Faculty Profiles - Deguchi Daisuke

Presentations -

Division display 421 - 440 of about 727 ／ All the affair displays >>

Removal of Moving Objects From Frontal In-vehicle Camera Videos Based on Adaptive Selection of Reference Image

Toru Kotsuka, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

17th Meeting on Image Recognition & Understanding (MIRU2014) 2014.7.29

A Collaborative Report on The Open Apereo 2013 Conference

Yuji Tokiwa, Makoto Miyazaki, Daisuke Deguchi, Naoshi Hiraoka, Shoji Kajita

IPSJ SIG Technical Report (CLE) 2013.12.15

複数フレーム型と学習型の組み合わせによる低解像度顔画像の超解像

吉田智成, 高橋友和, 出口大輔, 井手一郎, 村瀬洋

動的画像処理実利用化ワークショップ(DIA) 2013.3.7

Event date： 2013

Language：Japanese Presentation type：Oral presentation (general)

Venue：静岡大学浜松キャンパス

監視カメラ映像中の顔画像認識において，顔領域が低解像度であることは認識精度の低下を招く．この問題に対して我々は，顔向きの変化を利用した複数フレーム型超解像手法を提案してきた．しかし，複数フレーム型超解像には，画像が低解像度になるにつれ，必要な高周波成分を復元することが困難になるという問題がある．そこで本発表では，複数フレーム型超解像と学習型超解像を組み込み合わせた手法を提案する．学習型超解像は，大量の顔画像から高解像度化に必要な顔の知識を学習し，それを用いて低解像度画像から高解像度画像を生成する技術である．これにより，複数フレーム型超解像では復元しきれなかった高周波成分の復元を図る．実際に撮影した動画像を用いた実験の結果から提案手法の有効性を確認した．

An evaluation of a visibility estimation method for traffic signals under rainy weather condition using a large dataset.

Ryuhei Sato, Keisuke Doman, Daisuke Deguchi, Yoshito Mekada, Ichiro Ide, Hiroshi Murase, Naoki Nitanda

Dynamic Image Processing for Real Application workshop (DIA2013) 2013.3.7

Pedestrian detection with iterative generative learning

Hidefumi Yoshida, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase, Takashi Machida, Yoshiko Kojima

Dynamic Image Processing for Real Application workshop (DIA2013) 2013.3.7

Detection of Biased Broadcast Sports Video Highlights by Attribute-Based Tweets Analysis International conference

Takashi Kobayashi, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

Lecture Note in Computer Science 2013.1.9

Event date： 2013

Language：English Presentation type：Poster presentation

Venue：黄山柏瑞酒店（中華人民共和国安徽省黄山市）

We propose a method for detecting biased-highlights in a broadcast sports video according to viewers’ attributes obtained from a large number of tweets. Recently, Twitter is widely used to make real-time play-by-play comments on TV programs, especially on sports games. This trend enables us to effectively acquire the viewers’ interests in a large mass. In order to make use of such tweets for highlight detection in broadcast sports video, the proposed method first performs an attribute analysis on the set of tweets issued by each user to classify which team he/she supports. It then detects biased-highlights by referring to the number of tweets made by viewers with a specific attribute.

DOI： 10.1007/978-3-642-35728-2_35

A Study of Gaze Estimation Using Head and Body Pose Information International conference

Nobuhiro Funatsu, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

International Workshop on Advanced Image Technology (IWAIT) 2013 2013.1.8

Event date： 2013

Language：English Presentation type：Oral presentation (general)

Venue：Nagoya University

Gaze estimation from an image is an important technique for tasks such as driver monitoring and measuring of advertising effectiveness. For this, most existing methods require a high quality image of eyes. However, It is difficult to obtain the eye images when eye occlusion occurs due to sunglasses or face rotation. Another approach approximates head directions to gaze directions. However, the gaze direction is affected not only by the head direction but also by the body direction. Thus, we are studying a method that estimates the gaze direction accurately using information on both head and body pose directions. Experimental result showed that the proposed method could estimate gaze directions more accurately than by using the information on only head directions.

