Takiguchi, Tetsuya - Department of Computer and Systems Engineering, Kobe University

• (MIRU2009) 2009 7 E-mail: katsu0920@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• Random Projection Random Matrix
• Ball and Player Positional Estimation in 3D from Monocular Image Sequence Takuro Nishino
• AdaBoost Saliency Map Graph Cuts Automatic Segmentation of object region using Graph Cuts based on AdaBoost and Saliency Map
• 3 Real AdaBoost (VAD: Voice Activity Detection)
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Sudden Noise Reduction Based on GMM with Noise Power Estimation Nobuyuki Miyake, Tetsuya Takiguchi and Yasuo Ariki
• 6578501 11 E-mail: sakoats@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Extraction of Human Daily Activities from videos as Action Sequences using PrefixSpan Takuya Tonaru
• 3D Human Posture Estimation Using the HOG Features from Monocular Image Katsunori Onishi Tetsuya Takiguchi Yasuo Ariki
• IPSJ SIG Technical Report Acoustic Model Adaptation using Random Projection
• 2009 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS 2009) December 7-9, 2009 978-1-4244-5016-9/09/$25.00 c 2009 IEEE 445
• Confusion Network Confusion Network
• GRAPH CUTS BY USING LOCAL TEXTURE FEATURES OF WAVELET COEFFICIENT FOR IMAGE SEGMENTATION
• NOISE DETECTION AND CLASSIFICATION IN SPEECH SIGNALS WITH BOOSTING Nobuyuki Miyake, Tetsuya Takiguchi and Yasuo Ariki
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• pLSA[3] Topic HMM[4] (o1, , oN ) D = (W1, , WN )
• OP = HP (1) Estimation of sound source direction using active microphone with parabolic reflection board. by Ryoichi
• (MIRU2009) 2009 7 6578501 11
• Buried Markov Model Hidden Markov
• hk(r) = I(r)I(r + a 1 ) I(r + a
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• w(i, j) = wx(xi, xj)wd(di, dj) (2) Feature extraction using bilateral filter for noisy environment speech recognition. by YAMADA, Keishiro,
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• PCA (Principal Component Metamodel[4]
• Language Modeling using PLSA-Based Topic HMM Atsushi SAKO1
• (a) Training mode (b) Synthesis mode Estimate modeling
• System Request Detection in Human Conversation Based on Multi-Resolution Gabor Wavelet Features
• LSA One-Class SVM One-Class SVM[2]
• ISSN 1975-4736 MITA 20091 Situation Recognition Using 3D Positional Information of
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• MFCC(Mel-Frequency Cepstrum Coefficient) Effectiveness of bilateral filter for noisy speech recognition. by YAMADA, Keishiro, TAKIGUCHI, Tet-
• (Gaussian Mixture Model) HMM (Hidden Markov
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• INTEGRATION OF HMM COMPOSITION AND A MICROPHONE ARRAY FOR OVERLAPPING SPEECH RECOGNITION
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Improvement of In-Car Speech Recognition by Acoustic Echo Canceller with Maximum Likelihood
• (MIRU2009) 2009 7 6578501 11
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Image Annotation with Concept Level Feature Using PLSA+CCA
• Multimodal Speech Recognition of a Person with Articulation Disorders Using AAM and MAF
• Speech Synthesis by Modeling Harmonics Structure with Multiple Function Toru Nakashika1
• STRUCTURING A GENE NETWORK USING A MULTIRESOLUTION INDEPENDENCE Takayuki Yamamoto, Tetsuya Takiguchi and Yasuo Ariki
• Echo Canceller for Multi-Loudspeakers Based on Maximum Likelihood Using an Acoustic Model
• MATHEMATICAL MODELING OF HARMONIC-TIMBRE STRUCTURE WITH MULTI-BETA-DISTRIBUTION
• Estimation of Ground Surface Displacement from Microwave Radar Images by Using Phase-only Correlation
• Generic Object Recognition using CRF by Incorporating BoF as Global Features Takeshi Okumura*
• 3D Human Posture Estimation Using the HOG Features from Monocular Image
• Audio-Based Video Editing with Two-Channel Microphone Tetsuya Takiguchi
• Multiple Classifier Based on Fuzzy C-Means for a Flower Image Retrieval Keita Fukuda
• Tagging Video Contents with Positive/Negative Interest Based on User's Facial Expression
• ESTIMATION OF ROOM ACOUSTIC TRANSFER FUNCTION USING SPEECH MODEL Tetsuya Takiguchi, Yuji Sumida, Yasuo Ariki
• PCA-Based Feature Extraction for Fluctuation in Speaking Style of Articulation Disorders
• ROBUST FEATURE EXTRACTION USING KERNEL PCA Tetsuya Takiguchi and Yasuo Ariki
