English

SocialVAE: Human Trajectory Prediction using Timewise Latents

Computer Vision and Pattern Recognition 2022-11-01 v4 Machine Learning

Abstract

Predicting pedestrian movement is critical for human behavior analysis and also for safe and efficient human-agent interactions. However, despite significant advancements, it is still challenging for existing approaches to capture the uncertainty and multimodality of human navigation decision making. In this paper, we propose SocialVAE, a novel approach for human trajectory prediction. The core of SocialVAE is a timewise variational autoencoder architecture that exploits stochastic recurrent neural networks to perform prediction, combined with a social attention mechanism and a backward posterior approximation to allow for better extraction of pedestrian navigation strategies. We show that SocialVAE improves current state-of-the-art performance on several pedestrian trajectory prediction benchmarks, including the ETH/UCY benchmark, Stanford Drone Dataset, and SportVU NBA movement dataset. Code is available at: https://github.com/xupei0610/SocialVAE.

Keywords

Cite

@article{arxiv.2203.08207,
  title  = {SocialVAE: Human Trajectory Prediction using Timewise Latents},
  author = {Pei Xu and Jean-Bernard Hayet and Ioannis Karamouzas},
  journal= {arXiv preprint arXiv:2203.08207},
  year   = {2022}
}

Comments

In the 17th European Conference on Computer Vision (ECCV 2022). Code: https://github.com/xupei0610/SocialVAE

R2 v1 2026-06-24T10:14:47.430Z