• Journal of Broadcast Engineering
• /
• 제21권6호
• /
• pp.913-919
• /
• 2016

In this paper, we propose the visual tracking algorithm that takes advantage of multiple random walkers. We first show the tracking method based on support vector machine as [1] and suggest a method that suppresses feature vectors extracted from backgrounds while preserve features vectors from foregrounds. We also show how to discriminate between foregrounds and backgrounds. Learned by reducing influences of backgrounds, support vector machine can clearly distinguish foregrounds and backgrounds from the image whose target objects are similar to backgrounds and occluded by another object. Thus, the algorithm can track target objects well. Furthermore, we introduce a simple method improving tracking speed. Finally, experiments validate that proposed algorithm yield better performance than the state-of-the-art trackers on the widely-used benchmark dataset with high speed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3976-3990
• /
• 2022

Currently, cross-attention is widely used in Siamese trackers to replace traditional correlation operations for feature fusion between template and search region. The former can establish a similar relationship between the target and the search region better than the latter for robust visual object tracking. But existing trackers using cross-attention only focus on rich semantic information of high-level features, while ignoring the appearance information contained in low-level features, which makes trackers vulnerable to interference from similar objects. In this paper, we propose a Multi-level Cross-attention Siamese network(MCSiam) to aggregate the semantic information and appearance information at the same time. Specifically, a multi-level cross-attention module is designed to fuse the multi-layer features extracted from the backbone, which integrate different levels of the template and search region features, so that the rich appearance information and semantic information can be used to carry out the tracking task simultaneously. In addition, before cross-attention, a target-aware module is introduced to enhance the target feature and alleviate interference, which makes the multi-level cross-attention module more efficient to fuse the information of the target and the search region. We test the MCSiam on four tracking benchmarks and the result show that the proposed tracker achieves comparable performance to the state-of-the-art trackers.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• 제5권1호
• /
• pp.1-6
• /
• 2012

In this paper, we propose a HMM(Hidden Markov Model)-based segmentation method to model shadows as well as foreground and background regions. The shadow of moving objects often keeps from visual tracking. We propose an HMM-based segmentation method which classifies each object in real time. In the case of traffic monitoring movies, the effectiveness of the proposed method was proved by experiments.

• Proceedings of the KIEE Conference
• /
• 대한전기학회 1999년도 추계학술대회 논문집 학회본부 B
• /
• pp.664-666
• /
• 1999

영상 추적(visual tracking)은 로봇의 시각기반제어, 교통정보시스템, 무인감시시스템 등 다양한 분야에 적용 가능하기 때문에 지정된 혹은 운동이 감지된 물체를 지속적이고 빠르게 추적하는 데 목적을 둔다. 이 때 어려운 문제 중 하나는 시간이 지나면서 위치이동은 물론 회전에 와해 물체의 모양이 변한다는 것이다. 이에 본 논문에서는 물체의 3차원 회전에 대응 가능한 실시간 영상추적 알고리듬을 제안한다. 이 알고리듬은 SSD(sum-of-squared differences)를 기반으로 하되, 물체의 배경이 바뀔 때나 물체가 영상추적 윈도우보다 작은 경우에도 추적이 가능하고 3차원 회전에 대응 가능하다. 이것은 3차원 회전으로 인하여 추적목표를 잃어버리는 것을 막기 위하여 기준 영역이 회전할 때 제안된 성능지수에 따라 영상추적 영역과 기준 영상을 새롭게 설정해줌으로써 구현된다. 제안된 알고리듬은 PC기반 실시간 시각시스템에서 성공적으로 실험되었다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• 제32B권12호
• /
• pp.1652-1662
• /
• 1995

In this paper, a hybrid neural network is proposed to improve the learning ability of a neural network. The union of the characteristics of a Self-Organizing Neural Network model and of multi-layer perceptron model using the backpropagation learning method gives us the advantage of reduction of the learning error and the learning time. In learning process, the proposed hybrid neural network reduces the number of nodes in hidden layers to reduce the calculation time. And this proposed neural network uses the fuzzy feedback values, when it updates the responding region of each node in the hidden layer. To show the effectiveness of this proposed hybrid neural network, the boolean function(XOR, 3Bit Parity) and the solution of inverse kinematics are used. Finally, this proposed hybrid neural network is applied to the visual tracking control of a PUMA560 robot, and the result data is presented.

