• IEIE Transactions on Smart Processing and Computing
• /
• 제1권2호
• /
• pp.73-77
• /
• 2012

This paper reports a combined depth- and model-based face detection and tracking approach. The proposed algorithm consists of four functional modules; i) color-based candidate region extraction, ii) generation of the depth histogram for handling occlusion, iii) rotation-invariant face region detection using ellipse fitting, and iv) face tracking based on motion prediction. This technique solved the occlusion problem under complicated environment by detecting the face candidate region based on the depth-based histogram and skin colors. The angle of rotation was estimated by the ellipse fitting method in the detected candidate regions. The face region was finally determined by inversely rotating the candidate regions by the estimated angle using Haar-like features that were robustly trained robustly by the frontal face.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(4)
• /
• pp.69-72
• /
• 2002

Human face detection has many applications such as face recognition, face or facial feature tracking, pose estimation, and expression recognition. We present a new method for automatically segmentation and face detection in color images. Skin color alone is usually not sufficient to detect face, so we combine the color segmentation and shape analysis. The algorithm consists of two stages. First, skin color regions are segmented based on the chrominance component of the input image. Then regions with elliptical shape are selected as face hypotheses. They are certificated to searching for the facial features in their interior, Experimental results demonstrate successful detection over a wide variety of facial variations in scale, rotation, pose, lighting conditions.

• International Journal of Contents
• /
• 제10권3호
• /
• pp.17-25
• /
• 2014

In this paper, we describe an adaptive Markov chain Monte Carlo-based particle filter that effectively addresses real-time multi-face tracking on mobile platforms. Because traditional approaches based on a particle filter require an enormous number of particles, the processing time is high. This is a serious issue, especially on low performance devices such as mobile phones. To resolve this problem, we developed a tracker that includes a more sophisticated likelihood model to reduce the number of particles and maintain the identity of the tracked faces. In our proposed tracker, the number of particles is adjusted during the sampling process using an adaptive sampling scheme. The adaptive sampling scheme is designed based on the average acceptance ratio of sampled particles of each face. Moreover, a likelihood model based on color information is combined with corner features to improve the accuracy of the sample measurement. The proposed tracker applied on various videos confirmed a significant decrease in processing time compared to traditional approaches.

• 한국항공우주학회지
• /
• 제40권2호
• /
• pp.118-128
• /
• 2012

본 논문은 쿼드로터형 무인 비행체를 비전센서를 이용한 목표 추적 위치 제어기 설계하였고, 이를 시뮬레이션 및 실험을 통해서 확인하였다. 우선 제어기 설계에 앞서 쿼드로터의 동역학 분석 및 실험데이터를 통한 모델링을 수행하였다. 이때, 모델의 계수들은 실제 비행 데이터를 이용한 PEM(Prediction Error Method)을 이용하여 얻었다. 이 추정된 모델을 바탕으로 LQR(Linear Quadratic Regulator) 기법을 이용한 임의의 목표를 따라가는 위치 제어기를 설계하였으며, 이때 위치 정보는 비전센서의 색 정보를 이용한 Color Tracking기능을 이용하여 쿼드로터와 물체의 상대적인 위치를 얻어내었고, 초음파 센서를 이용하여 고도 정보를 얻어 내었다. 마지막으로 실제 움직이는 물체의 추적 제어 실험을 수행하여 LQR 제어기 성능을 평가하였다.

• 인터넷정보학회논문지
• /
• 제24권5호
• /
• pp.29-35
• /
• 2023

CCTV는 범죄 예방, 공공 안전 강화, 교통 관리 등 다양한 목적으로 사용된다. 그러나 카메라의 범위와 해상도가 향상됨에 따라 영상에서 개인의 신상정보가 노출되는 위험성이 있다. 따라서 영상에서 개인 정보를 보호함과 동시에 개인을 식별할 수 있는 새로운 기술의 필요성이 존재한다. 본 논문에서는 객체 식별 및 추적을 위한 히스토그램 기반 특이값 분해를 제안한다. 제안하는 방법은 객체의 색상 정보를 이용하여 영상에 존재하는 서로 다른 객체를 구분한다. 객체 인식을 위하여 YOLO와 DeepSORT를 이용해 영상에 존재하는 사람을 탐지 및 추출한다. 탐지된 사람의 위치 정보를 이용해 흑백 히스토그램으로 색상 값을 추출한다. 추출한 색상 값 중 유의미한 정보만을 추출하여 사용하기 위해 특이값 분해를 이용한다. 특이값 분해를 이용할 때 결과에서 상위 특이값의 평균을 이용함으로 객체 색상 추출의 정확도를 높인다. 특이값 분해를 이용해 추출한 색상 정보를 다른 영상에 존재하는 색상과 비교하며 서로 다른 영상에 존재하는 동일 인물을 탐지한다. 색상 정보 비교를 위해 유클리드 거리를 이용하며 정확도 평가는 Top-N을 이용한다. 평가 결과 흑백 히스토그램과 특이값 분해를 사용하여 동일 인물을 탐지할 때 최대 100%에서 최소 74%를 기록하였다.

