Advancements in self-driving car technology have the potential to revolutionize transportation by enhancing safety, efficiency, and accessibility. Nonetheless, the successful integration of autonomous vehicles into ou...
详细信息
Advancements in self-driving car technology have the potential to revolutionize transportation by enhancing safety, efficiency, and accessibility. Nonetheless, the successful integration of autonomous vehicles into our urban landscapes necessitates robust and reliable pedestrian detection and tracking systems. As we frequently observe pedestrians moving together, tracking them as a group becomes a beneficial approach, mitigating occlusion and enhancing both the accuracy and speed of object detection and tracking. However, utilizing a human-view camera in an autonomous vehicle presents challenges as pedestrians occupy varied fields of view. In some instances, pedestrians closer to the camera may overlap with those farther away, as seen from the camera's viewpoint, which causes the mis-groupings to happen. To address these challenges, we proposed a strategy to divide the image into distinct zones and perform grouping within each, significantly minimizing mis-groupings. First, an object detection method extracted pedestrians and their bounding boxes from an image. Second, zonedetection was applied to separate the image into several zones. Third, clustering methods were applied to detect pedestrian groups within each zone. Last, the object tracking method was utilized to track pedestrian groups. We repeated the process over a ten-frame sequence to achieve better performance, with object detection executed in the first frame and object tracking in the remaining nine frames. The comparison of processing times of different groupdetection methods indicated that tracking pedestrian groups is more time-efficient than tracking individuals and achieved a 4.5% to 14.1% improvement. Furthermore, according to the Adjusted Rand Index (ARI) evaluation metric, our proposed zone-based group detection method outperforms the other commonly used approaches by achieving scores of 0.635 on the MOT17 dataset and 0.781 on the KITTI dataset. In addition, the proposed approach surpasses the
Pedestrian trajectory prediction is a critical aspect of computer vision, aimed at predicting a pedestrian's future locations by analyzing their past movements. Traditional trajectory prediction models primarily f...
详细信息
Pedestrian trajectory prediction is a critical aspect of computer vision, aimed at predicting a pedestrian's future locations by analyzing their past movements. Traditional trajectory prediction models primarily focus on individuals, which can be challenging in densely populated areas due to occlusions. These occlusions not only complicate the re-identification of pedestrians once they reappear but also increase processing times due to the more complex procedures and the greater number of objects involved. However, a common observation is that pedestrians often travel in groups. This insight led us to propose a novel approach, predicting the future trajectories of pedestrian groups instead of individuals. This strategy effectively addresses the complexities of predicting movements in crowded environments and the issues related to pedestrian occlusion. In this work, we introduced a distinctive methodology for identifying pedestrian groups, re-identifying them, and predicting their future trajectories. Our approach, unlike traditional state-of-the-art re-identification and trajectory prediction methods of individual pedestrians, focuses on re-identifying pedestrian groups and predicting their future trajectories while emphasizing processing time reduction with great accuracy. The process started with object detection to ascertain pedestrian coordinates. Subsequently, a zone-based clustering method was employed to form groups. Following this, a specific group re-identification was utilized to construct continuous trajectories for these groups, rather than for individual pedestrians. Finally, the group trajectory prediction technique was applied to estimate the future movements of these groups. Both the object detection and groupdetection methods were applied every five frames to generate these trajectories. The effectiveness of our approach has been validated using several evaluation metrics, including Average Displacement Error (ADE), Final Displacement Error (FDE),
暂无评论