Social-STAGE: Spatio-Temporal Multi-Modal Future Trajectory Forecast

TL;DR

This paper introduces Social-STAGE, a novel multi-modal trajectory forecasting model that incorporates social interactions, spatio-temporal attention, and new evaluation metrics to improve prediction diversity and confidence assessment.

Contribution

It presents a new graph-based multi-attention model for multi-modal trajectory prediction with a novel ranking and evaluation framework.

Findings

Outperforms state-of-the-art on ETH and UCY datasets.

Effectively models social interactions and multi-modality.

Provides new metrics for diversity and confidence in predictions.

Abstract

This paper considers the problem of multi-modal future trajectory forecast with ranking. Here, multi-modality and ranking refer to the multiple plausible path predictions and the confidence in those predictions, respectively. We propose Social-STAGE, Social interaction-aware Spatio-Temporal multi-Attention Graph convolution network with novel Evaluation for multi-modality. Our main contributions include analysis and formulation of multi-modality with ranking using interaction and multi-attention, and introduction of new metrics to evaluate the diversity and associated confidence of multi-modal predictions. We evaluate our approach on existing public datasets ETH and UCY and show that the proposed algorithm outperforms the state of the arts on these datasets.