiry activities released the limitation to meet the goal of science education, namely scientific literacy, as it has been laid out by the 7th Science Educational Curriculum. This study revealed that the features of scientific inquiry which are most frequently used in the astronomy sections of science textbooks are 'data collection' and 'form explanation', whereas the features of 'oriented-question', 'evaluate explanations' and 'communicate and justify' rarely appeared. The analysis of inquiry activities by SAPA showed that the basic inquiry skills of 'observing', 'communicating' and 'manipulating materials' were used with increasing frequency according to grade level, and the integrated skills of 'investigating', 'creating models', 'interpreting data' and 'experimenting' were more emphasized in the textbooks. Therefore, it is suggested that students be provided with more opportunities to experience all the features of scientific inquiry and scientific processes as envisioned by the 7th Science Educational Curriculum in order to achieve the stated goal of scientific literacy. Science educators should be required to develop new lesson modules which will allow students to experience authentic scientific inquiry. It is crucial for science teachers to reflect upon and develop their understanding and teaching strategies regarding scientific inquiry through professional development programs in teacher education.

• Journal of the Korean earth science society
• /
• v.29 no.2
• /
• pp.163-174
• /
• 2008

In this study, we compared textbook knowledge organization with students' mental models to contribute to a more well-designed instruction scheme. The selected science content was the cause of moon phases. We investigated 9 textbooks and 25 third-year middle school students. Patterns and features in participants' mental models were identified through cross inter-rater data analysis by 9 researchers, including in service teachers and experts in science education. According to the results, observing and modeling are the main activities engaged in when dealing with moon phases. The activities consisted of such concepts as: lunar revolution, the sun's parallel rays, the illuminated half of moon, and the relative positions of the sun, moon, and earth. Each concept involved inquiry skills such as: creating and manipulating models, utilizing the relationship between time and space, and communicating. However, the most important skills which are required for authentic scientific inquiry, namely controlling variables and formulating hypotheses, were missing. We categorized students' mental models into three types: scientific models, mixed models, and alternative models. The knowledge structure of each of the models was also discussed in this paper. Consequently, it was found that, typically, students were not given enough opportunities to strengthen the connection among ideas.

• Journal of KIISE:Software and Applications
• /
• v.28 no.2
• /
• pp.179-191
• /
• 2001

Comparison shopping is a merchant brokering process that finds the best price for the desired product from several Web-based online stores. To get a scalable comparison shopper, we need an agent that automatically constructs a simple information extraction procedure, called a wrapper, for each semi-structured store. Automatic construction of wrappers for HTML-based Web stores is difficult because HTML only defines how information is to be displayed, not what it means, and different stores employ different ways of manipulating customer queries and different presentation formats for displaying product descriptions. Wrapper induction has been suggested as a promising strategy for overcoming this heterogeneity. However, previous scalable comparison-shoppers such as ShopBot rely on a strong bias in the product descriptions, and as a result, many stores that do not confirm to this bias were unable to be recognized. This paper proposes a more scalable comparison-shopping agent named MORPHEUS. MORPHEUS presents a simple but robust inductive learning algorithm that antomatically constructs wrappers. The main idea of the proposed algorithm is to recognize the position and the structure of a product description unit by finding the most frequent pattern from the sequence of logical line information in output HTML pages. MORPHEUS successfully constructs correct wtappers for most stores by weakening a bias assumed in previous systems. It also tolerates some noises that might be present in production descriptions such as missing attributes. MORPHEUS generates the wrappers rapidly by excluding the pre-processing phase of removing redundant fragments in a page such as a header, a tailer, and advertisements. Eventually, MORPHEUS provides a framework from which a customized comparison-shopping agent can be organized for a user by facilitating the dynamic addition of new stores.

• Cartoon and Animation Studies
• /
• s.33
• /
• pp.55-78
• /
• 2013

The issues on the representation of digital images, the uncanny is not considered aesthetically but must be overcome. It seems that the future of visual media using digital images, depends on the development and advancement of the digital technology. However, the Brothers Quay regarded as unique and experimental animators, focus their attention towards the past rather than the future. They trace back to the origin of the media, reproducing, manipulating, and representing gestures in oder to restore the shock and thrill delivered in film and animation. The Brothers Quay explore the uncanny of gestures through the mechanism of medium. They work in a wide range of genres, from animation, live action to dance film. The creaking spasms, which create the feeling that the movement does not belong to this world, and the strange gestures, reveals the flicker of the still projection and make audiences recognize gestures in the true meaning of the word. While they animate live actors and dead objects like automata, the hidden desires are revealed over screen. In this paper, focusing on the works of Brothers Quay, I tried to explore the uncanny of gesture based on Freud's and Jentsch' theory of uncanny.

