# LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding

DevFeed: [LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding](<https://devfeed.tech/articles/lsd-3d-large-scale-3d-driving-scene-generation-with-geometry-grounding-119979.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2508.19204>)

Author: Julian Ost, Andrea Ramazzina, Amogh Joshi, Maximilian B\\"omer, Mario Bijelic, Felix Heide

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [3D](<https://devfeed.tech/topics/3d.md>), [3D rendering](<https://devfeed.tech/topics/3d-rendering.md>), [Machine Perception](<https://devfeed.tech/topics/machine-perception.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [allows](<https://devfeed.tech/tags/allows.md>), [approach](<https://devfeed.tech/tags/approach.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [bridge](<https://devfeed.tech/tags/bridge.md>), [can](<https://devfeed.tech/tags/can.md>), [causality](<https://devfeed.tech/tags/causality.md>), [combines](<https://devfeed.tech/tags/combines.md>), [complex](<https://devfeed.tech/tags/complex.md>), [control](<https://devfeed.tech/tags/control.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-gr](<https://devfeed.tech/tags/cs-gr.md>), [data](<https://devfeed.tech/tags/data.md>), [driving](<https://devfeed.tech/tags/driving.md>), [generation](<https://devfeed.tech/tags/generation.md>), [grounding](<https://devfeed.tech/tags/grounding.md>), [large](<https://devfeed.tech/tags/large.md>), [scale](<https://devfeed.tech/tags/scale.md>), [scene-generation](<https://devfeed.tech/tags/scene-generation.md>), [with](<https://devfeed.tech/tags/with.md>)

## AI overview

LSD-3D generates controllable, geometry-grounded 3D driving scenes to address limits of both reconstruction from captured data and image or video diffusion. It combines a proxy geometry and environment representation with score distillation from learned 2D image priors. According to the abstract, scenes can be prompt-guided and conditioned on map layouts, with explicit 3D geometry estimation and causal novel-view synthesis that preserves object permanence. The authors report realistic, geometrically consistent scenes with high-fidelity texture and structure; the excerpt does not provide quantitative evaluation details.

## Source excerpt

arXiv:2508.19204v2 Announce Type: replace-cross Abstract: Large-scale scene data is essential for training and testing in robot learning. Neural reconstruction methods have promised the capability of reconstructing large physically-grounded outdoor scenes from captured sensor data. However, these methods have baked-in static environments and only allow for limited scene control -- they are functionally constrained in scene and trajectory diversity by the captures from which they are reconstructed. In contrast, generating driving data with recent image or video diffusion models offers control, however, at the cost of geometry grounding and causality. In this work, we aim to bridge this gap and present a method that directly generates large-scale 3D driving scenes with accurate geometry, allowing for causal novel view synthesis with object permanence and explicit 3D geometry estimation. The proposed method combines the generation of a proxy geometry and environment representation with score distillation from learned 2D image priors. We find that this approach allows for high controllability, enabling the prompt-guided geometry and high-fidelity texture and structure that can be conditioned on map layouts -- producing realistic and geometrically consistent 3D generations of complex driving scenes.