• Journal of Information Processing Systems
• /
• v.15 no.2
• /
• pp.399-409
• /
• 2019

Three-dimensional (3D) human pose reconstruction from single-view image is a difficult and challenging topic. Existing approaches mostly process frame-by-frame independently while inter-frames are highly correlated in a sequence. In contrast, we introduce a novel spatial-temporal 3D human pose reconstruction framework that leverages both intra and inter-frame relationships in consecutive 2D pose sequences. Orthogonal matching pursuit (OMP) algorithm, pre-trained pose-angle limits and temporal models have been implemented. Several quantitative comparisons between our proposed framework and recent works have been studied on CMU motion capture dataset and Vietnamese traditional dance sequences. Our framework outperforms others by 10% lower of Euclidean reconstruction error and more robust against Gaussian noise. Additionally, it is also important to mention that our reconstructed 3D pose sequences are more natural and smoother than others.

• Asian-Australasian Journal of Animal Sciences
• /
• v.25 no.7
• /
• pp.921-926
• /
• 2012

In order to protect the genetic resource of native horse breeds, the genetic diversity of mitochondrial DNA (mtDNA) D-loop of three native horse breeds in western China were investigated. Forty-three 600 bp mtDNA D-loop sequences were analyzed by PCR and sequencing techniques, 33 unique haplotypes with 70 polymorphic sites were detected in these horses, which account for 11.67% of 600 bp sequence analyzed, showing the abundant genetic diversity of the three native horse breeds in western China. The Neighbour-Joining (NJ) phylogenetic tree based on 247 bp of 43 D-loop sequences demonstrated the presence of seven major lineages (A to G), indicating that the three native horse breeds in western China originated from multiple maternal origins. Consistent with the front, the NJ phylogenetic tree based on 600 bp of mtDNA D-loop sequences of 43 Chinese western native horses and 81 sequences of six horse breeds from GenBank indicated that the three horse breeds had distributed into the seven major lineages (A to G). The structure of the phylogenic tree is often blurred because the variation in a short segment of the mitochondrial genome is often accompanied by high levels of recurrent mutations. Consequently, longer D-loop sequences are helpful in achieving a higher level of molecular resolution in horses.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.28-32
• /
• 2007

Sufficient Conditions are proposed herein for analytically obtaining fully orthotropic (A16=A26=0, Bij=0, D16=D26=0) laminate stacking sequences together with a brief literature survey. A number of specially chosen anti-symmetric balanced stacking sequences are analytically studied, in which all coupling stiffnesses including B16 and B26 are identically zero. Those specially chosen anti-symmetric balanced stacking sequences are then arranged symmetrically with respect to the laminate mid-plane to obtain a number of symmetric stacking balanced stacking sequences of which the elastic stiffnesses are fully orthotropic.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.53 no.7
• /
• pp.525-530
• /
• 2004

A classifier was constructed by using a generalized regression neural network (GRU) and random generator (RG), which was applied to classify DNA sequences. Three data sets evaluated are eukaryotic and prokaryotic sequences (Data-I), eukaryotic sequences (Data-II), and prokaryotic sequences (Data-III). For each data set, the classifier performance was examined in terms of the total classification sensitivity (TCS), individual classification sensitivity (ICS), total prediction accuracy (TPA), and individual prediction accuracy (IPA). For a given spread, the RG played a role of generating a number of sets of spreads for gaussian functions in the pattern layer Compared to the GRNN, the RG-GRNN significantly improved the TCS by more than 50%, 60%, and 40% for Data-I, Data-II, and Data-III, respectively. The RG-GRNN also demonstrated improved TPA for all data types. In conclusion, the proposed RG-GRNN can effectively be used to classify a large, multivariable promoter sequences.

• Korean Journal of Microbiology
• /
• v.31 no.3
• /
• pp.189-196
• /
• 1993

Nucleotide sequences of nodD and nodA from Bradyrhizobium sp. SNUOOI were determined. The open reading frame (ORF) of nodD was 942 bp in length and encoded 314 amino acids. while ORF of nodA, sequence of which is the first one among legume symbionts Bradyrhizobium, was 630 bp and encoded 210 amino acids. The nucleotide sequence of nodD showed 99.4% homology with nodDI of B. japonicum USDAllO. while that of nodA showed 81.5% with B. sp. (Parasponial. At the 5' of nodYAB operon and nodD, consensus nod box sequences composed of 9 bp unit repeated four times and two times respectively were found. Also an A.T-rich sequence was found at 5' of nodD.