• 전자공학회논문지CI
• /
• 제47권3호
• /
• pp.1-10
• /
• 2010

손동작을 통한 입력방법은 컴퓨터와 디지털 기기의 발전에 따라 요구되는 새로운 HCI(Human-Computer Interaction) 방법으로써 그 가능성을 가지고 있으며 이에 대한 다양한 시도가 있었다. 본 논문에서는 컴퓨터 비전을 기반으로 단일 카메라를 사용하는 손 영역 검출 및 추적방법을 제시하고 이에 의한 컴퓨터 인터페이스를 제안한다. 기존에 많이 쓰이는 피부색 매치 방법에 추가하여 형태 정보를 더함으로써 손 영역 검출능력을 향상 시켰다. 이러한 형태 정보를 추출하는 방법으로써 주요 방향 에지 기술자라는 방법을 제안하였고 이는 강력하여 학습 시간 없이 한 가지 손 모델만을 사용하여 손 영역 검출을 할 수 있다. 또한 손 영역 검출과 추적하는 방법을 나누어 추적할 때는 회전에 대한 자유도를 높이도록 설계 하였다. 위 방법을 이용하여 3차원 공간에 그려지는 필기체 숫자 인식에 적용해 보았으며 분류 방법으로 DNAC 알고리즘을 사용하였다. 결과적으로 손 영역 검출은 82%의 검출률을 보였고 필기체 숫자 인식은 90%의 인식률을 보였다.

• ETRI Journal
• /
• 제34권6호
• /
• pp.932-941
• /
• 2012

A powerful interaction mechanism is one of the key elements for the success of smart TVs, which demand far more complex interactions than traditional TVs. This paper proposes a novel interface based on the famous touch interaction model but utilizes long-range bare hand tracking to emulate touch actions. To satisfy the essential requirements of high accuracy and immediate response, the proposed hand tracking algorithm adopts a fast color-based tracker but with modifications to avoid the problems inherent to those algorithms. By using online modeling and motion information, the sensitivity to the environment can be greatly decreased. Furthermore, several ideas to solve the problems often encountered by users interacting with smart TVs are proposed, resulting in a very robust hand tracking algorithm that works superbly, even for users with sleeveless clothing. In addition, the proposed algorithm runs at a very high speed of 82.73 Hz. The proposed interface is confirmed to comfortably support most touch operations, such as clicks, swipes, and drags, at a distance of three meters, which makes the proposed interface a good candidate for interaction with smart TVs.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.615-618
• /
• 2002

This paper proposes a real-time face detection and tracking a method in complex backgrounds. The proposed method is based on the principal component analysis (PCA) technique. For the detection of a face, first, we use a skin color model and motion information. And then using the PCA technique the detected regions are verified to determine which region is indeed the face. The tracking of a face is based on the Euclidian distance in eigenspace between the previously tracked face and the newly detected faces. Camera control for the face tracking is done in such a way that the detected face region is kept on the center of the screen by controlling the pan/tilt platform. The proposed method is extensible to other systems such as teleconferencing system, intruder inspection system, and so on.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권5호
• /
• pp.1856-1869
• /
• 2015

This paper addresses the issues of 3D human activity detection, tracking and recognition from RGB-D video sequences using a feature structured framework. During human tracking and activity recognition, initially, dense depth images are captured using depth camera. In order to track human silhouettes, we considered spatial/temporal continuity, constraints of human motion information and compute centroids of each activity based on chain coding mechanism and centroids point extraction. In body skin joints features, we estimate human body skin color to identify human body parts (i.e., head, hands, and feet) likely to extract joint points information. These joints points are further processed as feature extraction process including distance position features and centroid distance features. Lastly, self-organized maps are used to recognize different activities. Experimental results demonstrate that the proposed method is reliable and efficient in recognizing human poses at different realistic scenes. The proposed system should be applicable to different consumer application systems such as healthcare system, video surveillance system and indoor monitoring systems which track and recognize different activities of multiple users.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권1호
• /
• pp.305-326
• /
• 2019

Discriminative correlation filter (DCF) based tracking algorithms have recently shown impressive performance on benchmark datasets. However, amount of recent researches are vulnerable to heavy occlusions, irregular deformations and so on. In this paper, we intend to solve these problems and handle the contradiction between accuracy and real-time in the framework of tracking-by-detection. Firstly, we propose an innovative strategy to combine the template and color-based models instead of a simple linear superposition and rely on the strengths of both to promote the accuracy. Secondly, to enhance the discriminative power of the learned template model, the spatial regularization is introduced in the learning stage to penalize the objective boundary information corresponding to features in the background. Thirdly, we utilize a discriminative multi-scale estimate method to solve the problem of scale variations. Finally, we research strategies to limit the computational complexity of our tracker. Abundant experiments demonstrate that our tracker performs superiorly against several advanced algorithms on both the OTB2013 and OTB2015 datasets while maintaining the high frame rates.