Background Subtraction using Adaptive Singular Value Decomposition
This work addresses background subtraction for sensor data processing, presenting an incremental improvement with computational efficiency.
The paper tackled background subtraction in sensor data by developing an adaptive singular value decomposition method that updates singular vectors efficiently and supports block-wise updates, achieving state-of-the-art performance in qualitative and quantitative evaluations.
An important task when processing sensor data is to distinguish relevant from irrelevant data. This paper describes a method for an iterative singular value decomposition that maintains a model of the background via singular vectors spanning a subspace of the image space, thus providing a way to determine the amount of new information contained in an incoming frame. We update the singular vectors spanning the background space in a computationally efficient manner and provide the ability to perform block-wise updates, leading to a fast and robust adaptive SVD computation. The effects of those two properties and the success of the overall method to perform a state of the art background subtraction are shown in both qualitative and quantitative evaluations.