Segmentation of Subspaces in Sequential Data
This work addresses subspace segmentation for sequential data, which is incremental as it builds on existing methods like Sparse Subspace Clustering.
The paper tackles the problem of segmenting data from sequentially ordered subspaces by proposing Ordered Subspace Clustering (OSC), which adds a penalty term to Sparse Subspace Clustering to handle sequential data, and shows that OSC outperforms state-of-the-art methods like Spatial Subspace Clustering, Low-Rank Representation, and SSC on infrared hyperspectral, video, and motion capture data.
We propose Ordered Subspace Clustering (OSC) to segment data drawn from a sequentially ordered union of subspaces. Similar to Sparse Subspace Clustering (SSC) we formulate the problem as one of finding a sparse representation but include an additional penalty term to take care of sequential data. We test our method on data drawn from infrared hyper spectral, video and motion capture data. Experiments show that our method, OSC, outperforms the state of the art methods: Spatial Subspace Clustering (SpatSC), Low-Rank Representation (LRR) and SSC.