Presentation | 2006/3/15 Interpretation of Human Actions for Each Body Part Based on Object Information and Similitude Juanda LOKMAN, Junichi Imai, Masahide KANEKO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper describes the interpretation method of human actions in daily lives, which are rich of varieties in motion and very ambiguous. We present an approach for extracting human object from the scene using distance cue, face detection and skin color. We use only one stereo camera without any information about its location, except the optical axis is parallel to the ground. Human actions can be very ambiguous and moreover some actions involve only one body part while the others could involve more than one body part. Thus object as a context and the relationship among objects that share some level of similitude (e.g., dimensionality, texture/outline, material) can be used as a cue for interpreting ambiguous human actions in daily lives. We will show some examples of the advantage of object context and relationship among objects for the recognition of human actions. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Interpretation of Human Motion / Action Recognition / Multiple Object Segmentation and Tracking |
Paper # | HCS2005-72 |
Date of Issue |
Conference Information | |
Committee | HCS |
---|---|
Conference Date | 2006/3/15(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 | Human Communication Science (HCS) |
---|---|
Language | ENG |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Interpretation of Human Actions for Each Body Part Based on Object Information and Similitude |
Sub Title (in English) | |
Keyword(1) | Interpretation of Human Motion |
Keyword(2) | Action Recognition |
Keyword(3) | Multiple Object Segmentation and Tracking |
1st Author's Name | Juanda LOKMAN |
1st Author's Affiliation | The University of Electro-Communications() |
2nd Author's Name | Junichi Imai |
2nd Author's Affiliation | The University of Electro-Communications |
3rd Author's Name | Masahide KANEKO |
3rd Author's Affiliation | The University of Electro-Communications |
Date | 2006/3/15 |
Paper # | HCS2005-72 |
Volume (vol) | vol.105 |
Number (no) | 681 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |