Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
MI |
2023-03-06 16:25 |
Okinawa |
OKINAWA SEINENKAIKAN (Primary: On-site, Secondary: Online) |
Domain Generalization for Mitosis Detection Yusuke Kurose (U-Tokyo), Yuko Hiroshima, Hiroshi Nanjo (Akita-U), Tatsuya Harada (U-Tokyo) MI2022-93 |
In pathological image analysis, it is known that the existence of inter-institutional differences (domain shift) causes ... [more] |
MI2022-93 pp.100-104 |
PRMU, CNR |
2019-02-28 10:50 |
Tokushima |
|
PRMU2018-113 CNR2018-36 |
[more] |
PRMU2018-113 CNR2018-36 p.3 |
PRMU, CNR |
2019-02-28 11:10 |
Tokushima |
|
PRMU2018-114 CNR2018-37 |
Traditional image recognition methods only consider objects belonging to already learned classes. However, since trainin... [more] |
PRMU2018-114 CNR2018-37 p.5 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2017-09-15 11:00 |
Tokyo |
|
Alternating Circulant Random Features for Semigroup Kernels Yusuke Mukuta, Yoshitaka Ushiku, Tatsuya Harada (UTokyo) PRMU2017-42 IBISML2017-14 |
In this paper, we propose novel random features termed ``alternating circulant random features,'' which are comprised of... [more] |
PRMU2017-42 IBISML2017-14 pp.27-34 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2017-09-15 09:30 |
Tokyo |
|
PRMU2017-43 IBISML2017-15 |
(To be available after the conference date) [more] |
PRMU2017-43 IBISML2017-15 pp.35-42 |
PRMU |
2015-10-08 13:05 |
Chiba |
|
[Special Talk]
Recent Trend in Image Recognition Tatsuya Harada (UTokyo) PRMU2015-92 |
[more] |
PRMU2015-92 pp.1-20 |
PRMU |
2014-10-09 15:00 |
Chiba |
|
Learning Similarities for Rigid and Non-Rigid Object Detection Asako Kanezaki (Univ. of Tokyo), Emanuele Rodola, Daniel Cremers (Technische Univ. Munchen), Tatsuya Harada (Univ. of Tokyo) PRMU2014-56 |
We propose an optimization method for estimating parameters in graph-theoretical formulations of the matching problem fo... [more] |
PRMU2014-56 pp.13-18 |
IBISML |
2013-11-13 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Probabilistic Partial Canonical Correlation Analysis Yusuke Mukuta, Tatsuya Harada (Univ. of Tokyo) IBISML2013-58 |
Partial Canonical Correlation Analysis (Partial CCA) is a statistical method to estimate a pair of linear projections on... [more] |
IBISML2013-58 pp.169-176 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2013-09-03 11:00 |
Tottori |
|
Canonical Contextual Anomaly Detection for Multi-modal Data Chie Kamada, Tatsuya Harada (Univ. of Tokyo) PRMU2013-48 IBISML2013-28 |
In prior studies of anomaly detection, relationships between modalities have not been considered much. In this paper, we... [more] |
PRMU2013-48 IBISML2013-28 pp.167-172 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2013-09-03 16:15 |
Tottori |
|
Image annotation from wild using semantic hierarchies Masatoshi Hidaka, Naoyuki Gunji, Tatsuya Harada (Univ. of Tokyo) PRMU2013-52 IBISML2013-32 |
For many-class image annotation system, automatic construction of large-scale dataset is needed. Wild web is regarded as... [more] |
PRMU2013-52 IBISML2013-32 pp.201-206 |
PRMU |
2013-03-14 09:30 |
Tokyo |
|
Fine-grained classification with various local descriptors and Passive-Aggressive Naoyuki Gunji, Takayuki Higuchi, Koki Yasumoto, Hiroshi Muraoka, Yoshitaka Ushiku (Univ. of Tokyo), Tatsuya Harada (Univ. of Tokyo/JST), Yasuo Kuniyoshi (Univ. of Tokyo) PRMU2012-184 |
[more] |
PRMU2012-184 pp.25-30 |
PRMU, IBISML, IPSJ-CVIM (Joint) [detail] |
2012-09-03 09:30 |
Tokyo |
|
