# Global Optimality Guarantees for Nonconvex Unsupervised Video Segmentation

## TL;DR

This paper establishes conditions under which nonconvex optimization methods can reliably achieve globally optimal unsupervised video object segmentation, offering theoretical guarantees for practical algorithms.

## Contribution

It introduces a nonnegative robust PCA formulation for video segmentation and derives conditions ensuring global optimality with local search methods.

## Key findings

- Global optimality guarantees under specific data conditions
- Nonconvex formulation is computationally more tractable than convex relaxations
- Validated optimality criteria on real video data

## Abstract

In this paper, we consider the problem of unsupervised video object segmentation via background subtraction. Specifically, we pose the nonsemantic extraction of a video's moving objects as a nonconvex optimization problem via a sum of sparse and low-rank matrices. The resulting formulation, a nonnegative variant of robust principal component analysis, is more computationally tractable than its commonly employed convex relaxation, although not generally solvable to global optimality. In spite of this limitation, we derive intuitive and interpretable conditions on the video data under which the uniqueness and global optimality of the object segmentation are guaranteed using local search methods. We illustrate these novel optimality criteria through example segmentations using real video data.