• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1278-1281
• /
• 2022

현재 Action classification model은 computational resources의 제약으로 인해 video전체의 frame으로 학습하지 못한다. Model에 따라 다르지만, 대부분의 경우 하나의 action을 학습시키기 위해 보통 많게는 32frame, 적게는 8frame으로 model을 학습시킨다. 본 논문에서는 이 한계를 극복하기 위해 하나의 video의 많은 frame들을 mix-up과정을 거쳐 한장의 frame에 여러장의 frame 정보를 담고자 한다. 이 과정에서 video의 시간에 따른 변화(temporal- dynamics)를 손상시키지 않기 위해 linear mix-up이라는 방법을 제안하고 그 성능을 증명하며, 여러장의 frame을 mix-up시켜 모델의 성능을 향상시키는 가능성에 대해 논하고자 한다.

• International Journal of Concrete Structures and Materials
• /
• v.10 no.2
• /
• pp.237-247
• /
• 2016

Progressive collapse resistance of RC buildings can be analyzed by considering column loss scenarios. Using finite element analysis and a static test, the progressive collapse process of a RC frame under monotonic vertical displacement of a side column was investigated, simulating a column removal scenario. A single-story 1/3 scale RC frame that comprises two spans and two bays was tested and computed, and downward displacement of a side column was placed until failure. Our study offers insight into the failure modes and progressive collapse behavior of a RC frame. It has been noted that the damage of structural members (beams and slabs) occurs only in the bay where the removal side column is located. Greater catenary action and tensile membrane action are mobilized in the frame beams and slabs, respectively, at large deformations, but they mainly happen in the direction where the frame beams and slabs are laterally restrained. Based on the experimental and computational results, the mechanism of progressive collapse resistance of RC frames at different stages was discussed further. With large deformations, a simplified calculation method for catenary action and tensile membrane action is proposed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.7
• /
• pp.1280-1286
• /
• 2001

In this Study, we try to suggest a system that recognize worker's regular motion more effectively First, based on frame difference that separates still background from movable object to video that make a film of worker's motion. The next, with edge detection, estimating the center of motion could recognize continuous motion. By action cognition system that design in this research films worker's action using fixed CCTV to supplement problem of action awareness system that is applied in existent industry spot, various mountings to get action information minimized. Also, shorten session that need in awareness enforcing action awareness through image subtraction and edge detection between frame to reduce time necessary to draw worker's body part special quality, expense designed inexpensive action cognition system as being efficient.

• Electronics and Telecommunications Trends
• /
• v.36 no.2
• /
• pp.75-82
• /
• 2021

Online action detection (OAD) in a streaming video is an attractive research area that has aroused interest lately. Although most studies for action understanding have considered action recognition in well-trimmed videos and offline temporal action detection in untrimmed videos, online action detection methods are required to monitor action occurrences in streaming videos. OAD predicts action probabilities for a current frame or frame sequence using a fixed-sized video segment, including past and current frames. In this article, we discuss deep learning-based OAD models. In addition, we investigated OAD evaluation methodologies, including benchmark datasets and performance measures, and compared the performances of the presented OAD models.

• Coupled systems mechanics
• /
• v.4 no.2
• /
• pp.173-189
• /
• 2015

In reality, masonry infill modifies the seismic response of reinforced concrete (r.c.) frame structures by increasing the overall rigidity of structure which results in: increasing of total seismic load value, decreasing of deformations and period of vibration, therefore masonry infill frame structures have larger capacity of absorbing and dissipating seismic energy. The aim of the paper is to explore and assess actual influence of masonry infill on seismic response of r.c. frame structures, to determine whether it's justified to disregard masonry infill influence and to determine appropriate way to consider infill influence by design. This was done by modeling different structures, bare frame structures as well as masonry infill frame structures, while varying masonry infill to r.c. frame stiffness ratio and seismic intensity. Further resistance envelope for those models were created and compared. Different structures analysis have shown that the seismic action on infilled r.c. frame structure is almost always twice as much as seismic action on the same structure with bare r.c. frames, regardless of the seismic intensity. Comparing different models resistance envelopes has shown that, in case of lower stiffness r.c. frame structure, masonry infill (both lower and higher stiffness) increased its lateral load capacity, in average, two times, but in case of higher stiffness r.c. frame structures, influence of masonry infill on lateral load capacity is insignificant. After all, it is to conclude that the optimal structure type depends on its exposure to seismic action and its masonry infill to r.c. frame stiffness ratio.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2000.10a
• /
• pp.348-355
• /
• 2000

The characteristics on non linear behavior and the failure mechanism of RC space frame structure serving railway under seismic action have been investigated by numerical analysis in time domain. The structure concerned is modeled in 3 dimensional extent and RC frame elements with fibers are employed. Fibers are characterized as RC one and PL one to distinguish different energy release after cracking. Due to deviation of mass center and stiffness center of entire structure the complex behavior under seismic action is shown. The excessive shear force is concentrated on the pier beside flexible one relatively, which leads to failure of bridge concerned.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.5
• /
• pp.55-61
• /
• 2021

The character motion control based on physics simulation using reinforcement learning continue to being carried out. In order to solve a problem using reinforcement learning, the network structure, hyperparameter, state, action and reward must be properly set according to the problem. In many studies, various combinations of states, action and rewards have been defined and successfully applied to problems. Since there are various combinations in defining state, action and reward, many studies are conducted to analyze the effect of each element to find the optimal combination that improves learning performance. In this work, we analyzed the effect on reinforcement learning performance according to the state representation, which has not been so far. First we defined three coordinate systems: root attached frame, root aligned frame, and projected aligned frame. and then we analyze the effect of state representation by three coordinate systems on reinforcement learning. Second, we analyzed how it affects learning performance when various combinations of joint positions and angles for state.

• Steel and Composite Structures
• /
• v.30 no.3
• /
• pp.253-269
• /
• 2019

This paper investigates the effects of the tensile catenary action on dynamic increase factor (DIF) in the nonlinear static analysis for progressive collapse of steel-frame buildings. Numerical analyses were performed to verify the accuracy of the empirical and analytical expressions proposed in the literature in cases where the catenary action is activated. For this purpose, nonlinear static and dynamic analyses of a series of steel moment frame buildings with a different number of spans and stories were carried out following the alternate path method. Different column removal scenarios were considered as separate load cases. The dynamic increase factor that approximately compensates for the dynamic effects in the nonlinear static analysis was selected so to match results from the nonlinear dynamic analysis. The study results showed that the many expressions in literature may not work in cases where the catenary stage is fully developed.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.23 no.1
• /
• pp.1-8
• /
• 2019

In this study, the effectiveness of a multi-action hybrid damper (MHD) composed of lead rubber bearing (LRB) and friction pad was verified in terms of seismic performance improvement of a frame structure. The LRB and the friction elements are connected in series, so the LRB governs the intial small deformation and the friction determines large deformation behavior. Cyclic loading tests were conducted by using a half scale steel frame structure with the MHD, and the results indicated that the structure became to have the stable trilinear hysteresis with large initial stiffness and first yielding due to the LRB, and the second yielding due to the friction. The MHD could significantly increase the energy dissipation capacity of the structure and the hysteresis curves obtained by tests were almost identical to the analytically estimated ones.

• Communications of the Korean Mathematical Society
• /
• v.36 no.4
• /
• pp.829-833
• /
• 2021

We investigate the Möbius transformation action of PSL(2, ℝ) on the set of mutually distinct ordered triple points of ℝ ∪ {∞}.