Eigencontours: Novel Contour Descriptors Based on Low-Rank Approximation
This work addresses contour representation for instance segmentation, offering an incremental improvement over existing descriptors.
The paper tackled the problem of representing object boundaries in instance segmentation by proposing eigencontours based on low-rank approximation, resulting in more effective and efficient representation in low-dimensional space and meaningful performance on datasets.
Novel contour descriptors, called eigencontours, based on low-rank approximation are proposed in this paper. First, we construct a contour matrix containing all object boundaries in a training set. Second, we decompose the contour matrix into eigencontours via the best rank-M approximation. Third, we represent an object boundary by a linear combination of the M eigencontours. We also incorporate the eigencontours into an instance segmentation framework. Experimental results demonstrate that the proposed eigencontours can represent object boundaries more effectively and more efficiently than existing descriptors in a low-dimensional space. Furthermore, the proposed algorithm yields meaningful performances on instance segmentation datasets.