Simultaneous training of multi-class object detectors via large scale image dataset
-- introduction of target specific negative classes -- Asako Kanezaki, Sho Inaba, Yoshitaka Ushiku, Yuya Yamashita, Hiroshi Muraoka, Tatsuya Harada, Yasuo Kuniyoshi (Univ. Tokyo) PRMU2012-42 IBISML2012-25 |
We propose an efficient method to train multiple object detectors simultaneously using a large-scale image dataset. The ... [more] |
PRMU2012-42 IBISML2012-25 pp.105-112 |
PRMU |
2011-02-17 10:00 |
Saitama |
|
Large Scale Image Classification using Metric based on Correlation between Multiple Image Features and Class Labels Yoshitaka Ushiku, Yuya Yamashita, Jun Imura, Hideki Nakayama, Tatsuya Harada, Yasuo Kuniyoshi (Tokyo Univ.) PRMU2010-208 |
In this paper, we propose a scalable image recognition method for both image classification and image annotation. We fir... [more] |
PRMU2010-208 pp.1-6 |
PRMU |
2011-02-17 17:10 |
Saitama |
|
[Special Talk]
Large-Scale Generic Image Recognition and Image Representation Tatsuya Harada (The Univ. of Tokyo) PRMU2010-218 |
In recent years, it is becoming possible to obtain a large amount of image and additional information (e.g., tags) from ... [more] |
PRMU2010-218 pp.61-74 |
PRMU |
2009-11-26 11:10 |
Ishikawa |
Ishikawa Industrial Promotion Center |
Semantics-Based Search from Global Features of Web Images and Texts
-- Toward a novel approach to realization of generic image recognition on the Web -- Yoshitaka Ushiku, Hideki Nakayama (Tokyo Univ.), Tatsuya Harada (Tokyo Univ./JST), Yasuo Kuniyoshi (Tokyo Univ.) PRMU2009-100 |
In these days, methods for generic image recognition are actively researched. Many works use web images as training data... [more] |
PRMU2009-100 pp.45-50 |
PRMU |
2009-11-27 10:00 |
Ishikawa |
Ishikawa Industrial Promotion Center |
Multiple Weighted Region Features for Fast Object Recognition Tatsuya Harada (The Univ. of Tokyo/JST), Yasuo Kuniyoshi (The Univ. of Tokyo) PRMU2009-108 |
[more] |
PRMU2009-108 pp.109-114 |
PRMU |
2009-11-27 13:30 |
Ishikawa |
Ishikawa Industrial Promotion Center |
Application of High-speed 3D Object Recognition in the Real World Using Integral Features and Linear Subspace Method Asako Kanezaki, Tatsuya Harada, Yasuo Kuniyoshi (Univ. of Tokyo.) PRMU2009-123 |
As a previous work, we proposed a method of high-speed 3D object recognition using linear subspace method and our 3D fea... [more] |
PRMU2009-123 pp.207-212 |
PRMU |
2007-12-14 11:00 |
Hyogo |
Kobe Univ. |
Ultra high speed image annotation/retrieval method by learning the conceptual relationship between images and labels Hideki Nakayama, Tatsuya Harada, Yasuo Kuniyoshi (Tokyo Univ.), Nobuyuki Otsu (AIST) PRMU2007-147 |
Content-based image recognition and retrieval are challenging
problems which have wide application.
In this paper, we ... [more] |
PRMU2007-147 pp.65-70 |
PRMU, HIP |
2006-01-20 12:15 |
Osaka |
Osaka Univ. |
Learning Environment Responding to User's Behavior Mehrab Hosseinbor, Taketoshi Mori, Tatsuya Harada, Tomomasa Sato (Tokyo Univ.) |
[more] |
PRMU2005-158 HIP2005-128 pp.53-58 |
PRMU, HIP |
2006-01-20 15:40 |
Osaka |
Osaka Univ. |
Development of Individual Identification System utilizing Time Series of Floor Pressure Distribution Toshiaki Yamanishi, Taketoshi Mori, Hiroshi Morishita, Tatsuya Harada, Tomomasa Sato (Tokyo Univ.) |
This paper proposes an individual identification system utilizing the change of foot-floor contact area shape during wal... [more] |
PRMU2005-167 HIP2005-137 pp.105-110 |