The paper presents an experimental survey of visual tracking methods, evaluating 19 trackers on 315 video fragments to assess their performance under various conditions such as illumination changes, occlusion, and camera motion. The study aims to provide an objective evaluation of tracking algorithms by using statistical tests, survival curves, and Grubbs testing. It finds that the F-score is as effective as the object tracking accuracy (OTA) score in evaluating tracking performance. The survey includes a diverse set of trackers, ranging from older methods (1999-2006) to recent ones from major conferences. The evaluation is conducted on a real-life dataset, ALOV++, which includes a wide variety of dynamic situations. The study highlights the strengths and weaknesses of different tracking methods, showing that the top-performing trackers do not share a common underlying method. The paper also discusses various tracking approaches, including matching-based, appearance-based, constraint-based, and discriminative classification methods. It evaluates different performance metrics, such as the F-score, OTA, and PBM, to assess the accuracy and robustness of tracking algorithms. The study concludes that the performance of trackers can be objectively evaluated using statistical tests, and that the results provide valuable insights into the strengths and weaknesses of different tracking methods.The paper presents an experimental survey of visual tracking methods, evaluating 19 trackers on 315 video fragments to assess their performance under various conditions such as illumination changes, occlusion, and camera motion. The study aims to provide an objective evaluation of tracking algorithms by using statistical tests, survival curves, and Grubbs testing. It finds that the F-score is as effective as the object tracking accuracy (OTA) score in evaluating tracking performance. The survey includes a diverse set of trackers, ranging from older methods (1999-2006) to recent ones from major conferences. The evaluation is conducted on a real-life dataset, ALOV++, which includes a wide variety of dynamic situations. The study highlights the strengths and weaknesses of different tracking methods, showing that the top-performing trackers do not share a common underlying method. The paper also discusses various tracking approaches, including matching-based, appearance-based, constraint-based, and discriminative classification methods. It evaluates different performance metrics, such as the F-score, OTA, and PBM, to assess the accuracy and robustness of tracking algorithms. The study concludes that the performance of trackers can be objectively evaluated using statistical tests, and that the results provide valuable insights into the strengths and weaknesses of different tracking methods.