Utilizing an Intention-aware Denoising Diffusion Model (IDM) can improve trajectory prediction accuracy in autonomous driving systems by modeling uncertainty and reducing inference time.
Di-Long proposes a knowledge distillation framework for long-term trajectory prediction, achieving state-of-the-art performance.
A progressive interaction network enhances agent feature representation by incorporating map information at different stages, improving trajectory prediction in autonomous driving.
提案されたプログレッシブインタラクションネットワークは、自律走行の軌道予測においてエージェントの特徴表現をより良く学習し、シーンコンテキスト情報を捉えることができます。
SingularTrajectory proposes a diffusion-based universal trajectory prediction framework to reduce the performance gap across various trajectory prediction tasks by unifying human dynamics representations.
A global-to-local generation approach that mitigates the accumulated error, introduces spatial-temporal constraints among future steps, and selects the optimal granularity for each trajectory to enhance prediction accuracy and kinematic feasibility.