• Journal of Institute of Control, Robotics and Systems
• /
• 제9권7호
• /
• pp.515-526
• /
• 2003

Tutorial contents of kinematics and dynamics of a wheeled drive mobile robot are presented. Based on the dynamic model, simulation studies of position tracking of a mobile robot are performed. The control structure of several position control algorithms using visual feedback are proposed and their performances are compared. In order to compensate for uncertainties from unknown dynamics and ignored dynamic effects such as slip conditions, neural network based position control schemes are proposed. Experiments are conducted and the results show the performance of the vision based neural network control scheme fumed out to be the best among several proposed schemes.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 한국컴퓨터정보학회 2015년도 제52차 하계학술대회논문집 23권2호
• /
• pp.232-233
• /
• 2015

디지털 미디어의 발전으로 인해 모션그래픽의 구성요소들은 다양한 방식으로 제작되고 있다. 모션그래픽을 활용한 영상콘텐츠 제작기법은 새로운 이미지를 어떻게 제작하느냐에 따라 보다 창의적이고 효율적인 영상으로 표현되기도 한다. 모션그래픽의 활용도는 웹, 게임, 및 영화 방송 등 다양한 콘텐츠의 시각적 정보전달의 의미를 가지고 있다. 모션그래픽의 새롭고 실험적인 작업은 어떠한 제작방식에 따라 그 결과는 아주 다르게 나타난다. 또한 영상제작에 대한 프로세스는 새로운 뉴미디어의 기술의 결합으로 영상콘텐츠의 창의적인 작품으로 된다. 따라서 본 논문은 모션그래픽의 제작방식에서 실사카메라의 움직임을 가상카메라로 트래킹하여 입체적인 모션그래픽을 구현하는 제작방법을 제시한다.

• Journal of information and communication convergence engineering
• /
• 제3권1호
• /
• pp.56-61
• /
• 2005

In this paper, we present a video rate control scheme for storage and delivery in which the time-varying viewing interests are controlled by human gaze. To track the gaze, the pupil's movement is detected using the three-step process : detecting face region, eye region, and pupil point. To control bit rates, the quantization parameter (QP) is changed by considering the static parameters, the video object priority derived from the pupil tracking, the target PSNR, and the weighted distortion value of the coder. As results, we achieved human interfaced visual model and corresponding region-of-interest rate control system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 한국해양정보통신학회 2009년도 추계학술대회
• /
• pp.285-288
• /
• 2009

동영상에서 물체를 추적하기 위한 많은 방법들은 각 방법마다 다양한 요소들을 이용한다. 하지만 다양한 요소들은 그 관계를 명확히 나타낼 수 없다. 본 논문에서는 동영상에서 물체 추적을 위해 시각적 기본 요소의 동일성을 나타내고 설명하기 위한 방법을 제안한다. 시각적 기본 요소란 이미지에서 얻을 수 있는 순수한 시각적 기본 요소들인 색상, 명암, 윤곽선 정보를 말하며 이 정보를 이용하여 연속된 프레임을 비교해 동일한 물체인지를 판단한다. 그리하여 판단된 정보를 가지고 추적에 사용되는 요소들의 지속적인 갱신을 통해 대상을 모델 없이 추적하고 인식 과정에 대하여 제어가 가능하도록 한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.617-620
• /
• 1997

We have built a robot soccer system to participate in MIROSOT97. This paper represents hardware specification of our system and our strategy. We select a centralized on-line system for a soccer game. The paper explains hardware specifications of our system for later development. Also, the paper explains our strategy from two viewpoints. From the viewpoint of cooperation, some heuristic ideas are implemented. From the viewpoint of path plan, Cubic spline is used with cost function which minimized time, radius of curvature for smoothness, and obstacle potential field. Direct comparison will be realized in MIROSOT97.