Naoya Chiba / 千葉 直也

Selected Publications

Sparse Estimation of Light Transport Matrix under Saturated Condition

Naoya Chiba, Koichi Hashimoto. British Machine Vision Conference (BMVC), 2018.

Ultra-Fast Multi-Scale Shape Estimation of Light Transport Matrix for Complex Light Reflection Objects

Naoya Chiba, Koichi Hashimoto. IEEE International Conference on Robotics and Automation (ICRA), 2018. Awarded "IEEE RAS-JJC Young Award (ICRA2018)")

3D Measurement by Estimating Homogeneous Light Transport (HLT) Matrix

Naoya Chiba, Koichi Hashimoto. IEEE International Conference on Mechatronics and Automation (ICMA), 2017.

Feedback Projection for 3D Measurements Under Complex Lighting Conditions

Naoya Chiba, Shogo Arai, Koichi Hashimoto. American Control Conference (ACC), 2017.

Profile

About me

Ph. D. Student at Tohoku University, Japan.

I belong to Hashomoto-Lab (Intelligent Control Systems Lab.), Graduate School of Information Sciences (GSIS).

Research Field: Robot Vision, 3D Measurement, Industrial Robots

Research Interests: Light Transport on Projector-Camera System, Pointcloud Processing, Application of Pointcloud Processing methods to Biochemistry

Biography / Education

  • 2018.04 - : JSPS Research Fellow (DC1)
  • 2018.04 - : Ph. D. Student
System Information Science, Graduate School of Informaion Science, Tohoku University.
  • 2016.04 - 2018.03: Master of Information Science
System Information Science, Graduate School of Informaion Science, Tohoku University.(Master Thesis: 3D Shape Estimation Based on Sparse Sensing of Light Transport (LT) Matrix)
  • 2012.04 - 2016.03: Bachelor of Engineering
Department of Mechanical and Aerospace Engineering, School of Engineering, Tohoku University.(Graduation Thesis: Adaptive 3D Measurement for Target Objects Optical Specification)
  • 2009.04 - 2012.03: High School
Department of Science and Mathematics, Iwate Prefectural Morioka First High School.

Work / Teaching Experience

  • 2017.04 - 2018.03: Research Assistant, Tohoku University
  • 2016.09 - 2017.03: Teaching Assistant, Tohoku University
Student Experiment II, Department of Mechanical and Aerospace Engineering
  • 2013.12 - 2014.05: Web Developer, GENGOCROSS

Projects / Grants

  • 2018.04 - 2020.03: JST ACT-I Acceleration Phase
Title: 3D Measurements Based on Estimation Light Transport Matrix between Camera and Projector
  • 2018.04 - 2021.03: JSPS DC1
Title: General Purpose Robot Vision System via Dynamic Projection MeasurementTitle: 3D Measurements Based on Estimation Light Transport Matrix between Camera and Projector
  • 2017.06 - 2018.03: Student Project, GSIS, Tohoku University
Title: Large Scale Exploration of Peptizymes Using 3D Structure MatchingCollaborated with Yamato Komatsu in Suga Lab., Department of Chemistry, The University of Tokyo and Yuki Kagaya in Kinoshita and Obayashi Lab. Department of Applied Information Sciences, Tohoku University.
  • 2016.06 - 2017.03: Student Project, GSIS, Tohoku University
Title: Large Scale Exploration of Peptizymes Using 3D Structure MatchingCollaborated with Yamato Komatsu in Suga Lab., Department of Chemistry, The University of Tokyo.

