Tracking Dynamic Gaussian Density with a Theoretically Optimal Sliding Window Approach
This work addresses dynamic density estimation for applications like computer vision and signal processing, offering a theoretically optimal solution that is incremental over existing heuristic methods.
The authors tackled the problem of dynamic density estimation by deriving the exact mean integrated squared error for sliding window Gaussian kernel density estimators applied to evolving Gaussian densities, and they provided a principled method to choose the optimal weight sequence, showing empirical improvements over heuristic approaches.
Dynamic density estimation is ubiquitous in many applications, including computer vision and signal processing. One popular method to tackle this problem is the "sliding window" kernel density estimator. There exist various implementations of this method that use heuristically defined weight sequences for the observed data. The weight sequence, however, is a key aspect of the estimator affecting the tracking performance significantly. In this work, we study the exact mean integrated squared error (MISE) of "sliding window" Gaussian Kernel Density Estimators for evolving Gaussian densities. We provide a principled guide for choosing the optimal weight sequence by theoretically characterizing the exact MISE, which can be formulated as constrained quadratic programming. We present empirical evidence with synthetic datasets to show that our weighting scheme indeed improves the tracking performance compared to heuristic approaches.