Study of extracting multiple independently moving objects and estimating the egomotion from active stereo sequences (ヒューマン情報処理)

スポンサーリンク

概要

• 論文の詳細を見る

• In this paper, we propose a modified RANSAC based method that can detect multiple independently moving objects from the video sequences acquired by active stereo cameras, and can estimate the egomotion of the stereo cameras. From the optical flows that computed by matching the feature descriptors within temporal frames, a modified RANSAC is adopted to extract the valid egomotion models, under which enough homogenous optical flows can be found. Each of the egomotion models corresponds to either the background or one of the independent moving individuals. The optical flows of the background are then utilized to calculate the real egomotion. Experiments using real stereo sequences demonstrate the effectiveness of the proposed method.

• 社団法人電子情報通信学会の論文
• 2010-03-08

著者

• Ohya Jun

  Student Of Gits Waseda University

• Xie Yingdi

  Student Of Gits Waseda University

関連論文

• Study of extracting multiple independently moving objects and estimating the egomotion from active stereo sequences (ヒューマン情報処理)
• A modified RANSAC based method for extracting primitive shapes from image (パターン認識・メディア理解)
• Multiple shapes extraction with guidance to random sampling (コンピュータビジョンとイメージメディア)
• Independent motion areas extraction from a modified RANSAC mechanism
• A modified RANSAC based method for extracting primitive shapes from image (ヒューマン情報処理)

スポンサーリンク