Abstract for presentation (Poster or Podium) with a Paper in the Conference Proceedings
Data Sensing and Analytics
Yufei Ji (he/him/his)
Undergraduate
School of Transportation, Southeast University
Nanjing, Jiangsu, China (People's Republic)
Pei Liu, n/a
Master Student
School of Transportation, Southeast University
Nanjing, Jiangsu, China (People's Republic)
Liu Siwen
student
Southeast University
nan jing shi, Jiangsu, China (People's Republic)
Nanfang Zheng, n/a
undergraduate
School of Transportation, Southeast University
Nanjing, Jiangsu, China (People's Republic)
Ziyuan Pu, n/a
professor
School of Transportation, Southeast University
Nanjing, Jiangsu, China (People's Republic)
Yufei Ji
School of Transportation, Southeast University
Nanjing, Jiangsu, China (People's Republic)
Accurate pedestrian trajectory prediction is crucial for enhancing pedestrian and autonomous vehicle safety. In comparison to the bird's eye view, ego-centric trajectory prediction is more challenging because the vehicle's motion can dynamically alter the scale of pedestrian bounding boxes. As the distance between the vehicle and pedestrians decreases, pedestrian bounding boxes are magnified, subsequently influencing the predicted pedestrian trajectories within the image plane. To address this challenge, the paper proposes a deep learning network based on LSTM encoder-decoder architecture that leverages trajectory, ego-vehicle motion information and local-visual context as the additional input information. Specifically, the motion parameters of the in-vehicle camera are introduced to represent self-vehicle motion information at a coarse granularity level, reducing the impact of noise when representing self-vehicle speed by optical flow. Meanwhile, due to the disproportionate impact of pedestrians at different distances on the optimization process, the loss function is normalized using the actual width and height of the pedestrian's bounding box, effectively reducing the weight assigned to pedestrians in close proximity during optimization. The attention mechanism is utilized to better capture the long-term temporal variations. Several typical pedestrian trajectory prediction benchmark datasets are used for the algorithm validation, including PIE, JAAD, and PSI. According to the results, the proposed method outperformed the slected state-of-the-art algorithms with about 14% improvements in trajectory prediction accuracy.