<p>Object tracking has been an important research topic in computer vision serving a variety of applications. Existing methods often struggle with challenges such as occlusion, background clutter, and long-term object appearance variation. In this paper, we propose a novel rank-based context-aware multi-regularized correlation filter (CAMR) tracker to address these issues. Our model integrates rank-based contextual information with multiple regularization techniques, which enables effective group feature selection across both spatial and channel dimensions. The main innovation lies in the introduction of a selective spatial regularization technique that mitigates boundary effects while retaining key target information, and a multi-feature integration method that adaptively combines deep feature weights with handcrafted features to enhance target representation. Through extensive experiments on several publicly available datasets, we demonstrate that our method outperforms existing state-of-the-art techniques in terms of both accuracy and robustness. Specifically, in the GOT-10k dataset, our method improves the success rate by 30% over the baseline tracker. Our analysis further shows that each component of the proposed approach contributes significantly to its improved performance.</p>