Skip to main content

Spatio-Temporal Video Object Segmentation via Scale-Adaptive 3D Structure Tensor


To address multiple motions and deformable objects' motions encountered in existing region-based approaches, an automatic video object (VO) segmentation methodology is proposed in this paper by exploiting the duality of image segmentation and motion estimation such that spatial and temporal information could assist each other to jointly yield much improved segmentation results. The key novelties of our method are (1) scale-adaptive tensor computation, (2) spatial-constrained motion mask generation without invoking dense motion-field computation, (3) rigidity analysis, (4) motion mask generation and selection, and (5) motion-constrained spatial region merging. Experimental results demonstrate that these novelties jointly contribute much more accurate VO segmentation both in spatial and temporal domains.

Author information



Corresponding author

Correspondence to Hai-Yun Wang.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Wang, HY., Ma, KK. Spatio-Temporal Video Object Segmentation via Scale-Adaptive 3D Structure Tensor. EURASIP J. Adv. Signal Process. 2004, 672709 (2004).

Download citation

Keywords and phrases

  • video object segmentation
  • 3D structure tensor
  • rigidity analysis