Unsupervised Spatio-Temporal Segmentation with Sparse Spectral Clustering - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Unsupervised Spatio-Temporal Segmentation with Sparse Spectral Clustering

Abstract

Spatio-temporal cues are powerful sources of information for segmentation in videos. In this work we present an efficient and simple technique for spatio-temporal segmentation that is based on a low-rank spectral clustering algorithm. The complexity of graph based spatio-temporal segmentation is dominated by the size of the graph, which is proportional to the number of pixels in a video sequence. In contrast to other works, we avoid oversegmenting the images into super-pixels and instead generalize a simple graph based image segmentation. Our graph construction encodes appearance and motion information with temporal links based on optical flow. For large scale data sets naïve graph construction is computationally and memory intensive, and has only been achieved previously using a high power compute cluster. We make feasible for the first time large scale graph-based spatio-temporal segmentation on a single core by exploiting the sparsity structure of the problem and a low rank factorization that has strong approximation guarantees. We empirically demonstrate that constructing the low rank approximation using a subset of pixels (30%-50%) achieves performance exceeding the state-of-the-art on the Hopkins 155 dataset, while enabling the graph to fit in core memory.
Fichier principal
Vignette du fichier
unsupervised-spatio-temporal.pdf (2.75 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01034903 , version 1 (22-07-2014)

Identifiers

  • HAL Id : hal-01034903 , version 1

Cite

Mahsa Ghafarianzadeh, Matthew Blaschko, Gabe Sibley. Unsupervised Spatio-Temporal Segmentation with Sparse Spectral Clustering. British Machine Vision Conference (BMVC), Sep 2014, Nottingham, United Kingdom. ⟨hal-01034903⟩
377 View
1071 Download

Share

Gmail Facebook Twitter LinkedIn More