<technical report>
Model Fitting of Articulated Objects

Creator
Language
Publisher
Date
Source Title
Vol
Issue
First Page
Last Page
Publication Type
Access Rights
Rights
Related DOI
Related URI
Relation
Abstract 本稿では人体や手などに代表される多関節物体の三次元姿勢を画像から推定する モデルフィッティングの技術についてサーベイする. 画像によるモデルフィッティングの枠組みを, 1)推定に利用される画像特徴, 2)画像と照合するモデルの表現と照合のパラメー タ空間, 3)照合時の評価関数と最適解の探索手法,にわけて多関節物体の三次元 姿勢推定に特徴的な要素を上記三つの観点から比較整理することを試みる.
In ...this paper, we present a survey report for the model fitting method to estimate 3-D posture of articulated objects such as human body and hand. We decompose the model fitting framework into the following three elements: 1) image feature, 2) model description and parameter space for model-image matching and 3) matching function and its optimization. From the viewpoint of these three issues, we try to compare the various methods of model fitting to each other and summarize them.show more

Hide fulltext details.

pdf ShimadaCVIM06 pdf 431 KB 767  

Details

Record ID
Peer-Reviewed
Subject Terms
Notes
Type
Created Date 2009.04.22
Modified Date 2020.11.02

People who viewed this item also viewed