• Two-Channel-Based Noise Reduction in a Complex Spectrum Plane
• Recognition of Hands-free Speech and Hand Pointing Action for Conversational TV
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• (MIRU2010) 2010 7 6578501 11
• (MIRU2009) 2009 7 Gaussian Processes for Regression AAM
• (MIRU2010) 2010 7 657-8501 1-1
• (MIRU2010) 2010 7 6578501 11
• (MIRU2010) 2010 7 E-mail: {katsu0920,bogeli}@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Regular Paper Extracting Why Text Segment from Web
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• IPSJ SIG Technical Report Buried Markov Model
• (MIRU2009) 2009 7 6578501 1-1
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• (MIRU2008) 2008 7 E-mail: tonaru@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• (MIRU2008) 2008 7 E-mail: katsu0920@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• (MIRU2008) 2008 7 SIFT Graph Cuts
• (MIRU2008) 2008 7 AdaBoost Saliency Map Graph Cuts
• NetTv: Cross-Platform Video Retrieval and QA System with Speech Interface Katsuyuki Tanaka.1
• 6578501 11 6578501 11
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• (MIRU2007) 2007 7 E-mail: {masudaken,masamax777,matsuda}@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• (MIRU2007) 2007 7 E-mail: fukuda@me.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp
• (MIRU2007) 2007 7 657-8501 1-1
• 6578501 11 6578501 11
• 6578501 11 6578501 11
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Generic Object Recognition by Tree Conditional Random Field based on Hierarchical Segmentation Takeshi Okumura
• Buried Markov Model J. Bilmes Buried Markov
• GENERIC OBJECT RECOGNITION BASED ON WEIGHTED INTEGRATION OF MULTIPLE FEATURES
• Buried Markov Model Buried Markov Model (BMM) [1]
• A Study on Dysarthric Speech Recognition using Local Features, by Chikoto Miyamoto, Tetsuya Takiguchi, Yasuo Ariki (Kobe Univ.), Ichao Li (Otemon Univ.) and Toshitaka Nakabayashi (Kobe Univ.)
• Estimation of Ground Surface Displacement from SAR Satellite Image Using High-Accuracy Image Matching
• Content Analysis based on Human Face Images Tomoko Okada Tetsuya Takiguchi Yasuo Ariki
• [1] 2ch 1ch xL xR 2ch h
• STC(Spectro-Temporal structured Clustering)[1]
• Fig. 1(b) x0(t) O xm(t)(m = 1, , M)
• Pose Robust and Person Independent Facial Expressions Recognition using AAM of Model Selection Tomoko Okada
• Bottom-Up Top-Down 3D Human Pose Estimation Integrating Bottom-Up and Top-Down Approach from Monocular Image
• Fig. 2 x0(t) O xn(t)(n = 1, , N)
• SVM CART AdaBoost (VAD: Voice Activity Detection)
• SIFT Graph Cuts OBJECT RECOGNITION AND SEGMENTATION USING SIFT AND GRAPH CUTS
• System Request Fig. 1 One System + Two individuals dialog
• x(t) x(t)h(t) h(t) (1) [1]
• (Eigen Phoneme Space (Phoneme Vector : PV)
• (VAD: Voice Activity Detection) 3 3rd order cumu-
• Boosting AdaBoost AdaBoost xi
• O(; t) = H()S(; t) (1) log O(; t) = log H() + log S(; t) (2)
• PLSA(Probabilistic Latent Semantic Analysis) P(w|z)P(z|d) (1)
• Delay-and-Sum(DS) Griffith-Jim
• EVALUATION OF RANDOM-PROJECTION-BASED FEATURE COMBINATION ON SPEECH RECOGNITION
• DTA-Kernel PCA Latent Semantic Analysis (LSA)
• Object Recognition and Segmentation Using SIFT and Graph Cuts Keita Fukuda
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• GENERIC OBJECT RECOGNITION BASED ON WEIGHTED INTEGRATION OF MULTIPLE FEATURE Tetsuya Takiguchi
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• HMM-BASED SEPARATION OF ACOUSTIC TRANSFER FUNCTION FOR SINGLE-CHANNEL SOUND SOURCE LOCALIZATION
• Voice Activity Detection by Lip Shape Tracking Using Masaki Aoki
• (MIRU2009) 2009 7 6578501 11
• Tagging for Video Contents Based on User's Facial Expression Clustering Masanori Miyahara Masaki Aoki Tetsuya Takiguchi Yasuo Ariki
• {o1 ot} W = {w1 wN } ^W = argmax
• Subharmonic Summation (SHS) [1] [2]
• System Request Detection in Conversation Based on Acoustic and Speaker Alternation Features
• (MIRU2010) 2010 7 6578501 1-1
• O(; n) H()S(; n) (1) Ocep(d; n) Hcep(d) + Scep(d; n) (2)
• (MIRU2010) 2010 7 6578501 11
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• Latent Semantic Analysis (LSA)[1] Probabilistic LSA (pLSA)
• ( ), , ( IBM), , ( ) Concatenative
• FBANK Gabor Wavelet Gabor Wavelet
