Spectral Filter Tracking

TL;DR

Spectral Filter Tracking (SFT) introduces a robust, localized graph-based approach for visual object tracking by leveraging spectral graph filters and polynomial parameterization to improve accuracy and resilience to background clutter.

Contribution

The paper presents a novel spectral graph filter method for tracking that operates on localized regions and avoids eigenvalue decomposition, enhancing robustness and efficiency.

Findings

Effective in handling local variations and cluttered backgrounds.

Operates efficiently without eigenvalue decomposition.

Improves tracking robustness over holistic correlation filter methods.

Abstract

Visual object tracking is a challenging computer vision task with numerous real-world applications. Here we propose a simple but efficient Spectral Filter Tracking (SFT)method. To characterize rotational and translation invariance of tracking targets, the candidate image region is models as a pixelwise grid graph. Instead of the conventional graph matching, we convert the tracking into a plain least square regression problem to estimate the best center coordinate of the target. But different from the holistic regression of correlation filter based methods, SFT can operate on localized surrounding regions of each pixel (i.e.,vertex) by using spectral graph filters, which thus is more robust to resist local variations and cluttered background.To bypass the eigenvalue decomposition problem of the graph Laplacian matrix L, we parameterize spectral graph filters as the polynomial of L by spectral graph theory, in which L k exactly encodes a k-hop local neighborhood of each vertex. Finally, the filter parameters (i.e., polynomial coefficients) as well as feature projecting functions are jointly integrated into the regression model.