Multi-Object Tracking (MOT) is crucial for real-world applications like video surveillance, where it aims to detect and maintain consistent identifiers for objects across video frames. However, MOT methods often struggle with objects that are heavily overlapped due to occlusion or exhibit diverse poses due to non-linear motion. In this paper, we propose a robust tracking-by-detection method named DIMATrack. It incorporates the Kalman Filter for precise trajectory prediction, and a novel Dimension Aware Intersection-over-Union (DIMA-IoU) metric for enhanced data association. DIMA-IoU improves upon standard IoU by integrating both height-aware and width-aware measurements, improving association accuracy in complex scenarios and during occlusions. By integrating these components, DIMATrack effectively leverages weak cues that are often overlooked by conventional methods, which rely on appearance or spatial information. Extensive experiments on three benchmarks demonstrate the superior performance of our DIMATrack, particularly in challenging tracking environments. The code is available at https://github.com/Melikamuliyih/DIMATrack .