Publications (International)

Journal

  • Naoya Chiba, Koichi Hashimoto. Homogeneous light transport matrix estimation based 3D shape measurement. International Journal of Mechatronics and Automation (IJMA), Vol. 6, No. 2/3, pp.63-70. DOI: 10.1504/IJMA.2018.10015629. (Refereed)

Conference / Symposium

  • Naoya Chiba, Koichi Hashimoto. Sparse Estimation of Light Transport Matrix under Saturated Condition. The 29th British Machine Vision Conference (BMVC2018), Posters 1-68, Newcastle, UK. September 3rd-6th (Presentation in 4th), 2018. (Refereed, Poster)
  • Naoya Chiba, Koichi Hashimoto. LTM (Light Transport Matrix) estimation method under saturated condition by using Saturation ADMM. The 5th Case Western Reserve University - Tohoku University Joint Workshop, Poster 006, pp. 50, Sendai, Japan, August 2nd-3rd (Presentation in 2nd), 2018. (Non-refereed, Poster, Awarded "Best Poster Presentation Award")
  • Naoya Chiba, Koichi Hashimoto. Ultra-Fast Multi-Scale Shape Estimation of Light Transport Matrix for Complex Light Reflection Objects. The 2018 IEEE International Conference on Robotics and Automation (ICRA2018), ThA@O.5, Brisbane, Australia, May 21st-25th (Presentation in 24th) 2018. (Refereed, Poster, Awarded "2018 IEEE Robotics and Automation Society Japan Joint Chapter Young Award (ICRA2018)")
  • Naoya Chiba, Koichi Hashimoto. Fast Light Transport Matrix Estimation for 3D Shape Acquisition. Graduate Program in Data Science International Kick-off Symposium, Poster 20, pp. 27, Sendai, Japan, February 22nd-23rd (Presentation in 23rd), 2018. (Non-refereed, Poster, Awarded "Best Poster Presentation Award")
  • Naoya Chiba, Koichi Hashimoto, 3D Measurement by Estimating Homogeneous Light Transport (HLT) Matrix. The 2017 IEEE International Conference on Mechatronics and Automation (ICMA2017), WA2-2(3), pp. 1763-1768, Takamatsu Japan. August 6th-9th (Presentation in 9th), 2017. DOI: 10.1109/ICMA.2017.8016084. (Refereed, Oral)
  • Naoya Chiba, Koichi Hashimoto, 3D Measurement by Estimating Light Transport Matrix. The third Annual Data and Life Science Collaboration and Symposium. Cleveland, USA. August 3rd-4th (Presentation in 3rd and 4th), 2017. (Non-refereed, Poster)
  • Naoya Chiba, Shogo Arai, Koichi Hashimoto, Feedback Projection for 3D Measurements Under Complex Lighting Conditions. The 2017 American Control Conference (ACC2017), FrA18.5, Seattle, USA, May 24th-26th (Presentation in 26th), 2017, pp. 4649-4656. DOI: 10.23919/ACC.2017.7963673. (Refereed, Oral)

Publications (Domestic)

Journal

  • [採録決定] 千葉直也,橋本浩一.LT行列行分解推定による三次元計測.計測自動制御学会論文集.(査読あり)
Naoya Chiba, Koichi Hashimoto. 3D Measurement Based on Row-wise LT Matrix Estimation. Transactions of the Society of Instrument and Control Engineers. (Refereed)
  • 福地 伸晃,千葉直也,橋本浩一.CADモデルを用いないばら積みシーンからの物体検出.日本ロボット学会誌,Vol. 36,No. 5,pp. 368-375,2018.DOI: 10.7210/jrsj.36.368.(査読あり)
Nobuaki Fukuchi, Naoya Chiba, Koichi Hashimoto. Object Detection in Bin Scene without CAD Model. Journal of the Robotics Society of Japan, 2018, Vol. 36, No. 5, pp. 368-375. DOI: 10.7210/jrsj.36.368. (Refereed)

