DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States

TL;DR

DeMo introduces a novel framework that decouples motion forecasting into directional intentions and dynamic states, improving multi-modal trajectory prediction for autonomous driving.

Contribution

It proposes a new decoupling approach and combined Attention and Mamba techniques for better trajectory representation and prediction accuracy.

Findings

Achieves state-of-the-art results on Argoverse 2 and nuScenes benchmarks.

Effectively models dynamic agent states and directional intentions separately.

Enhances multi-modal trajectory forecasting performance.

Abstract

Accurate motion forecasting for traffic agents is crucial for ensuring the safety and efficiency of autonomous driving systems in dynamically changing environments. Mainstream methods adopt a one-query-one-trajectory paradigm, where each query corresponds to a unique trajectory for predicting multi-modal trajectories. While straightforward and effective, the absence of detailed representation of future trajectories may yield suboptimal outcomes, given that the agent states dynamically evolve over time. To address this problem, we introduce DeMo, a framework that decouples multi-modal trajectory queries into two types: mode queries capturing distinct directional intentions and state queries tracking the agent's dynamic states over time. By leveraging this format, we separately optimize the multi-modality and dynamic evolutionary properties of trajectories. Subsequently, the mode and state queries are integrated to obtain a comprehensive and detailed representation of the trajectories. To achieve these operations, we additionally introduce combined Attention and Mamba techniques for global information aggregation and state sequence modeling, leveraging their respective strengths. Extensive experiments on both the Argoverse 2 and nuScenes benchmarks demonstrate that our DeMo achieves state-of-the-art performance in motion forecasting.