• Journal of Advanced Navigation Technology
• /
• v.25 no.6
• /
• pp.557-561
• /
• 2021

A meatadata has become an essential element in order to recommend video content to users. However, it is passively generated by video content providers. In the paper, a method for automatically generating metadata was studied in the existing manual metadata input method. In addition to the method of extracting emotion tags in the previous study, a study was conducted on a method for automatically generating metadata for genre and country of production through movie audio. The genre was extracted from the audio spectrogram using the ResNet34 artificial neural network model, a transfer learning model, and the language of the speaker in the movie was detected through speech recognition. Through this, it was possible to confirm the possibility of automatically generating metadata through artificial intelligence.

• Journal of Information Processing Systems
• /
• v.17 no.5
• /
• pp.867-878
• /
• 2021

In context awareness and user intention tasks, dataset construction is expensive because specific domain data are required. Although pretraining with a large corpus can effectively resolve the issue of lack of data, it ignores domain knowledge. Herein, we concentrate on data domain knowledge while addressing data scarcity and accordingly propose a multi-channel long short-term memory (LSTM). Because multi-channel LSTM integrates pretrained vectors such as task and general knowledge, it effectively prevents catastrophic forgetting between vectors of task and general knowledge to represent the context as a set of features. To evaluate the proposed model with reference to the baseline model, which is a single-channel LSTM, we performed two tasks: voice phishing with context awareness and movie review sentiment classification. The results verified that multi-channel LSTM outperforms single-channel LSTM in both tasks. We further experimented on different multi-channel LSTMs depending on the domain and data size of general knowledge in the model and confirmed that the effect of multi-channel LSTM integrating the two types of knowledge from downstream task data and raw data to overcome the lack of data.

• The KIPS Transactions:PartC
• /
• v.12C no.2 s.98
• /
• pp.217-222
• /
• 2005

The IEEE 802.11 based wireless local area networks are candidates to lead the broadband connectivity in the home and office scenarios. Recently IEEE proposed the 802.11e as a new standard to provide appropriate Quality of Services to a plethora of emerging real-time multimedia and high demanding applications such as high definition movie and audio distribution, video-conference and voice over IP. This paper studies the IEEE 802.11e/IEEE 802.11b interactions focusing on potential unfairness problems that might appear in networks with heterogeneous wireless LAN technologies as well as in the IEEE 802.11e deployment phase.

• ETRI Journal
• /
• v.37 no.2
• /
• pp.395-405
• /
• 2015

We address the problem of uncooperative person recognition through continuous monitoring. Multiple modalities, such as face, height, clothes color, and voice, can be used when attempting to recognize a person. In general, not all modalities are available for a given frame; furthermore, only some modalities will be useful as some frames in a video sequence are of a quality that is too low to be able to recognize a person. We propose a method that makes use of stochastic information updates of temporal modalities and environment estimators to improve person recognition performance. The environment estimators provide information on whether a given modality is reliable enough to be used in a particular instance; such indicators mean that we can easily identify and eliminate meaningless data, thus increasing the overall efficiency of the method. Our proposed method was tested using movie clips acquired under an unconstrained environment that included a wide variation of scale and rotation; illumination changes; uncontrolled distances from a camera to users (varying from 0.5 m to 5 m); and natural views of the human body with various types of noise. In this real and challenging scenario, our proposed method resulted in an outstanding performance.

• (The) Research of the performance art and culture
• /
• no.43
• /
• pp.5-46
• /
• 2021