• Journal of Korea Multimedia Society
• /
• v.22 no.6
• /
• pp.655-664
• /
• 2019

In this paper, we propose a novel way of combining multiple deep convolutional neural network (DCNN) architectures which work well for accurate video face identification by adopting a serial combination of 3D and 2D DCNNs. The proposed method first divides an input video sequence (to be recognized) into a number of sub-video sequences. The resulting sub-video sequences are used as input to the 3D DCNN so as to obtain the class-confidence scores for a given input video sequence by considering both temporal and spatial face feature characteristics of input video sequence. The class-confidence scores obtained from corresponding sub-video sequences is combined by forming our proposed class-confidence matrix. The resulting class-confidence matrix is then used as an input for learning 2D DCNN learning which is serially linked to 3D DCNN. Finally, fine-tuned, serially combined DCNN framework is applied for recognizing the identity present in a given test video sequence. To verify the effectiveness of our proposed method, extensive and comparative experiments have been conducted to evaluate our method on COX face databases with their standard face identification protocols. Experimental results showed that our method can achieve better or comparable identification rate compared to other state-of-the-art video FR methods.

• Journal of Broadcast Engineering
• /
• v.2 no.2
• /
• pp.125-135
• /
• 1997

This paper is concerned with realistic mainpulation of content in video sequences. Manipulation of content in video sequences is one of the content-based functionalities for MPEG-4 Visual standard. We present an approach to synthesizing video sequences by using the intermediate outputs of three-dimensional (3D) motion and depth analysis. For concreteness, we focus on video showing 3D motion of an observer relative to a scene containing planar runways (or roads). We first present a simple runway (or road) model. Then, we describe a method of identifying the runway (or road) boundary in the image using the Point of Heading Direction (PHD) which is defined as the image of, the ray along which a camera moves. The 3D motion of the camera is obtained from one of the existing 3D analysis methods. Then, a video sequence containing a runway is manipulated by (i) coloring the scene part above a vanishing line, say blue, to show sky, (ii) filling in the occluded scene parts, and (iii) overlaying the identified runway edges and placing yellow disks in them, simulating lights. Experimental results for a real video sequence are presented.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.12 no.1
• /
• pp.21-32
• /
• 2002

In this paper, for any prime q, new cyclic difference sets with Singer parameter equation omitted are constructed by using the q-ary sequences (d-homogeneous functions) of period $q_n$-1. When q is a power of 3, new cyclic difference sets with Singer parameter equation omitted are constructed from the ternary sequences of period $q_n$-1 with ideal autocorrealtion found by Helleseth, Kumar and Martinsen.

• Journal of applied mathematics & informatics
• /
• v.20 no.1_2
• /
• pp.541-548
• /
• 2006

We find a graded Artinian level O-sequence of the form $H\;:\;h_0\;h_1\;\cdots\;h_{d-1}\;h_d\cdots$ $^{(d+1-1_)-st}h_d$ < $h_{d+s}$ not having the Weak-Lefschetz property. We also introduce several algorithms for construction of some examples of non-unimodal level O-sequences using a computer program called CoCoA.

• Journal of Korea Technical Association of The Pulp and Paper Industry
• /
• v.29 no.2
• /
• pp.36-45
• /
• 1997

Newly bleaching sequence using oxygen, ozone and hydrogen peroxide has introduced to avoid pollution hazards from chlorinated organic compounds, because chlorine dioxide substitution bleaching was produced a little chlorinated organic substance. Oxygen-type chemicals replaced for chlorine has attracted much research attention. Bleachability of ozone was improved at low temperature and high pulp consistency. In third bleaching followed OZ bleaching, addition of O2 and NaBH4 in alkali extraction was effective than only alkali extraction. Bleachability of pulps in ozone bleaching(Z) was improved at higher consistency and lower temperature The addition O2 and NaBH4 in alkali extraction after OZ bleaching sequence improved brightness, when compared to those obtained by only alkaline extraction. Pulps bleached by ECF bleaching sequences such as OZEoD and OZEopD was obtained by 90% ISO brightness. The brightness of pulp bleached by TCF sequences with the ozone dosage of 1.6% was approached to target brightness (88~90%ISO). Pulps bleached Z stage combined bleaching sequence was reduced the viscosity to a little greater extent. However, physical properties of pulps was not great different compared to those bleached by conventional bleaching sequences. A pollution index of bleaching effluente by ozone combined bleaching sequences was lower than by conventional bleaching sequence, but somewhat higher than multistage bleaching sequences combined C/D stage.