Vehicle re-identification (Re-ID) distinguishes between the same vehicle and other vehicles in images. It is challenging due to significant intra-instance differences between identical vehicles from different views and subtle inter-instance differences of similar vehicles. Researchers have tried to address this problem by extracting features robust to variations of viewpoints and environments. More recently, they tried to improve performance by using additional metadata such as key points, orientation, and temporal information. Although these attempts have been relatively successful, they all require expensive annotations. Therefore, this paper proposes a novel deep neural network called a multi-attention-based soft partition (MUSP) network to solve this problem. This network does not use metadata and only uses multiple soft attentions to identify a specific vehicle area. This function was performed by metadata in previous studies. Experiments verified that MUSP achieved state-of-the-art (SOTA) performance for the VehicleID dataset without any additional annotations and was comparable to VeRi-776 and VERI-Wild.