Presentation 1999/3/18
Articulated Object Extraction from Incomplete Image
Kazuma Ishikawa, Tsuyoshi Yamamura, Noboru Ohnishi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Object extraction from an image is indispensable in image recognition. Various methods such as template matching and Geometric Hashing have been studied in object extraction. However, problems arise when the image has much noise or occlusion. Moreover such methods are not effective for articulated object extraction. In these cases, it is most effctive to perform extraction by a top-down processing using a model. We created a model which is invariant to aspect, scale, pose change, false edges by noise, and occlusion. We propose an articulated object recognition system by top-down processing using this model.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) incomplete image / top-down processing / model-based matching / articulated object extraction
Paper # PRMU98-252
Date of Issue

Conference Information
Committee PRMU
Conference Date 1999/3/18(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Articulated Object Extraction from Incomplete Image
Sub Title (in English)
Keyword(1) incomplete image
Keyword(2) top-down processing
Keyword(3) model-based matching
Keyword(4) articulated object extraction
1st Author's Name Kazuma Ishikawa
1st Author's Affiliation Graduate School of Engineering, Nagoya University()
2nd Author's Name Tsuyoshi Yamamura
2nd Author's Affiliation Faculty of Information Science and Technology, Aichi Prefectual University
3rd Author's Name Noboru Ohnishi
3rd Author's Affiliation RIKEN BMC Research Center:Graduate School of Engineering, Nagoya University
Date 1999/3/18
Paper # PRMU98-252
Volume (vol) vol.98
Number (no) 677
Page pp.pp.-
#Pages 8
Date of Issue