• The Journal of the Acoustical Society of Korea
• /
• v.18 no.5
• /
• pp.44-51
• /
• 1999

This paper presents a massively parallel computational model for the efficient integration of speech and natural language understanding. The phoneme model is based on continuous Hidden Markov Model with context dependent phonemes, and the language model is based on a knowledge base approach. To construct the knowledge base, we adopt a hierarchically-structured semantic network and a memory-based parsing technique that employs parallel marker-passing as an inference mechanism. Our parallel speech recognition algorithm is implemented in a multi-Transputer system using distributed-memory MIMD multiprocessors. Experimental results show that the parallel speech recognition system performs better in recognition accuracy than a word network-based speech recognition system. The recognition accuracy is further improved by applying code-phoneme statistics. Besides, speedup experiments demonstrate the possibility of constructing a realtime parallel speech recognition system.

• ETRI Journal
• /
• v.33 no.4
• /
• pp.569-579
• /
• 2011

This paper presents a novel approach for human activity recognition (HAR) using the joint angles from a 3D model of a human body. Unlike conventional approaches in which the joint angles are computed from inverse kinematic analysis of the optical marker positions captured with multiple cameras, our approach utilizes the body joint angles estimated directly from time-series activity images acquired with a single stereo camera by co-registering a 3D body model to the stereo information. The estimated joint-angle features are then mapped into codewords to generate discrete symbols for a hidden Markov model (HMM) of each activity. With these symbols, each activity is trained through the HMM, and later, all the trained HMMs are used for activity recognition. The performance of our joint-angle-based HAR has been compared to that of a conventional binary and depth silhouette-based HAR, producing significantly better results in the recognition rate, especially for the activities that are not discernible with the conventional approaches.

• Asian-Australasian Journal of Animal Sciences
• /
• v.24 no.4
• /
• pp.457-462
• /
• 2011

The purpose of this study was to detect QTL for carcass quality on bovine chromosome (BTA) 6 using a high density SNP map in a Hanwoo population. The data set comprised 45 sires and their 427 Hanwoo steers that were born between spring of 2005 and fall of 2007. The steers that were used for progeny testing in the Hanwoo Improvement Center in Seosan, Korea, were genotyped with the 2,535SNPs on BTA6 that were embedded in the Illumina bovine SNP 50K chip. Four different linkage disequilibrium (LD) mapping models were applied to detect significant SNPs for carcass quality traits; the fixed model with a single marker, the random model with a single marker, the random model with haplotype effects using two adjacent markers, and the random model at hidden state. A total of twelve QTL were detected, for which four, one, three and four SNPs were detected on BTA6 under the respective models (p<0.001). Among the detected QTL, four, two, five and one QTL were associated with carcass weight, backfat thickness, longissimus dorsi muscle area, and marbling score, respectively (p<0.001). Our results suggest that the use of multiple LD mapping approaches may be beneficial in increasing power to detect QTL given a limited sample size and magnitude of QTL effect.

• Journal of KIISE:Software and Applications
• /
• v.33 no.5
• /
• pp.499-507
• /
• 2006

Recently human gait has been considered as a useful biometric supporting high performance human identification systems. This paper proposes a view-based pedestrian identification method using the dynamic silhouettes of a human body modeled with the Hidden Markov Model(HMM). Two types of gait models have been developed both with an endless cycle architecture: one is a discrete HMM method using a self-organizing map-based VQ codebook and the other is a continuous HMM method using feature vectors transformed into a PCA space. Experimental results showed a consistent performance trend over a range of model parameters and the recognition rate up to 88.1%. Compared with other methods, the proposed models and techniques are believed to have a sufficient potential for a successful application to gait recognition.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.1
• /
• pp.257-263
• /
• 2011

Recently as AR(Augmented Reality) is focus of attention, AR is applied to various fields and is expected its valuable use. In this paper, we implemented the system based on Goblin XNA which supports high resolution model file and higher AR. We confirmed the relation of model output among the number of marker, the location and changes of camera distance. And we produced the infantile studying contents using AR and embodied. In implemented contents, we showed the familiar character to infants on each page marker. As the result of it, we can raise their concentration and at a time studying supporters can use the contents easily as well. Also we put 3 marker on each page of contents to recognize it smoothly in case one part of it is hidden by any obstacle. Finally we maximized the learning effect such as presence and immersion in studying through reinforcing 3D models according to the every situation.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.396-401
• /
• 2009

This paper describes the improvements made on hardware and software of the remote controller for realistic contents. The controller can provide vibrotactile feedback which uses both of a voice-coil actuator and a vibration motor. A vision tracking system for the 3D position of the controller is optimized with respect to the marker size and the camera parameters. We also present the improvements of motion recognition due to the effective motion segmentation and the fusion of vision and acceleration data. We apply the developed controller to realistic contents and validate its usability.