Structure entropy minimization-based dynamic social interaction modeling for trajectory prediction

• Published in: Information sciences 2022-10, Vol.614, p.170-184
• Main Authors: Jin, Yuhui, Yang, Sixun, Lv, Weifeng, Yu, Haitao, Zhu, Sainan, Huang, Jian
• Format: Article
• Language: English
• Subjects: Intrascene consistency of motion patterns; Multi-time-scale motion tendency; Social interaction; Structure entropy minimization; Trajectory prediction
• ISSN: 0020-0255; 1872-6291
• DOI: 10.1016/j.ins.2022.10.024

•A novel dynamic social interaction modeling mechanism based on structure entropy minimization.•A separate scene node to model the motion patterns and behavioral preferences.•A multi-time-scale motion tendency modeling module.•State-of-the-art results on the ETH, UCY and SDD datasets. Trajectory is an important basis for reflecting the behavior of moving agents and can be used for various applications. Autonomous systems navigating in complex scenes should have the ability to predict the future locations of surrounding agents and avoid collisions. Agents in a scene interact with each other constantly, which is the greatest challenge of trajectory prediction. However, existing methods have problems with insufficient, superfluous or inaccurate interactions, whether they are distance-, attention- or dense graph-based interactive mechanisms. Moreover, the dynamic complexity and validity of social interactions cannot be quantified. In this paper, we propose a dynamic social interaction modeling mechanism based on structure entropy minimization. Specifically, structure entropy minimization provides a principle for detecting and quantifying the natural or true interactions between agents. Additionally, considering that most agents under the same scenario often have similar behavioral preferences, we introduce scene features to model the interscene variance and intrascene consistency of motion patterns. Finally, through multi-time-scale motion tendency modeling, we can simultaneously model short-term and long-term intentions to alleviate accumulated errors for pedestrians avoiding collisions. We evaluate our proposed method on three trajectory datasets, and the experiments and comparisons demonstrate that our method outperforms comparative state-of-the-art methods while demonstrating the capacity to model social interaction and motion patterns.