• Nilsback [4] port Vector Machine)
• (MIRU2010) 2010 7 Image Annotation by Concept Level Search Using PLSA
• SINGLE-CHANNEL MULTI-TALKER-LOCALIZATION BASED ON MAXIMUM Ryoichi Takashima, Tetsuya Takiguchi and Yasuo Ariki
• Real Adaboost (VAD: Voice Activity Detec-
• (MIRU2006) 2006 7 E-mail: {masudaken,matsuda,inoue}@me.cs.scitec.kobe-u.ac.jp, {ariki,takigu}@kobe-u.ac.jp,
• 242-8502 1623 14 E-mail: ftakigu, nisimurag@jp.ibm.com
• FACE AND GAZE ANGLE ESTIMATION USING AAM AND REGRESSION Manabu Takatani
• Bottom-up Top-down Katsunori Onishi Tetsuya Takiguchi Yasuo Ariki
• (MIRU2007) 2007 7 6578501 11
• i (i = 1 M) i i (; n) (1)
• 2. M(w) wi 1 M(w) sim(wi, wj)
• HMM (Hidden Markov Model) O(; n) H(; n)S(; n) (1)
• (MIRU2010) 2010 7 657-8501 1-1
• Pose Robust and Person Independent Facial Expressions Recognition Using AAM Selection
• A Fast Algorithm for Eye Detection Using Two-Dimensional CSP Akiko SUZUKI Tetsuya TAKIGUCHI Yasuo ARIKI
• score-based approach Structuring Gene Network Using Multiresolution Independence
• Speaker Independent Phoneme Recognition Based on Fisher Weight Map Takashi Muroi, Tetsuya Takiguchi, Yasuo Ariki
• Situation Based Speech Recognition for Structuring Baseball Live Games Atsushi Sako, Tetsuya Takiguchi and Yasuo Ariki
• ACOUSTIC MODEL ADAPTATION USING FIRST ORDER PREDICTION FOR REVERBERANT SPEECH
• Gaze Estimation Using Regression Analysis and AAMs Parameters Selected Based on
• (MIRU2010) 2010 7 6578501 11
• (Probabilistic Latent Semantic Analysis) P(w|z)P(z|d) (1)
• ( ) Jeff BILMES(University of Washington) Speech Feature Extraction Using Random Projection, by Mariko YOSHII, Tetsuya TAKIGUCHI, Yasuo
• O(; n) H(; n)S(; n) (1) Ocep(d; n) Hcep(d; n) + Scep(d; n) (2)
• Generic Object Recognition by Tree Conditional Random Field based on Hierarchical Segmentation
• Automatic Segmentation of Object Region Using Graph Cuts Based on Saliency Maps and AdaBoost
• Integration of Metamodel and Acoustic Model for Speech Recognition Hironori Matsumasa1
• MFCC (Mel-Frequency Cepstral Coefficient) SVM (Support Vector Machine)
• Human-Robot Interface Using System Request Utterance Detection Based on Acoustic Features
• DIGITAL CAMERA WORK FOR SOCCER VIDEO PRODUCTION WITH EVENT RECOGNITION AND ACCURATE BALL TRACKING BY SWITCHING SEARCH METHOD
• (1 ) (0 1) N(0,1) n k
• Estimation of Sound Source Direction Using Parabolic Reflection Board Tetsuya Takiguchi
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• m1(t) m2(t) m1(t) = s(t) + n(t)
• PCA (Principal Component Analysis)
• (MIRU2011) 2011 7 ActiveAppearanceModel
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• [1, 2, 3] Spe-v(x) = h(x) u(x) (1)
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• (MIRU2011) 2011 7 6578501 11
• CRF Confusion Network Confusion Network
• (MIRU2011) 2011 7 6578501 11
• Confusion Network CRF CRF (Conditional Random Field)
• NMF Matrix Generation Using Probabilistic Spectrum Envelope for Mixed Music Analysis
• SIC (MUltiple SIgnal Classification) CSP (Cross-power Spectrum Phase)
• THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS
• FEATURE SELECTION BASED ON MULTIPLE KERNEL LEARNING FOR SINGLE-CHANNEL SOUND SOURCE LOCALIZATION USING THE ACOUSTIC TRANSFER
• (Probabilistic Spectrum Envelope; PSE) 0 1000 2000 3000 4000 5000 6000 7000 8000-0.5
• GENERIC OBJECT RECOGNITION USING AUTOMATIC REGION EXTRACTION AND DIMENSIONAL FEATURE INTEGRATION UTILIZING MULTIPLE KERNEL LEARNING
• (MIRU2011) 2011 7 6578501 11
• Single-channel Head Orientation Estimation Based on Discrimination of Acoustic Transfer Function
• Probabilistic Spectrum Envelope: Categorized Audio-features Representation for NMF-based Sound Decomposition
• SIC (MUltiple SIgnal Classification) CSP (Cross-power Spectrum Phase)
• Statistical voice conversion based on GMM for articulation disorders, by Ryo Ishii, Tetsuya Takiguchi, Yasuo Ariki (Kobe University)
• 12th International Society for Music Information Retrieval Conference (ISMIR 2011) CONSTRAINED SPECTRUM GENERATION USING A PROBABILISTIC
• xL xR 2ch h (FL FR RL RR)
• F0 Specmurt Specmurt[1]
• Conditional Ran-dom Fields (CRF) [1]
• SIC (MUltiple SIgnal Classification) CSP (Cross-power Spectrum Phase)
• Audio-Visual Speech Recognition Based on AAM Parameter and Phoneme Analysis