The film Moranbong is the product of a trip to North Korea in 1958, when Armangati, Chris Marker, Claude Lantzmann, Francis Lemarck and Jean-Claude Bonardo left at the invitation of Joseon Film. However, for political reasons, the film was not immediately released, and it was not until 2010 that it was rediscovered and received attention. The movie consists of the narratives of Young-ran and Dong-il, set in the Korean War, that are folded into the narratives of Chunhyang and Mongryong in the classic Chunhyangjeon of Joseon. At this time, Joseon's classics are reproduced in the form of the drama Chunhyangjeon, which shares the time zone with the two main characters, and the two narratives are covered in a total of six scenes. There are two layers of middle-story frames in the movie, and if the same narrative is set in North Korea in the 1950s, there is an epic produced by the producers and actors of the Changgeuk Chunhyangjeon and the Changgeuk Chunhyangjeon as a complete work. In the outermost frame of the movie, Dong-il is the main character, but in the inner double frame, Young-ran, who is an actor growing up with the Changgeuk Chunhyangjeon and a character in the Changgeuk Chunhyangjeon, is the center. The following three OST albums are Corée Moranbong released in France in 1960, Musique de corée released in 1970, and 朝鮮の伝統音樂-唱劇 「春香伝」と伝統樂器- released in 1968 in Japan. While Corée Moranbong consists only of the music from the film Moranbong, the two subsequent albums included additional songs collected and recorded by Pyongyang National Broadcasting System. However, there is no information about the movie Moranbong on the album released in Japan. Under the circumstances, it is highly likely that the author of the record label or music commentary has not confirmed the existence of the movie Moranbong, and may have intentionally excluded related contents due to the background of the film's ban on its release. The results of analyzing the detailed scenes of the Changgeuk Chunhyangjeon, Farewell Song, Sipjang-ga, Chundangsigwa, Bakseokti and Prison Song in the movie Moranbong or OST album in the 1950s are as follows. First, the process of establishing the North Korean Changgeuk Chunhyangjeon in the 1950s was confirmed. The play, compiled in 1955 through the Joseon Changgeuk Collection, was settled in the form of a Changgeuk that can be performed in the late 1950s by the Changgeuk Chunhyangjeon between 1956 and 1958. Since the 1960s, Chunhyangjeon has no longer been performed as a traditional pansori-style Changgeuk, so the film Moranbong and the album Corée moranbong are almost the last records to capture the Changgeuk Chunhyangjeon and its music. Second, we confirmed the responses of the actors to the controversy over Takseong in the North Korean creative world in the 1950s. Until 1959, there was a voice of criticism surrounding Takseong and a voice of advocacy that it was also a national characteristic. Shin Woo-sun, who almost eliminated Takseong with clear and high-pitched phrases, air man who changed according to the situation, who chose Takseong but did not actively remove Takseong, Lim So-hyang, who tried to maintain his own tone while accepting some of modern vocalization. Although Cho Sang-sun and Lim So-hyang were also guaranteed roles to continue their voices, the selection/exclusion patterns in the movie Moranbong were linked to the Takseong removal guidelines required by North Korean musicians in the name of Dang and People in the 1950s. Second, Changgeuk actors' response to the controversy over the turbidity of the North Korean Changgeuk community in the 1950s was confirmed. Until 1959, there were voices of criticism and support surrounding Taksung in North Korea. Shin Woo-sun, who showed consistent performance in removing turbidity with clear, high-pitched vocal sounds, Gong Gi-nam, who did not actively remove turbidity depending on the situation, Cho Sang-sun, who accepted some of the vocalization required by the party, while maintaining his original tone. On the other hand, Cho Sang-seon and Lim So-hyang were guaranteed roles to continue their sounds, but the selection/exclusion patterns of Moranbong was independently linked to the guidelines for removing turbidity that the Gugak musicians who crossed to North Korea had been asked for.

• The Journal of the Korea Contents Association
• /
• v.17 no.9
• /
• pp.60-68
• /
• 2017

Through this thesis, I wanted to understand the form of free indirect discourse of modern films. To this end, I first explored the notion of the polyphonie as a mixture of the speaker and the character' voice in order to establish a concept related to free indirect discourse. However, I could not overlook the differences in the form of novels and movies to apply the following theory to films. Based on the concept of narrative distance, I sought to explore the possibility of free indirect discourse from the dual position of the camera. Next, I introduced the concept of free indirect discourse in the film by introducing the concept of Time in G. Deleuze' CinemaII. In other words, the time from Deleuze is the past and the present cycle, and he sees the Time circulating like the Non-Euclidean space. I wanted to understand the form of free indirect discourse in films by analyzing the concept of Time as an analysis of the movie .