Conference / Symposium

  • 戸田幸宏,千葉直也,橋本浩一.点群深層学習を用いた3次元物体の位置・姿勢推定.電子情報通信学会パターン認識・メディア理解(PRMU) 研究会 (2019-02-PRMU-CNR),PRMU2018-123 ,徳島,2月28日-3月1日(発表1日),2019.(査読なし,口頭発表)
Yukihiro Toda, Naoya Chiba, Koichi Hashimoto. Point cloud deep learning for estimating 3-D rigid body transformation. The Institute of Electronics, Information, and Communication Engineers (IEICE), Pattern Recognition and Media Understanding (PRMU) 2019-02-PRMU-CNR, PRMU2018-123, Tokushima, Japan, February 28th-March 1st (Presentation in 1st), 2019.
  • 千葉直也,今倉暁,橋本浩一.SMW公式を用いたADMM-L1最小化問題の高速解法.電子情報通信学会パターン認識・メディア理解(PRMU) 研究会 (2019-02-PRMU-CNR),PRMU2018-131,徳島,2月28日-3月1日(発表1日),2019.(査読なし,(代理)口頭発表)
Naoya Chiba, Akira Imakura, Koichi Hashimoto. Fast ADMM-L1 minimization method by using SMW formula. The Institute of Electronics, Information, and Communication Engineers (IEICE), Pattern Recognition and Media Understanding (PRMU) 2019-02-PRMU-CNR, PRMU2018-131 , Tokushima, Japan, February 28th-March 1st (Presentation in 1st), 2019.
  • 千葉直也,橋本浩一.LTM推定三次元計測法による金属物体のばら積みピッキング.第19回 計測自動制御学会システムインテグレーション部門講演会 (SICE SI2018),3D2-03,大阪,12月13-15日(発表15日),2018.(査読なし,ポスター発表)
  • 千葉直也,橋本浩一.高速LTMスパース推定による三次元形状計測とばら積みピッキングロボットへの応用 .ビジョン技術の実利用ワークショップ (ViEW2018),OS4-O1,横浜,12月6-7日(発表7日),2018.(査読なし,口頭発表)
  • 千葉直也,橋本浩一.三次元計測のためのサチュレーションに対応したスパース推定法.第21回 画像の認識・理解シンポジウム (MIRU2018),PS1-30,札幌,8月5-8日(発表6日),2018.(査読なし,ポスター発表)
  • 阿部理也,澤田あずさ,Chenhui Chu,千葉直也,南蒼馬,依田拓也.音声処理の最先端.第21回 画像の認識・理解シンポジウム (MIRU2018),WT-5,札幌,8月5-8日(発表6-7日),2018.(MIRU2018若手プログラム 異分野サーベイ発表)
  • 千葉直也,橋本浩一.多段解像度LT行列推定を用いた三次元計測法.第24回 画像センシングシンポジウム (SSII2018),IS3-04,横浜,6月13-15日(発表15日),2018.(査読なし,ポスター発表)
  • 千葉直也,橋本浩一.サチュレーションとガンマ補正を考慮したプロジェクタ・カメラシステムの光学キャリブレーションツールの開発.ロボティクス・メカトロニクス講演会 2018 (ROBOMECH2018),2A1-M13,北九州,6月2-5日(発表5日),2018.(査読なし,ポスター発表)ソースコード
Naoya Chiba, Koichi Hashimoto. Development of Calibration Tool for Projector-Camera System Considering Saturation And Gamma Correction. The 2017 JSME annual Conference on Robotics and Mechatronics (ROBOMECH2018), 2A1-M13, Kitakyushu, June 2nd-5th (Presentation in 5th), 2018. (Non-refereed, Poster) Source.
  • 小松大和,千葉直也,橋本浩一.三次元構造マッチングを用いたペプチド立体構造探索手法の構築.情報処理学会第80回全国大会,7ZF-02,東京,3月13-15日(発表15日),2018.情報処理学会全国大会講演論文集, Vol. 80, No. 4, pp. 4.933-4.934(査読なし,口頭発表)
  • 千葉直也,橋本浩一.LT 行列の推定に基づく三次元計測法.第18回 計測自動制御学会システムインテグレーション部門講演会 (SICE SI2017),1C2-16,仙台,12月20-22日(発表20日),2017.(査読なし,ポスター発表,第18回計測自動制御学会システムインテグレーション部門講演会 SI2017 優秀講演賞 受賞)
  • 千葉直也,橋本浩一.物理シミュレーションと透視投影カメラモデルを用いたバラ積みシーン点群生成プログラムの開発.ロボティクス・メカトロニクス講演会 2017 (ROBOMECH2017),2A2-O09,福島,5月10-13日(発表12日)2017.(査読なし,ポスター発表)動画ソースコード
Naoya Chiba, Koichi Hashimoto. Development of program for generating pointcloud of bin scene using physical simulation and perspective camera model, The 2017 JSME annual Conference on Robotics and Mechatronics (ROBOMECH2017), 2A2-O09, Fukushima, Japan, May 10th-13th (Presentation in 12th), 2017. (Non-refereed, Poster). Video. Source.
  • 千葉直也,荒井翔悟,橋本浩一.フィードバックプロジェクションを用いた三次元計測.第4回サステナブル/ロボティック・システム・デザイン研究会,栃木,9月7-8日(発表7日),2015.(査読なし,ポスター発表)
  • 荒井翔悟,千葉直也,福地伸晃,橋本浩一.3D Keypoint 対応付けの高精度化に関する検討.ロボティクス・メカトロニクス講演会 2015 (ROBOMECH2015),1A1-H03,京都,5月17-19日(発表18日),2015.(査読なし,ポスター発表)
Shogo Arai, Naoya Chiba, Nobuaki Fukuchi, Koichi Hashimoto. 3D Keypoint Matching with High Accuracy. The 2015 JSME annual Conference on Robotics and Mechatronics (ROBOMECH2015), 1A1-H03, Kyoto, Japan, May 17th-19th (Presentation in 18th), 2015. (Non-refereed, Poster)

