• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.518-528
• /
• 2014

In this paper, we propose a method to recognize the action direction of human by developing 4D space-time (4D-ST, [x,y,z,t]) features. For this, we propose 4D space-time interest points (4D-STIPs, [x,y,z,t]) which are extracted using 3D space (3D-S, [x,y,z]) volumes reconstructed from images of a finite number of different views. Since the proposed features are constructed using volumetric information, the features for arbitrary 2D space (2D-S, [x,y]) viewpoint can be generated by projecting the 3D-S volumes and 4D-STIPs on corresponding image planes in training step. We can recognize the directions of actors in the test video since our training sets, which are projections of 3D-S volumes and 4D-STIPs to various image planes, contain the direction information. The process for recognizing action direction is divided into two steps, firstly we recognize the class of actions and then recognize the action direction using direction information. For the action and direction of action recognition, with the projected 3D-S volumes and 4D-STIPs we construct motion history images (MHIs) and non-motion history images (NMHIs) which encode the moving and non-moving parts of an action respectively. For the action recognition, features are trained by support vector data description (SVDD) according to the action class and recognized by support vector domain density description (SVDDD). For the action direction recognition after recognizing actions, each actions are trained using SVDD according to the direction class and then recognized by SVDDD. In experiments, we train the models using 3D-S volumes from INRIA Xmas Motion Acquisition Sequences (IXMAS) dataset and recognize action direction by constructing a new SNU dataset made for evaluating the action direction recognition.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1216-1217
• /
• 2022

포인트 클라우드는 객체 또는 장면을 재구성하기 위한 3D 데이터의 표현 방식 중 하나로써 가상 및 증강 현실을 포함한 다양한 분야에서 활용되고 있다. 포인트 클라우드 데이터는 품질에 따라 수많은 포인트로 이루어질 수 있으며, 이와 관련된 데이터의 양은 2차원 영상의 데이터보다 상당히 많다. 따라서 포인트 클라우드 데이터를 사용하여 다양한 서비스를 제공하기 위해서는 포인트 클라우드의 특징을 고려한 효율적인 압축 기술이 요구되며, 이에 따라 국제 표준화 단체의 Moving Picture Experts Group은 포인트 클라우드 데이터의 효율적인 압축을 위한 V-PCC 표준을 제정하였다. V-PCC는 포인트 클라우드 데이터를 다수의 2차원 공간으로 투영하여 점유 맵, 기하 영상, 그리고 속성 영상을 생성하고 각 2차원 영상을 기존의 비디오 코덱을 활용하여 압축하는 방식이다. 기존의 코덱을 사용하여 압축함에 따라 활용성이 높지만, 3차원 데이터를 다수의 2차원 영상을 통하여 압축하기 때문에 압축의 효율성을 높이기 위한 많은 연구가 필요하다. 본 논문에서는 V-PCC의 부호화 효율을 높이기 위해 점유 맵의 투영 정보를 활용한 속성 영상의 효율적인 압축 방법을 소개하고 이를 위한 적응적 SAO 방법을 제안한다. 실험에서 제안 방법은 V-PCC의 속성 영상에 대해 약 3.2%의 부호화 효율을 보인다.

• The KIPS Transactions:PartD
• /
• v.10D no.2
• /
• pp.283-292
• /
• 2003

One-dimensional time-series data have been studied in various database applications such as data mining and data warehousing. However, in the current complex business environment, multidimensional data sequences (MDS') become increasingly important in addition to one-dimensional time-series data. For example, a video stream can be modeled as an MDS in the multidimensional space with respect to color and texture attributes. In this paper, we propose the effective similarity measures on which the similar pattern retrieval is based. An MDS is partitioned into segments, each of which is represented by various geometric and semantic features. The similarity measures are defined on the basis of these segments. Using the measures, irrelevant segments are pruned from a database with respect to a given query. Both data sequences and query sequences are partitioned into segments, and the query processing is based upon the comparison of the features between data and query segments, instead of scanning all data elements of entire sequences.

• Journal of Digital Contents Society
• /
• v.13 no.2
• /
• pp.255-262
• /
• 2012

Human motion tracking algorithm is receiving attention from many research areas, such as human computer interaction, video conference, surveillance analysis, and game or entertainment applications. Over the last decade, various tracking technologies for each application have been demonstrated and refined among them such of real time computer vision and image processing, advanced man-machine interface, and so on. In this paper, we introduce cost-effective and real-time human motion tracking algorithms based on depth image 3D point matching with a given superellipsoid body representation. The body representative model is made by using parametric volume modeling method based on superellipsoid and consists of 18 articulated joints. For more accurate estimation, we exploit initial inverse kinematic solution with classified body parts' information, and then, the initial pose is modified to more accurate pose by using 3D point matching algorithm.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.37 no.4
• /
• pp.345-353
• /
• 2004

Target strength (TS) of Japanese common squids (Todarodes pacificus) were measured using 38 and 120 kHz split beam scientific echosounders under the live condition. For the TS measurement of an individual, a total of 3 squids (mantle length (ML): 22.8, 25, and 27 cm) were used using small fishhook method, whereas for measurement of swimming angle, a total of 8 squids (ML: 21-27 cm) were used under live condition, confined with net cage with 2 m diameter At the same time, two underwater video cameras enabled continuous monitoring of squid behavior. Considering normal behavior, the mean TS at 38 and 120 kHz varied from -48.6 to -45.9 dB, and from -46.5 to -44.6 dB, respectively In both frequencies, mean TS at 120 kHz is relatively higher than that of 38 kHz, approximately 1.3-2.5 dB. From free living condition, the mean swimming angle of the squlds was $-24^{\circ}$. The results of the measurement will be provided basic information for conducting acoustic surveys of the squid.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.2A
• /
• pp.148-158
• /
• 2008