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.351-356
• /
• 2023

The purpose of this study is to analyze how the movie <The Book of Fish> (2021) represents Joseon, which is slowly collapsing with the Neo-Confucian order of the 19th century shaking, and to discuss its meaning. Prior to the analysis, the analysis framework of the historical drama was presented considering the narration characteristics of the historical drama. Using the analysis framework of historical dramas, we confirmed that <The Book of Fish> is representing the image of Jeong Yak-jeon and Jang Chang-dae living their lives as independent individuals between the limitations and possibilities of the times based on the plot structure of the narrative of exile. Through the central memory and surplus memory created through plot and style elements such as contrast between black and white and color images, voice-over narration, chinese poetry subtitles and music, the film asks us universal questions about what it takes to live as an independent individual.

• The Journal of the Korea Contents Association
• /
• v.13 no.2
• /
• pp.100-107
• /
• 2013

Agn$\grave{e}$s Varda is a French representative female film director. In particular, the method combining subjectivity and objectivity is the most outstanding characteristics. However, it can be said that the method to support this is not only theme of the film but also creation of structure including the exploration for the filmic form different from classical film form. Such an approach accords with No$\ddot{e}$l Burch's refined analysis of filmic form. Therefore, This study aims to aesthetical analysis of the form of producing the structure in modern movies through which is one of the representative work by Agn$\grave{e}$sVarda and won the Golden Lion award in 1985 Venice Film Festival. The theme of the film, the recovery of relationships among people and contacts through the tragic death of drifting life, created a new filmic structure by formative experiment of the film. The formative experiment is the fragmented and repetitive construction with the introductory voice-over, and consists of movement and editing of camera and specific use of flashback and sound through the representation of figures and situations, mixture of narrative and non-narrative style.

• Journal of Broadcast Engineering
• /
• v.22 no.5
• /
• pp.632-642
• /
• 2017

In this paper, we propose a DNN based speech detection system using acoustic characteristics and context information of media audio. The speech detection for discriminating between speech and non-speech included in the media audio is a necessary preprocessing technique for effective speech processing. However, since the media audio signal includes various types of sound sources, it has been difficult to achieve high performance with the conventional signal processing techniques. The proposed method improves the speech detection performance by separating the harmonic and percussive components of the media audio and constructing the DNN input vector reflecting the acoustic characteristics and context information of the media audio. In order to verify the performance of the proposed system, a data set for speech detection was made using more than 20 hours of drama, and an 8-hour Hollywood movie data set, which was publicly available, was further acquired and used for experiments. In the experiment, it is shown that the proposed system provides better performance than the conventional method through the cross validation for two data sets.

• The Journal of the Korea Contents Association
• /
• v.20 no.11
• /
• pp.38-48
• /
• 2020

The domestic AI speaker market is growing into a full-fledged early audience market beyond the innovative consumer market with 3 million domestic supply units at the end of 2018, but the reality is that for various reasons, we are not satisfied with the use. There are many previous papers on AI Speaker, but the majority of research so far tends to be biased towards the acceptance of the device's own performance. Many changes are being made, such as OTT providers trying to secure the market through collaboration with AI speaker providers. This study tried to identify the priorities for content services, which can be another major selection factor for AI speakers, excluding the factors of unsatisfactory technology. First, this study identified the priorities among AI speaker selection factors using AHP (Analytic Hierarchy Process), based on the AI speaker selection factors derived through literature research. The most important hierarchical factor are Concierge Service, Education Service, and Entertainment Service order in AI speaker selection, and the primary content among the individual factors was the one that ranked weather/temperature/fine dust (11.6%) and child caring content was in the second place (10.8%), and then music service was in the third place (9.8%). The three top priorities were derived from the items in the top tier 1, 2 and 3 priorities. Of the total 15 individual services, 6 sub-layers of Concierge Service (weather/temperature/fine dust, news, voice schedule notification) and Education Service (foreign language, toddler, reading books) were in the top 8, and two of the Entertainment Service Music service and movie service ranked third and sixth.