Long-range trajectories from global and local motion representations
详细信息 查看全文
- 作者：Eduardo M. Pereira ; ejmp@inesctec.pt" class="auth_mail" title="E-mail the corresponding author ; Jaime S. Cardoso jaime.cardoso@inesctec.pt" class="auth_mail" title="E-mail the corresponding author ; Ricardo Morla ricardo.morla@fe.up.pt" class="auth_mail" title="E-mail the corresponding author
- 关键词：Long trajectories ; Motion representations ; Flow information
- 刊名：Journal of Visual Communication and Image Representation
- 出版年：2016
- 出版时间：October 2016
- 年：2016
- 卷：40
- 期：part_PA
- 页码：265-287
- 全文大小：6997 K
文摘
- •
A system that extracts long-range trajectories from different scenarios is proposed.
- •
A rich fine-to-coarse representation of flow vectors is obtained.
- •
A new data-driven flow vector outlier removal technique is presented and justified.
- •
A global re-correlation algorithm at the tracklet-level is formulated.
- •
Our system largely outperforms state-of-the-art methods for motion segmentation.