Point-cloud skeleton conditions and a Reset-and-Roll inference scheme enable stable frame-wise autoregressive driving video generation for closed-loop autonomous driving simulation.
Vista: A generalizable driving world model with high fidelity and versatile controllability
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
DriveCtrl is a depth-conditioned controllable framework that generates realistic driving videos from simulation while preserving annotations and scene dynamics.
citing papers explorer
-
Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation
Point-cloud skeleton conditions and a Reset-and-Roll inference scheme enable stable frame-wise autoregressive driving video generation for closed-loop autonomous driving simulation.
-
DriveCtrl: Conditioned Sim-to-Real Driving Video Generation
DriveCtrl is a depth-conditioned controllable framework that generates realistic driving videos from simulation while preserving annotations and scene dynamics.