Jul 2026
End-to-end Conditional Diffusion for Realistic and Controllable Visual Traffic Scenario Generation
Conditioned on front-view visual observations, E2E-CDiff jointly denoises future motion states and executable low-level controls for route-interacting background vehicles, and mitigates the planning-control mismatch in conventional two-stage trajectory-then-controller pipelines.
Jingzheng Li, Yu-Fei Ge, Zhijun Chen et al.
· arXiv.org · 0 citations