Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories With GANs

By Javad Amirian, Jean-Bernard Hayet, Julien Pettré. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2019

324 citations recorded by the source corpus.

This paper proposes a novel approach for predicting the motion of pedestrians interacting with others. It uses a Generative Adversarial Network (GAN) to sample plausible predictions for any agent in the scene. As GANs are very susceptible to mode collapsing and dropping, we show that the recently proposed Info-GAN allows dramatic improvements in multi-modal pedestrian trajectory prediction to avoid these issues. We also left out L2-loss in training the generator, unlike some previous works, because it causes serious mode collapsing though faster convergence. We show through experiments on real and synthetic data that the proposed method leads to generate more diverse samples and to preserve the modes of the predictive distribution. In particular, to prove this claim, we have designed a toy example dataset of trajectories that can be used to assess the performance of different methods in preserving the predictive distribution modes.

Subfields: Artificial Intelligence, Automotive Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/6328723f58c8f296f9551c5be4910f0cf36330e1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Automotive Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Generative Adversarial Networks–Enabled Human–Artificial Intelligence Collaborative Applications for Creative and Design Industries: A Systematic Review of Current Approaches and Trends (2021, methodology): “Taking crowd simulation as an example, we see several GAN-based solutions for modeling behavior (Gupta et al., 2018; Amirian et al., 2019), but these models only take current agent states into account.”
Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks (2019, methodology): “Furthermore, our model uses graph attention networks [22] to more efﬁciently and robustly model the interactions between the agents in the scene, whereas prior research [12, 31] depend on hand-deﬁned rules.”
Human motion trajectory prediction: a survey (2019, background): “2018; Amirian et al. 2019; Kosaraju et al. 2019). For instance, Gupta et al. (2018) extend the Social-LSTM by using Generative Adversarial Networks and a novel variety loss which encourages the generative network to produce diverse multi-modal predictions.”