Reproduce “ParkDiffusion++: Ego Intention Conditioned Joint Multi-Agent Trajectory Prediction for Automated Parking using Diffusion Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiarong Wei, Anna Rehr, Christian Feist, Abhinav Valada
ParkDiffusion++ proposes six candidate ego destinations and predicts joint future trajectories for the surrounding scene conditioned on each destination, using beam-based assembly and learned selection. On validation data, its top-scene f-FDE is 0.66 m on DLP and 1.18 m on inD, with lower f-OR than listed baselines, but f-ADE is higher and counterfactuals remain unobserved.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.