Saved in:
Bibliographic Details
Main Authors: Gao, Lili, Xu, Yanbo, Koch, William, Ruffino, Samuele, Rowe, Luke, Chalaki, Behdad, Rivkin, Dmitriy, Ost, Julian, Girgis, Roger, Bijelic, Mario, Heide, Felix
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.17147
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911604568752128
author Gao, Lili
Xu, Yanbo
Koch, William
Ruffino, Samuele
Rowe, Luke
Chalaki, Behdad
Rivkin, Dmitriy
Ost, Julian
Girgis, Roger
Bijelic, Mario
Heide, Felix
author_facet Gao, Lili
Xu, Yanbo
Koch, William
Ruffino, Samuele
Rowe, Luke
Chalaki, Behdad
Rivkin, Dmitriy
Ost, Julian
Girgis, Roger
Bijelic, Mario
Heide, Felix
contents We introduce ScenarioControl, the first vision-language control mechanism for learned driving scenario generation. Given a text prompt or an input image, Scenario-Control synthesizes diverse, realistic 3D scenario rollouts - including map, 3D boxes of reactive actors over time, pedestrians, driving infrastructure, and ego camera observations. The method generates scenes in a vectorized latent space that represents road structure and dynamic agents jointly. To connect multimodal control with sparse vectorized scene elements, we propose a cross-global control mechanism that integrates crossattention with a lightweight global-context branch, enabling fine-grained control over road layout and traffic conditions while preserving realism. The method produces temporally consistent scenario rollouts from the perspectives different actors in the scene, supporting long-horizon continuation of driving scenarios. To facilitate training and evaluation, we release a dataset with text annotations aligned to vectorized map structures. Extensive experiments validate that the control adherence and fidelity of ScenarioControl compare favorable to all tested methods across all experiments. Project webpage: https://light.princeton.edu/ScenarioControl
format Preprint
id arxiv_https___arxiv_org_abs_2604_17147
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ScenarioControl: Vision-Language Controllable Vectorized Latent Scenario Generation
Gao, Lili
Xu, Yanbo
Koch, William
Ruffino, Samuele
Rowe, Luke
Chalaki, Behdad
Rivkin, Dmitriy
Ost, Julian
Girgis, Roger
Bijelic, Mario
Heide, Felix
Computer Vision and Pattern Recognition
Robotics
We introduce ScenarioControl, the first vision-language control mechanism for learned driving scenario generation. Given a text prompt or an input image, Scenario-Control synthesizes diverse, realistic 3D scenario rollouts - including map, 3D boxes of reactive actors over time, pedestrians, driving infrastructure, and ego camera observations. The method generates scenes in a vectorized latent space that represents road structure and dynamic agents jointly. To connect multimodal control with sparse vectorized scene elements, we propose a cross-global control mechanism that integrates crossattention with a lightweight global-context branch, enabling fine-grained control over road layout and traffic conditions while preserving realism. The method produces temporally consistent scenario rollouts from the perspectives different actors in the scene, supporting long-horizon continuation of driving scenarios. To facilitate training and evaluation, we release a dataset with text annotations aligned to vectorized map structures. Extensive experiments validate that the control adherence and fidelity of ScenarioControl compare favorable to all tested methods across all experiments. Project webpage: https://light.princeton.edu/ScenarioControl
title ScenarioControl: Vision-Language Controllable Vectorized Latent Scenario Generation
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2604.17147