Awards

  • The 5th Case Western Reserve University - Tohoku University Joint Workshop Best Poster Presentation Award
  • 2018 IEEE Robotics and Automation Society Japan Joint Chapter Young Award (ICRA2018)
  • Graduate Program in Data Science International Kick-off Symposium Best Poster Presentation Award
  • 第18回計測自動制御学会システムインテグレーション部門講演会 SI2017 優秀講演賞

Others (Domestic)

  • 発表.高速Light Transport Matrixスパース推定を用いた三次元計測法.情報系WINTER FESTA Episode4,東京,12月25-26日(発表26日),2018.
Presentation. 3D Shape Measurement by Using Fast Light Transport Matrix Sparse Estimation. WINTER FESTA Episode4, Tokyo, Japan. December 25th-26th (Presentation in 26th).
  • 講演.LTMスパース推定による三次元計測の紹介.第5回 カジュアルトーク(11月) 全脳アーキテクチャ若手の会東北支部,11月24日,仙台,2018.
  • 講演.LTMによる三次元計測とLTMスパース推定 .第2回 東北大学人工知能学研究会学生の会,11月8日,仙台,2018.
  • サーベイ資料.三次元点群を取り扱うニューラルネットワークのサーベイ.10月22日,2018.SlideShare
  • 発表.汎用的な三次元ロボットビジョン(研究課題:プロジェクタ・カメラ間の輝度伝達行列の推定に基づく三次元計測法).ACT-I「情報と未来」1期生成果発表会,東京,5月19日,2018.VideoPoster
  • 発表.三次元構造マッチングを用いたペプチザイムの大規模探索.H29学生プロジェクト成果発表会,仙台,2月19日,2018.
  • 講演.プロジェクタ・カメラ間のLight Transport行列の推定に基づく三次元計測法.東北大学人工知能学研究センター 第4回 人工知能学研究会,11月13日,仙台,2017.
  • 講演.三次元構造マッチングを用いたペプチザイムの大規模探索の研究紹介.木下・大林研究室有志メンバーによる研究会,仙台,3月31日,2017.
  • 発表.三次元構造マッチングを用いたペプチザイムの大規模探索.H28学生プロジェクト成果発表会,仙台,2月28日,2017.

Contact

  • Address: Hashimoto Lab. Graduate School of Information Sciences, Tohoku University, 6-6-01 Aramaki Aza Aoba, Aoba-ku, Sendai 980-8579, Japan
  • Phone Numbers: +81-22-795-7021
  • E-mail Address: chiba_at_ic.is.tohoku.ac.jp
  • Twitter: @n_chiba_