• Asian-Australasian Journal of Animal Sciences
• /
• v.24 no.11
• /
• pp.1483-1503
• /
• 2011

To advance the effectiveness of the current Hanwoo improvement system, we developed a general simulation that compared a series of breeding schemes under realistic user circumstances. We call this system the Integrated General Model (IGM) and it allows users to control the breeding schemes and selection methods by manipulating the input parameters. The Current Hanwoo Performance and Progeny Test (CHPPT) scheme was simulated with a Modified Hanwoo Performance and Progeny Test (MHPPT) scheme using a Hanwoo Breeding Farm cow population of the Livestock Improvement Main Center (LOMC) of the National Agricultural Cooperatives Federation (NACF). To compare the two schemes, a new method, the Simple Hanwoo Performance Test (SHPT), which uses ultrasound technology for measuring the carcass traits of live animals, was developed. These three models, including the CHPPT, incorporated three types of selection criteria: phenotype (PH), true breeding value (TBV), and estimated breeding value (EBV). The simulation was scheduled to mimic an actual Hanwoo breeding program; thus, the simulation was run to include the years 1983-2020 for each breeding method and was replicated 10 times. The parameters for simulation were derived from the literature. Approximately 642,000 animals were simulated per replication for the CHPPT scheme; 129,000 animals were simulated for the MHPPT scheme and 112,000 animals for the SHPT scheme. Throughout the 38-year simulation, all estimated parameters of each simulated population, regardless of population size, showed results similar to the input parameters. The deviations between input and output values for the parameters in the large populations were statistically acceptable. In this study, we integrated three simulated models, including the CHPPT, in an attempt to achieve the greatest genetic gains within major economic traits including body weight at 12 months of age (BW12), body weight at 24 months of age (BW24), average daily gain from 6 to 12 months (ADG), carcass weight (CWT), carcass longissimus muscle area (CLMA), carcass marbling score (CMS), ultrasound scanned longissimus muscle area (ULMA), and ultrasound scanned marbling score (UMS).

• Journal of Digital Convergence
• /
• v.10 no.10
• /
• pp.19-29
• /
• 2012

We made a regression analysis on the early IPO of venture capital investments in Korean IPO market. First, we found that it was likely to shorten the period to IPO in companies which were fast growing with a good operating cash flow, but these companies had a higher possibility of the earning management. Second, companies with more assets and larger size of the board of directors did not take companies public any earlier. Third, a better corporate governance also had no impact on the time period to IPO in the newly public firms. The findings above clearly show that venture-backed companies in Korea pursue the tendency of an early IPO. This phenomenon was much clearer when the companies were invested in by multiple venture capital firms than by a single investor. In general, venture capital firms invest in companies which are fast growing and which have a good operating cash flow. On the other side, venture capitals make investee companies go public earlier by manipulating operating earnings, so that they themselves may exit early. In conclusion, this research has shown that venture capitals in Korea do not play a positive role in the corporate transparency. This is the paradox of venture capital investment and this also shows the current status of Korean venture capital firms.

• The Journal of Society for e-Business Studies
• /
• v.20 no.1
• /
• pp.217-241
• /
• 2015

One of the challenges in online community management is what level of perceived anonymity can be granted to encourage active participation from members while discouraging unhealthy activities. Few studies developed a scheme of anonymity and investigated how different levels of anonymity influence community activities. This study develops a classification scheme of anonymity encompassing the following three different levels : (1) real name(no anonymity), (2) nickname (partial anonymity), and (3) random assignment of a temporal ID (complete anonymity). Then, it examines how different levels of anonymity influence trust and perceived risk, which in turn affect virtual team performance. A series of laboratory experiments were performed, manipulating the levels of anonymity, in the context of well-structured communities that allow prior interactions among community members. The data was collected from 364 laboratory participants and analyzed using ANOVA and PLS. The results indicate that the difference of anonymity between (2) and (3) had not be significant and the only (1) could not guarantee the anonymity. The impact of anonymity on trust and perceived risk could not be significant in this situation. These findings could contribute to make more beneficial member identification strategies in online community practice.