 This paper proposes a novel approach for multi-object tracking, MOT. It uses a triplet-based image feature to distinguish between similar-looking objects, which reduces the number of identity switches over time. Additionally, an attention-based re-identification model is used to extract the feature vectors from the images, allowing for more accurate association of objects. Experimental results show that the proposed method outperforms other methods on the ID switch metric and improves the detection performance of the tracking system. This article was authored by Wu Jin-an, Kung-Suk Ko, Mutei Glim, and others.