ILNet: Trajectory Prediction with Inverse Learning Attention for Enhancing Intention Capture

TL;DR

ILNet introduces an inverse learning attention mechanism and dynamic anchor selection to improve multi-agent trajectory prediction by better capturing complex interactions and intentions, achieving state-of-the-art results on benchmark datasets.

Contribution

The paper proposes ILNet, a novel trajectory prediction framework that models interactions through inverse learning attention and dynamic anchor selection, enhancing intention capture and adaptability.

Findings

Achieves state-of-the-art performance on INTERACTION and Argoverse datasets.

Performs better in challenging interaction scenarios with fewer parameters.

Produces more accurate and multimodal trajectory distributions.

Abstract

Trajectory prediction for multi-agent interaction scenarios is a crucial challenge. Most advanced methods model agent interactions by efficiently factorized attention based on the temporal and agent axes. However, this static and foward modeling lacks explicit interactive spatio-temporal coordination, capturing only obvious and immediate behavioral intentions. Alternatively, the modern trajectory prediction framework refines the successive predictions by a fixed-anchor selection strategy, which is difficult to adapt in different future environments. It is acknowledged that human drivers dynamically adjust initial driving decisions based on further assumptions about the intentions of surrounding vehicles. Motivated by human driving behaviors, this paper proposes ILNet, a multi-agent trajectory prediction method with Inverse Learning (IL) attention and Dynamic Anchor Selection (DAS) module. IL Attention employs an inverse learning paradigm to model interactions at neighboring moments, introducing proposed intentions to dynamically encode the spatio-temporal coordination of interactions, thereby enhancing the model's ability to capture complex interaction patterns. Then, the learnable DAS module is proposed to extract multiple trajectory change keypoints as anchors in parallel with almost no increase in parameters. Experimental results show that the ILNet achieves state-of-the-art performance on the INTERACTION and Argoverse motion forecasting datasets. Particularly, in challenged interaction scenarios, ILNet achieves higher accuracy and more multimodal distributions of trajectories over fewer parameters. Our codes are available at https://github.com/mjZeng11/ILNet.