The paper introduces the MOT16 benchmark, a new release of the MOTChallenge benchmark for multiple object tracking. The benchmark aims to provide a standardized and fair evaluation framework for multi-target tracking methods. Unlike the initial MOT15 release, MOT16 includes 14 sequences with more crowded scenarios, different viewpoints, camera motions, and weather conditions. The annotations for all sequences are carefully conducted by qualified researchers following a strict protocol, ensuring high accuracy. The benchmark covers multiple object classes, including pedestrians, vehicles, sitting people, and occluding objects, providing detailed information on the level of occlusion and cropping for each object. The paper details the annotation rules, data format, and evaluation metrics used in the benchmark. It also presents the quantitative results of several baseline methods and discusses the strengths and weaknesses of state-of-the-art multi-target tracking techniques. The goal is to advance the field of multi-target tracking by providing a rich ground truth dataset and a fair comparison platform.The paper introduces the MOT16 benchmark, a new release of the MOTChallenge benchmark for multiple object tracking. The benchmark aims to provide a standardized and fair evaluation framework for multi-target tracking methods. Unlike the initial MOT15 release, MOT16 includes 14 sequences with more crowded scenarios, different viewpoints, camera motions, and weather conditions. The annotations for all sequences are carefully conducted by qualified researchers following a strict protocol, ensuring high accuracy. The benchmark covers multiple object classes, including pedestrians, vehicles, sitting people, and occluding objects, providing detailed information on the level of occlusion and cropping for each object. The paper details the annotation rules, data format, and evaluation metrics used in the benchmark. It also presents the quantitative results of several baseline methods and discusses the strengths and weaknesses of state-of-the-art multi-target tracking techniques. The goal is to advance the field of multi-target tracking by providing a rich ground truth dataset and a fair comparison platform.