Archives

1．Technology

• Development Technology in Hashimoto Laboratory
• Amazon Picking Challenge: Important technologies
• Vector Pair Matching
Movie :An Object picking by using robot arms
References：
S.Akizuki and M.Hashimoto，Position and Pose Recognition of Randomly Stacked Objects using Highly Observable 3D Vector Pairs，Proceedings of the 40th Annual Conference of the IEEE Industrial Electronics Society（IECON），pp.5266-5271，Dallas/U.S.A.，2014


• Stable Position and Pose Estimation of Industrial Parts using Evaluation of Observability of 3D Vector Pairs
Movie :Simulate Observability of Object Surface
References：
S.Akizuki and M.Hashimoto, Stable Position and Pose Estimation of Industrial Parts using Evaluation of Observability of 3D Vector Pairs, Journal of Robotics and Mechatronics, Vol.27, No.2, pp.174-181, 2015/4.


• Global Reference Frame for position and orientation recognition of planar objects
Movie :3D tracking of planar objects
References：
S.Akizuki and M.Hashimoto, A Proposal of the Global Reference Frame for Surface Flatness-independent 3D Object Detection，Proceedings of Joint Conference of the International Workshop on Advanced Image Technology（IWAIT） and the International Forum on Medical Imaging in Asia（IFMIA），OS. 27，Tainan/Taiwan，2015．


• 2D Object Detection
• Co-occurrence Probability-based Template Matching(CPTM)
• Distinguishing Pixel Template Matching(DPTM)
• Stability Co-occurrence Probability Template Matching(SCPTM)
• Multi Flash Camera Template Matching(MFCTM)
• 3D Object Detection
• Vector Pair Matching(VPM)
• Combination of Curvatures and Difference of Normals(CCDoN)
• 3D Object Recognition using Effective Features Selected by Evaluating Performance of Discrimination
• Related Technology
• 2D Object Detection

Template Matching	ISC	Chamfer Matching

• 3D Object Detection

Point Pair Feature(PPF)

2．Experimental Database

• 2-D data
• LSI chip
• T-shaped metal plate
• Piece of puzzle
• A piece of Shogi
• 3-D data
• Data Captured by the Range Sensor
• Plug
• Gear
• Wooden Stick
• Joint
• Original
• Other
• Mono-6D dataset
Reference: Konishi, Y., Hanzawa, Y., Kawade, M., Hashimoto, M.: Fast 6D Pose Estimation from a Monocular Image Using Hierarchical Pose Trees，in ECCV, pp.398-413(2016)


• Experimental Image of Textureless Object Recognition
References：Yoshinori Konishi, Yuki Hanzawa, Masato Kawade, Manabu Hashimoto, Fast 6D Pose Estimation from a Monocular Image Using Hierarchical Pose Trees, 14th European Conference, Proceedings Part I, Volume 9905 of the series Lecture Notes in Computer Science pp 398-413, Amsterdam, Netherlands, Oct.9, 2016.

3．Presentation Document / List of Literature

• Lecture on the Nagoya CV-PRML Seminar（2015/3/21）
　Slides: "3D Features for Object Recognition" 　（5.5MB）

• Lecture on the DIA2015 workshop（2015/3/5）
　Paper "3D Features for point cloud data processing" 　（1.9MB）

• Special Talk on ViEW2014 （2014/12/4）
　Slides: "Fundamentals of 3D features for object recognition" 　（8.1MB）

• Invited Talk on the IAIP（2014/5/16）
　Slides: "Evolution of object recognition: Pixel-based and Feature-based"　（0.9MB）

• Tutorial Seminar of SSII2013（2013/6/12）
　Slides: Attractive research on Template Matching　（29.8MB）

• Robotics Seminar（2013/5/16）
　Seminar text: "High-speed object detection"　（3.7MB）
　Reference lists:　 Robust　 　High-speed　 　3D object recognition based on Key-points

4．Resume of Graduation thesis and Master Thesis

• Resume of Graduation thesis
• Resume of Master Thesis

▲Topへ

---