DeepKalPose: An enhanced deep-learning Kalman filter for temporally consistent monocular vehicle pose estimation

This paper presents DeepKalPose, a novel approach for enhancing temporal consistency in monocular vehicle pose estimation applied on video through a deep-learning-based Kalman filter. By integrating a bi-directional Kalman filter strategy utilizing forward and backward time-series processing, combined with a learnable motion model to represent complex motion patterns, the method significantly improves pose accuracy and robustness across various conditions, particularly for occluded or distant vehicles. Experimental validation on the KITTI dataset confirms that DeepKalPose outperforms existing methods in both pose accuracy and temporal consistency. In this paper, we propose a deep-learning-based Kalman filter to enhance temporal consistency in monocular vehicle pose estimation applied on video. A learnable motion model is integrated into a novel bi-directional time-series processing module, significantly improving pose accuracy and temporal coherency. The method demonstrates robustness across various conditions, particularly for occluded or distant vehicles, providing the potential for real-world applications. image