This paper proposes an efficient signal to noise ratio (SNR) estimation algorithm and its hardware implementation for adaptive transmission system using M-ary modulation scheme. In this paper, we present the implementation results of the proposed algorithm for the second generation digital video broadcasting via satellite (DVB-S2) system, and the proposed algorithm can be tailored to the other communication systems using adaptive transmissions. We built a look-up table (LUT) using the theoretical background of the received signal distribution, and by using this LUT we need just two comparators and a counter for the hardware implementation. For this reason, the hardware of the proposed scheme produces accurate estimation results even with extremely low complexity. The simulation results investigated in this paper reveal that the proposed method can produce estimation results within the specified SNR range in the DVB-S2 system, and it requires a few hundreds of samples for average estimation error of about 1 dB.

• Journal of the Korean Society of Radiology
• /
• v.16 no.4
• /
• pp.491-497
• /
• 2022

In this study, we tried to secure basic data to create an environment necessary for distance learning through a survey on professor demand. Among the 184 full-time faculty members of the university, 73 (39.89%) respondents in 2021 and 87 (47.28%) in 2022 were included. As a result of the research on professor demand, in the 2021 school year, 27 people (37%) were classified as LMS improvement items when checking attendance, 38 people (23.3%) were pin-mics as content development support items, and 26 people (35.6%), 33 people (45.2%) of GOM Mix and 25 people (34.2%) of the distance education support center wanted to learn video editing program as the item of video editing program they are currently using. In the 2022 school year, 27 people (31.03%) said mobile upgrade as an LMS improvement item, 52 people (59.8%) of pin-microphone as a content development support item, 33 people (37.9%), but currently using the content creation intention using a studio. As for the video editing program they are working on, 47 people (54%) of GOM Mix Pro and 23 people (26.4%) of the distance education support center want to learn content creation method. In addition, the intention to produce content using the studio for the 2021 and 2022 academic year and the desired educational topic of the distance education support center in the future appeared insignificant (p > 0.05). In this distance education support center, we are working to solve the class of LMS attendance, upgrade mobile, and plan to distribute pin microphones. We are planning to increase the usability of the studio and provide training on how to use video editing programs and how to create video content. In order for a smooth class to take place in a university distance class, the university authorities should seek ways to support the instructor so that he/she does not have difficulties in performing his/her role as a teaching designer, such as setting learning goals, organizing and organizing content, motivating learning, and establishing effective class participation plans. there is a need

• Journal of Broadcast Engineering
• /
• v.22 no.4
• /
• pp.448-461
• /
• 2017

This paper proposes a vehicle counting and speed estimation method for traffic situation analysis in road CCTV videos. The proposed method removes a distortion in the images using Inverse perspective Mapping, and obtains specific region for vehicle counting and speed estimation using lane detection algorithm. Then, we can obtain vehicle counting and speed estimation results from using optical flow at specific region. The proposed method achieves stable accuracy of 88.94% from several CCTV images by regional groups and it totally applied at 106,993 frames, about 3 hours video.

• Journal of Information Technology Applications and Management
• /
• v.27 no.1
• /
• pp.187-198
• /
• 2020

With the development of technology and the creation of an entertainment environment for leisure, various marketing strategies are being used in the film industry. Among them, the use of the credit cookie of ending credits was very effective in producing the series. The ending credit is the time it takes to show the names of the people who made the movie, which is meaningless to the audience. There is a cost to produce a ending credit but It wasn't made because no revenue was generated. The credit cookie was inserted into this ending credit area, which brought new pleasure to the audience. Most of them were epilogue images showing the story behind the movie, NG images showing the NG situation during film production, and In videos mentioned in the movie but not shown in the movie itself. As various ideas about credit cookie were connected with marketing, a series movie and a spin-off foretelling the derivative works after the screening work were produced and have a new meaning. As a result, the time of ending credits, which had no commercial value, became the methodology of the most powerful promotional strategy. Looking at the difference between live-action film and animation in producing such credit cookie, unlike live-action films that edit the remaining parts after shooting, the NG video of the animation has a lot of time and money to produce. So, it hasn't try very well, and it seems to have been actively produced when moving from 2D animation to 3D animation. This is because 3D animation, which has already been modeled, can create new NG scenes by simply adding animating based on the layout of the created scene. Since it is possible to produce an episode movie at a low cost and time, and to use the scenes of the movie after the production, it will be necessary to strategically produce credit cookie for promotion in animation.

• Journal of the Korean Society of Radiology
• /
• v.3 no.2
• /
• pp.5-10
• /
• 2009

Myocardial imaging in SPECT (Single Photon Emission Computed tomography) scan of the gamma-ray emitting radiopharmaceuticals to patients after intravenous radiopharmaceuticals evenly spread in the heart region of interest by recording changes in the disease caused by a computer using the PSA test is to diagnose. Containing information on the functional myocardial perfusion imaging is a useful way to examine non-invasive heart disease, but the argument by noise and low resolution of the physical landscape that is difficult to give. For this paper, the level of myocardial imaging by using the three algorithms to split the video into 3-D implementation of the partitioned area to help you read the proposed plan. To solve the difficulty of reading level, interest in using the sheet set, partitioned area of the left ventricle was ranked the partitioned area was modeled as a 3-D images.