Salvato in:
Dettagli Bibliografici
Autori principali: Carrión, Héctor, Bai, Yutong, Castro, Víctor A. Hernández, Panaganti, Kishan, Zenith, Ayush, Trang, Matthew, Zhang, Tony, Perona, Pietro, Malik, Jitendra
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2506.11302
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916802071625728
author Carrión, Héctor
Bai, Yutong
Castro, Víctor A. Hernández
Panaganti, Kishan
Zenith, Ayush
Trang, Matthew
Zhang, Tony
Perona, Pietro
Malik, Jitendra
author_facet Carrión, Héctor
Bai, Yutong
Castro, Víctor A. Hernández
Panaganti, Kishan
Zenith, Ayush
Trang, Matthew
Zhang, Tony
Perona, Pietro
Malik, Jitendra
contents World models aim to simulate environments and enable effective agent behavior. However, modeling real-world environments presents unique challenges as they dynamically change across both space and, crucially, time. To capture these composed dynamics, we introduce a Spatio-Temporal Road Image Dataset for Exploration (STRIDE) permuting 360-degree panoramic imagery into rich interconnected observation, state and action nodes. Leveraging this structure, we can simultaneously model the relationship between egocentric views, positional coordinates, and movement commands across both space and time. We benchmark this dataset via TARDIS, a transformer-based generative world model that integrates spatial and temporal dynamics through a unified autoregressive framework trained on STRIDE. We demonstrate robust performance across a range of agentic tasks such as controllable photorealistic image synthesis, instruction following, autonomous self-control, and state-of-the-art georeferencing. These results suggest a promising direction towards sophisticated generalist agents--capable of understanding and manipulating the spatial and temporal aspects of their material environments--with enhanced embodied reasoning capabilities. Training code, datasets, and model checkpoints are made available at https://huggingface.co/datasets/Tera-AI/STRIDE.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11302
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TARDIS STRIDE: A Spatio-Temporal Road Image Dataset and World Model for Autonomy
Carrión, Héctor
Bai, Yutong
Castro, Víctor A. Hernández
Panaganti, Kishan
Zenith, Ayush
Trang, Matthew
Zhang, Tony
Perona, Pietro
Malik, Jitendra
Computer Vision and Pattern Recognition
Artificial Intelligence
World models aim to simulate environments and enable effective agent behavior. However, modeling real-world environments presents unique challenges as they dynamically change across both space and, crucially, time. To capture these composed dynamics, we introduce a Spatio-Temporal Road Image Dataset for Exploration (STRIDE) permuting 360-degree panoramic imagery into rich interconnected observation, state and action nodes. Leveraging this structure, we can simultaneously model the relationship between egocentric views, positional coordinates, and movement commands across both space and time. We benchmark this dataset via TARDIS, a transformer-based generative world model that integrates spatial and temporal dynamics through a unified autoregressive framework trained on STRIDE. We demonstrate robust performance across a range of agentic tasks such as controllable photorealistic image synthesis, instruction following, autonomous self-control, and state-of-the-art georeferencing. These results suggest a promising direction towards sophisticated generalist agents--capable of understanding and manipulating the spatial and temporal aspects of their material environments--with enhanced embodied reasoning capabilities. Training code, datasets, and model checkpoints are made available at https://huggingface.co/datasets/Tera-AI/STRIDE.
title TARDIS STRIDE: A Spatio-Temporal Road Image Dataset and World Model for Autonomy
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2506.11302