A new trajectory based motion segmentation benchmark dataset (UdG-MS15)

Muhammad Habib Mahmood, Luca Zappella, Yago Díez, Joaquim Salvi, Xavier Lladó

研究成果: Conference contribution

2 被引用数 (Scopus)

抄録

Motion segmentation (MS) is an essential step in video analysis. Its quantitative and qualitative evaluation is largely dependent on the dataset used for testing. Although there are publicly available datasets such as Hopkins and FBMS, they have limitations in terms of number of motions, partial/complete occlusion, stopping motion, sequence length, and real life natural sequences. Due to these limitations, many recent proposals have reached nearly zero misclassification, especially for Hopkins, which leaves no room for quantitatively differentiating among proposals. In this paper, we present a new challenging trajectory based MS dataset of 15 sequences, where number of motions and sequence length have been largely increased as compared to the state of the art. An effort has been made to include all forms of distortions that are present in real life scenes. As a starting point, a preliminary benchmark evaluation using a recent and well known state of the art algorithm has been provided for this dataset.

本文言語English
ホスト出版物のタイトルPattern Recognition and Image Analysis - 7th Iberian Conference, IbPRIA 2015, Proceedings
編集者Jaime S. Cardoso, Roberto Paredes, Xosé M. Pardo
出版社Springer Verlag
ページ463-470
ページ数8
ISBN(電子版)9783319193892
DOI
出版ステータスPublished - 2015
イベント7th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2015 - Santiago de Compostela, Spain
継続期間: 2015 6 172015 6 19

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
9117
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Other

Other7th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2015
国/地域Spain
CitySantiago de Compostela
Period15/6/1715/6/19

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「A new trajectory based motion segmentation benchmark dataset (UdG-MS15)」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル