The paper introduces Staple, a real-time tracking algorithm that combines the strengths of template and color-based models to achieve robust performance in challenging scenarios. Staple leverages ridge regression to learn a model that is both fast and accurate, outperforming state-of-the-art trackers in multiple benchmarks while maintaining a frame rate of over 80 FPS. The method addresses the limitations of correlation filters, which are sensitive to deformation, and color-based models, which struggle with illumination changes. Staple's approach involves solving two independent ridge-regression problems for template and histogram scores, respectively, and combining their outputs in a dense translation search. This ensures that the model remains robust to both color changes and deformations, making it suitable for applications requiring high computational efficiency and robustness to environmental variations.The paper introduces Staple, a real-time tracking algorithm that combines the strengths of template and color-based models to achieve robust performance in challenging scenarios. Staple leverages ridge regression to learn a model that is both fast and accurate, outperforming state-of-the-art trackers in multiple benchmarks while maintaining a frame rate of over 80 FPS. The method addresses the limitations of correlation filters, which are sensitive to deformation, and color-based models, which struggle with illumination changes. Staple's approach involves solving two independent ridge-regression problems for template and histogram scores, respectively, and combining their outputs in a dense translation search. This ensures that the model remains robust to both color changes and deformations, making it suitable for applications requiring high computational efficiency and robustness to environmental variations.