Automatic Construction of Traffic Sign Detector Based on Simultaneous Gathering of Training Samples

Daisuke Deguchi, Hiroshi Murase

International Workshop on Advanced Image Technology (IWAIT) 2013 2013.1.7

Event date： 2013

Language：English Presentation type：Oral presentation (general)

Venue：Nagoya University, Japan

This paper proposes a method for automatic construction of traffic sign detector based on simultaneous gathering of training samples from in-vehicle camera image sequences. Most existing methods for object detection based on AdaBoost classifier require various training samples for constructing an accurate detector. In the case of traffic sign detection, traffic sign images having various appearances should be prepared as positives before constructing the detector to achieve accurate traffic sign detection from in-vehicle camera images. In addition to the traffic sign images, various background images should also be prepared as negatives to reduce false alarms. However, since they are prepared manually by specifying their clipping rectangles in in-vehicle camera images, their preparation cost is quite expensive. To solve this problem, our group has proposed a method for obtaining positive and negative training samples automatically from in-vehicle camera image sequences. By extending our previous work, this paper presents a method of automatic construction of traffic sign detector that uses the state-of-the-art image features and Ada-Boost classifier. We conducted experiments with in-vehicle camera images containing traffic signs with various backgrounds. Experimental results showed the effectiveness of the proposed method.

A study on detection of belongings from a human image using their category and direction

Yasuhiro Asai, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

IEICE Technical Report (PRMU) 2013.12.13

Summarizing a Domestic Cooking Video According to the Description of a Cooking Recipe

Yasuhiro Hayashi, Keisuke Doman, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

IEICE Technical Report (MVE) 2013.3.11

A Study on Exemplar Based Super Resolution for Low Resolution Images of Human Body

Kento Nishibori, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

IEICE Technical Report (PRMU) 2013.9.3

TransifexによるSakai CLE 翻訳

出口大輔

Ja Sakai アンカンファレンス 2013.8.21

Exemplar-Based Super-Resolution for Low-Resolution Images of Human Body

Kento Nishibori, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

16th Meeting on Image Recognition & Understanding (MIRU2013) 2013.7.31

エピポーラ幾何に基づく画像間距離を用いた車載カメラ映像データベース探索による自車位置推定

久徳遙矢, 出口大輔, 高橋友和, 目加田慶人, 井手一郎, 村瀬洋

画像の認識・理解シンポジウム(MIRU) 2013.7.30

A study on possessing detection of belongings from human image using human direction

Yasuhiro Asai, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

16th Meeting on Image Recognition & Understanding (MIRU2013) 2013.7.1

Localization of point heat sources using the spatial characteristic of a sensor in an infrared sensor array

Takashi Hosono, Tomokazu Takahashi, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

16th Meeting on Image Recognition & Understanding (MIRU2013) 2013.7.1

Automatic Pedestrian Segmentation Using Grab-cut and Active Shape Model Feedback

Esmaeil Pourjam, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase

16th Meeting on Image Recognition & Understanding (MIRU2013) 2013.7

Pedestrian Detection by Scene Dependent Classifiers with Generative Learning International conference

Hidefumi Yoshida, Daichi Suzuo, Daisuke Deguchi, Ichiro Ide, Hiroshi Murase, Takashi Machida, Yoshiko Kojima

2013 IEEE Intelligent Vehicles Symposium (IV2013) 2013.6.25

Event date： 2013

Language：English Presentation type：Poster presentation

Venue：Gold Coast, Australia

Recently, pedestrian detection from in-vehicle camera images is becoming an crucial technology for Intelligent Transportation Systems (ITS). However, it is difficult to detect pedestrians accurately in various scenes by obtaining training samples. To tackle this problem, we propose a method to construct scene dependent classifiers to improve the accuracy of pedestrian detection. The proposed method selects an appropriate classifier based on the scene information that is a category of appearance associated with location information. To construct scene dependent classifiers, the proposed method introduces generative learning for synthesizing scene dependent training samples. Experimental results showed that the detection accuracy of the proposed method outperformed the comparative method, and we confirmed that scene dependent classifiers improved the accuracy of pedestrian detection.

Surrounding environment recognition using in-vehicle camera

Daisuke Deguchi, Keisuke Doman, Ichiro Ide, Hiroshi Murase

2013.6.10

Segmentation of Human Instances Using Grab-cut and Active Shape Model Feedback International conference

Esmaeil Pourjam, Ichiro Ide, Daisuke Deguchi, Hiroshi Murase

IAPR Conference on Machine Vision Applications (MVA) 2013 2013.5.21

　 PREV - NEXT 　

<Deguchi Daisuke>

Presentations -