Deep Convolutional Likelihood Particle Filter for Visual Tracking

TL;DR

This paper introduces a convolutional likelihood particle filter that improves visual tracking by using correlation response maps for more reliable likelihood estimation, handling multiple modes, and adapting to challenging scenarios like occlusion and motion blur.

Contribution

The novel particle filter leverages correlation response maps for likelihood estimation and efficiently handles multiple modes, enhancing tracking accuracy and robustness.

Findings

Outperforms state-of-the-art methods on OTB100 benchmark.

Effectively handles occlusion by searching for multiple modes.

Adapts to motion blur by increasing likelihood variance.

Abstract

We propose a novel particle filter for convolutional-correlation visual trackers. Our method uses correlation response maps to estimate likelihood distributions and employs these likelihoods as proposal densities to sample particles. Likelihood distributions are more reliable than proposal densities based on target transition distributions because correlation response maps provide additional information regarding the target's location. Additionally, our particle filter searches for multiple modes in the likelihood distribution, which improves performance in target occlusion scenarios while decreasing computational costs by more efficiently sampling particles. In other challenging scenarios such as those involving motion blur, where only one mode is present but a larger search area may be necessary, our particle filter allows for the variance of the likelihood distribution to increase. We tested our algorithm on the Visual Tracker Benchmark v1.1 (OTB100) and our experimental results demonstrate that our framework outperforms state-of-the-art methods.