Motion Segmentation by SCC on the Hopkins 155 Database

TL;DR

This paper demonstrates that the Spectral Curvature Clustering (SCC) algorithm achieves superior motion segmentation accuracy on the Hopkins 155 database, outperforming existing methods with low misclassification rates.

Contribution

The paper applies SCC to a benchmark database and shows it outperforms all other state-of-the-art motion segmentation methods.

Findings

SCC achieves 1.41% misclassification on two-motion sequences.

SCC achieves 4.85% misclassification on three-motion sequences.

SCC outperforms all other methods on the Hopkins 155 database.

Abstract

We apply the Spectral Curvature Clustering (SCC) algorithm to a benchmark database of 155 motion sequences, and show that it outperforms all other state-of-the-art methods. The average misclassification rate by SCC is 1.41% for sequences having two motions and 4.85% for three motions.