This paper investigates the use of color attributes in visual tracking, a challenging problem in computer vision. Most existing trackers rely on luminance information or simple color representations, while sophisticated color features have shown superior performance in object recognition and detection. The authors propose an adaptive low-dimensional variant of color attributes, which are computationally efficient and maintain high discriminative power. They extend the CSK tracker, a fast and accurate tracker, to incorporate multi-dimensional color features. The proposed approach improves the baseline intensity-based tracker by 24% in median distance precision and outperforms state-of-the-art tracking methods while running at over 100 frames per second. The evaluation is conducted on 41 challenging benchmark sequences, demonstrating the effectiveness of the proposed method in various challenging scenarios such as illumination variation, occlusion, deformation, and in-plane rotation.This paper investigates the use of color attributes in visual tracking, a challenging problem in computer vision. Most existing trackers rely on luminance information or simple color representations, while sophisticated color features have shown superior performance in object recognition and detection. The authors propose an adaptive low-dimensional variant of color attributes, which are computationally efficient and maintain high discriminative power. They extend the CSK tracker, a fast and accurate tracker, to incorporate multi-dimensional color features. The proposed approach improves the baseline intensity-based tracker by 24% in median distance precision and outperforms state-of-the-art tracking methods while running at over 100 frames per second. The evaluation is conducted on 41 challenging benchmark sequences, demonstrating the effectiveness of the proposed method in various challenging scenarios such as illumination variation, occlusion, deformation, and in-plane rotation.