SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Dang, Chenxu, Wang, Jie, Li, Guang, Hou, Zhiwen, You, Zihan, Ye, Hangjun, Ma, Jie, Chen, Long, Wang, Yan
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914259918651392
author Dang, Chenxu
Wang, Jie
Li, Guang
Hou, Zhiwen
You, Zihan
Ye, Hangjun
Ma, Jie
Chen, Long
Wang, Yan
author_facet Dang, Chenxu
Wang, Jie
Li, Guang
Hou, Zhiwen
You, Zihan
Ye, Hangjun
Ma, Jie
Chen, Long
Wang, Yan
contents In autonomous driving, Vision Language Models (VLMs) excel at high-level reasoning , whereas semantic occupancy provides fine-grained details. Despite significant progress in individual fields, there is still no method that can effectively integrate both paradigms. Conventional VLMs struggle with token explosion and limited spatiotemporal reasoning, while semantic occupancy provides a unified, explicit spatial representation but is too dense to integrate efficiently with VLMs. To address these challenges and bridge the gap between VLMs and occupancy, we propose SparseOccVLA, a novel vision-language-action model that unifies scene understanding, occupancy forecasting, and trajectory planning powered by sparse occupancy queries. Starting with a lightweight Sparse Occupancy Encoder, SparseOccVLA generates compact yet highly informative sparse occupancy queries that serve as the single bridge between vision and language. These queries are aligned into the language space and reasoned by the LLM for unified scene understanding and future occupancy forecasting. Furthermore, we introduce an LLM-guided Anchor-Diffusion Planner featuring decoupled anchor scoring and denoising, as well as cross-model trajectory-condition fusion. SparseOccVLA achieves a 7% relative improvement in CIDEr over the state-of-the-art on OmniDrive-nuScenes, a 0.5 increase in mIoU score on Occ3D-nuScenes, and sets state-of-the-art open-loop planning metric on nuScenes benchmark, demonstrating its strong holistic capability.
format Preprint
id arxiv_https___arxiv_org_abs_2601_06474
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning
Dang, Chenxu
Wang, Jie
Li, Guang
Hou, Zhiwen
You, Zihan
Ye, Hangjun
Ma, Jie
Chen, Long
Wang, Yan
Computer Vision and Pattern Recognition
Artificial Intelligence
In autonomous driving, Vision Language Models (VLMs) excel at high-level reasoning , whereas semantic occupancy provides fine-grained details. Despite significant progress in individual fields, there is still no method that can effectively integrate both paradigms. Conventional VLMs struggle with token explosion and limited spatiotemporal reasoning, while semantic occupancy provides a unified, explicit spatial representation but is too dense to integrate efficiently with VLMs. To address these challenges and bridge the gap between VLMs and occupancy, we propose SparseOccVLA, a novel vision-language-action model that unifies scene understanding, occupancy forecasting, and trajectory planning powered by sparse occupancy queries. Starting with a lightweight Sparse Occupancy Encoder, SparseOccVLA generates compact yet highly informative sparse occupancy queries that serve as the single bridge between vision and language. These queries are aligned into the language space and reasoned by the LLM for unified scene understanding and future occupancy forecasting. Furthermore, we introduce an LLM-guided Anchor-Diffusion Planner featuring decoupled anchor scoring and denoising, as well as cross-model trajectory-condition fusion. SparseOccVLA achieves a 7% relative improvement in CIDEr over the state-of-the-art on OmniDrive-nuScenes, a 0.5 increase in mIoU score on Occ3D-nuScenes, and sets state-of-the-art open-loop planning metric on nuScenes benchmark, demonstrating its strong holistic capability.
title SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2601.06474