• Journal of Broadcast Engineering
• /
• v.18 no.3
• /
• pp.342-348
• /
• 2013

In this paper, a novel frame rate up conversion (FRUC) algorithm using adaptive extended bilateral motion estimation (AEBME) is proposed. Conventionally, extended bilateral motion estimation (EBME) conducts dual motion estimation (ME) processes on the same region, therefore involves high complexity. However, in this proposed scheme, a novel block type matching procedure is suggested to accelerate the ME procedure. We calculate the edge information using sobel mask, and the calculated edge information is used in block type matching procedure. Based on the block type matching, decision will be made whether to use EBME. Motion vector smoothing (MVS) is adopted to detect outliers and correct outliers in the motion vector field. Finally, overlapped block motion compensation (OBMC) and motion compensated frame interpolation (MCFI) are adopted to interpolate the intermediate frame in which OBMC is employed adaptively based on frame motion activity. Experimental results show that this proposed algorithm has outstanding performance and fast computation comparing with EBME.

• Journal of Korea Multimedia Society
• /
• v.17 no.7
• /
• pp.797-805
• /
• 2014

In this paper we propose a new frame rate up conversion scheme which is used to overcome the motion blur problem of liquid crystal display caused by its slow response. The conventional bilateral motion estimation method which is mainly used in the frame rate up conversion scheme has a drawback that it cannot find true motion vector if there are blocks with simple texture in the search range. To solve this problem, a texture adaptive bilateral motion estimation method that increases cost value of block with simple texture is proposed. Also a motion estimation scheme that utilizes neighboring motion vector effectively is proposed to reduce computation time required to estimate motion. Since the proposed scheme does not apply all available motion vectors within the search range, the execution time of frame rate up conversion can be reduced dramatically. Experimental results show that the interpolated frame image quality of the proposed method is improved in subjective as well as objective view point compared with that of the conventional method.

• Journal of Digital Contents Society
• /
• v.8 no.4
• /
• pp.441-449
• /
• 2007

In video coding, motion estimation is a process to estimate the pixel of the current frame from the reference frame, which affects directly the predictive quality and the encoding time. This paper is related to AHHS(Adaptive Hierarchical Hexagon Search) using spatio-temporal motion activity for fast motion estimation. The proposed method defines the spatio-temporal motion activity of the current macroblock using the motion vectors of its spatio-temporally adjacent macroblocks, and then conventional AHS(Adaptive Hexagon Search) is performed if the spatio-temporal motion activity is lower, otherwise, hierarchical hexagon search is performed on a multi-layered hierarchical space constructed by multiple sub-images with low frequency in wavelet transform. In the paper, based on computer simulation results for multiple video sequences with different motion characteristics, the performance of the proposed method was analysed and assessed in terms of the predictive quality and the computational time. Experimental results indicate that the proposed method is both suitable for (quasi-) stationary and large motion searches. The proposed method could keep the merit of the adaptive hexagon search capable of fast estimating motion vectors and also adaptively reduce the local minima occurred in the video sequences with higher spatio-temporal motion activity.

• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.1
• /
• pp.50-61
• /
• 2005

To provide adaptively video streaming services on network environment, video transcoding is introduced. The one of transcoding methods is the frame-rate conversion. it needs a re-estimation about a motion vector of the frame to refer a skipping frame. This re-estimation makes higher the computational complexity in video transcoding. To reduce the computational complexity of a motion vector refinement, this paper proposes a region & activity based motion vector composition scheme that refine the moving vector of a skipping frame. This scheme composes each motion vector from the weight based on the activity information of a macroblock and the site of the overlapped area. The experiment result shows that RABVC has a higher PSNR than the value of existing weight-based motion vector selection schemes though the computational complexity of our scheme is similar to that of other schemes.

• Journal of Korea Multimedia Society
• /
• v.14 no.8
• /
• pp.992-999
• /
• 2011

This paper proposed a human body posture recognition program based on haar-like feature and hand activity detection. Its distinguishing features are the combination of face detection and motion detection. Firstly, the program uses the haar-like feature face detection to receive the location of human face. The haar-like feature is provided with the advantages of speed. It means the less amount of calculation the haar-like feature can exclude a large number of interference, and it can discriminate human face more accurately, and achieve the face position. Then the program uses the frame subtraction to achieve the position of human body motion. This method is provided with good performance of the motion detection. Afterwards, the program recognises the human body motion by calculating the relationship of the face position with the position of human body motion contour. By the test, we know that the recognition rate of this algorithm is more than 92%. The results show that, this algorithm can achieve the result quickly, and guarantee the exactitude of the result.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.1118-1133
• /
• 2017

