Annealed Winner-Takes-All for Motion Forecasting

TL;DR

This paper introduces an annealed Winner-Takes-All loss (aWTA) for motion forecasting in autonomous driving, improving diversity and accuracy of predictions with fewer hypotheses and no post-selection, enhancing training stability and convergence.

Contribution

We integrate aWTA loss into existing motion forecasting models, demonstrating improved performance and stability with minimal hypotheses, reducing the need for post-selection.

Findings

Enhanced prediction diversity and accuracy.

Reduced number of hypotheses needed for effective forecasting.

Elimination of post-selection step during inference.

Abstract

In autonomous driving, motion prediction aims at forecasting the future trajectories of nearby agents, helping the ego vehicle to anticipate behaviors and drive safely. A key challenge is generating a diverse set of future predictions, commonly addressed using data-driven models with Multiple Choice Learning (MCL) architectures and Winner-Takes-All (WTA) training objectives. However, these methods face initialization sensitivity and training instabilities. Additionally, to compensate for limited performance, some approaches rely on training with a large set of hypotheses, requiring a post-selection step during inference to significantly reduce the number of predictions. To tackle these issues, we take inspiration from annealed MCL, a recently introduced technique that improves the convergence properties of MCL methods through an annealed Winner-Takes-All loss (aWTA). In this paper, we demonstrate how the aWTA loss can be integrated with state-of-the-art motion forecasting models to enhance their performance using only a minimal set of hypotheses, eliminating the need for the cumbersome post-selection step. Our approach can be easily incorporated into any trajectory prediction model normally trained using WTA and yields significant improvements. To facilitate the application of our approach to future motion forecasting models, the code is made publicly available: https://github.com/valeoai/MF_aWTA.