Computer vision-based human activity recognition (HAR) has become very famous these days due to its applications in various fields such as smart home healthcare for elderly people. A video-based activity recognition system basically has many goals such as to react based on people's behavior that allows the systems to proactively assist them with their tasks. A novel approach is proposed in this work for depth video based human activity recognition using joint-based motion features of depth body shapes and Deep Belief Network (DBN). From depth video, different body parts of human activities are segmented first by means of a trained random forest. The motion features representing the magnitude and direction of each joint in next frame are extracted. Finally, the features are applied for training a DBN to be used for recognition later. The proposed HAR approach showed superior performance over conventional approaches on private and public datasets, indicating a prominent approach for practical applications in smartly controlled environments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.7
• /
• pp.3599-3619
• /
• 2019

In human activity recognition system both static and motion information play crucial role for efficient and competitive results. Most of the existing methods are insufficient to extract video features and unable to investigate the level of contribution of both (Static and Motion) components. Our work highlights this problem and proposes Static-Motion fused features descriptor (SMFD), which intelligently leverages both static and motion features in the form of descriptor. First, static features are learned by two-stream 3D convolutional neural network. Second, trajectories are extracted by tracking key points and only those trajectories have been selected which are located in central region of the original video frame in order to to reduce irrelevant background trajectories as well computational complexity. Then, shape and motion descriptors are obtained along with key points by using SIFT flow. Next, cholesky transformation is introduced to fuse static and motion feature vectors to guarantee the equal contribution of all descriptors. Finally, Long Short-Term Memory (LSTM) network is utilized to discover long-term temporal dependencies and final prediction. To confirm the effectiveness of the proposed approach, extensive experiments have been conducted on three well-known datasets i.e. UCF101, HMDB51 and YouTube. Findings shows that the resulting recognition system is on par with state-of-the-art methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.77.5-77
• /
• 2002

A mechanical device was developed for assisting the grasping function of a person whose fingers suffered cervical injury and thus are unable to grasp. This device is composed of a mechanical glove put on the user's hand and a muscle sensor to measure the activity of his or her muscle. The mechanical glove consists of a finger frame, a base and an air cylinder mounted on the base. With the kinematics carefully designed, the finger frame can achieve the grasping motion under the actuation of the air cylinder. For controlling this motion, an innovative sensor was developed to detect the user's motion intention. The sensor measures the change of the muscle stiffness...

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.6
• /
• pp.547-552
• /
• 2015

In this paper, an ARS-EKF based motion counting algorithm for repetitive exercises such as calisthenics is proposed using a smartwatch. Raw sensor signals from accelerometers and gyroscopes are widely used for conventional smartwatch counting algorithms based on pattern recognition. However, generated features from raw data are not intuitive to reflect the movement of motions. The proposed motion counter algorithm is composed of navigation based feature generation and counting with error correction. The candidate features for each activity are velocity and attitude calculated through an ARS-EKF algorithm. In order to select those features which reveal the characteristics of each motion, an exercise frame from the initial sensor frame is introduced. Counting processes are basically based on the zero crossing method, and misdetected counts are eliminated via simple classification algorithms considering the frequency of the counted motions. Experimental results show that the proposed algorithm efficiently and accurately counts the number of exercises.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.1577-1580
• /
• 2004

여러 응용 서비스를 유 무선을 포함한 다양한 네트웍을 통해 제공하기 위해서는 네트웍에 적응할 수 있는 서비스 형태가 요구된다. 그 가운데 멀티미디어 서비스의 경우 네트웍이 서로 다른 환경에 적응할 수 있는 해결책으로 트랜스코딩 기술이 제시되었다. 하지만, 트랜스코딩을 위해 필요한 복호 부호의 처리 과정은 실시간으로 제공되는 멀티미디어 스트리밍의 경우에 제약조건으로 작용하고, 이에 따른 처리 과정을 대폭 줄이는 일부 기술들은 사용자 측의 서비스 품질에 문제점을 안고 있다. 본 논문에서는 트랜스코딩을 통한 처리 과정과 사용자 측 서비스 품질의 두 가지 측면을 고려하는 frame dropping 시의 모션 벡터 합성 기법에 대해서 언급한다. 또한, 본 논문에서는 기존의 기법과는 달리 양방향 예측 프레임이 포함된 경우에도 적용할 수 있는 확장성을 제공한다.