Saved in:
Bibliographic Details
Main Authors: Tze, Christina Ourania, Dauner, Daniel, Liao, Yiyi, Tsishkou, Dzmitry, Geiger, Andreas
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.19117
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910251186388992
author Tze, Christina Ourania
Dauner, Daniel
Liao, Yiyi
Tsishkou, Dzmitry
Geiger, Andreas
author_facet Tze, Christina Ourania
Dauner, Daniel
Liao, Yiyi
Tsishkou, Dzmitry
Geiger, Andreas
contents Existing approaches to 3D semantic urban scene generation predominantly rely on voxel-based representations, which are bound by fixed resolution, challenging to edit, and memory-intensive in their dense form. In contrast, we advocate for a primitive-based paradigm where urban scenes are represented using compact, semantically meaningful 3D elements that are easy to manipulate and compose. To this end, we introduce PrITTI, a latent diffusion model that leverages vectorized object primitives and rasterized ground surfaces for generating diverse, controllable, and editable 3D semantic urban scenes. This hybrid representation yields a structured latent space that facilitates object- and ground-level manipulation. Experiments on KITTI-360 show that primitive-based representations unlock the full capabilities of diffusion transformers, achieving state-of-the-art 3D scene generation quality with lower memory requirements, faster inference, and greater editability than voxel-based methods. Beyond generation, PrITTI supports a range of downstream applications, including scene editing, inpainting, outpainting, and photo-realistic street-view synthesis. The source code and more results can be found at https://raniatze.github.io/pritti/.
format Preprint
id arxiv_https___arxiv_org_abs_2506_19117
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PrITTI: Primitive-based Generation of Controllable and Editable 3D Semantic Urban Scenes
Tze, Christina Ourania
Dauner, Daniel
Liao, Yiyi
Tsishkou, Dzmitry
Geiger, Andreas
Computer Vision and Pattern Recognition
Existing approaches to 3D semantic urban scene generation predominantly rely on voxel-based representations, which are bound by fixed resolution, challenging to edit, and memory-intensive in their dense form. In contrast, we advocate for a primitive-based paradigm where urban scenes are represented using compact, semantically meaningful 3D elements that are easy to manipulate and compose. To this end, we introduce PrITTI, a latent diffusion model that leverages vectorized object primitives and rasterized ground surfaces for generating diverse, controllable, and editable 3D semantic urban scenes. This hybrid representation yields a structured latent space that facilitates object- and ground-level manipulation. Experiments on KITTI-360 show that primitive-based representations unlock the full capabilities of diffusion transformers, achieving state-of-the-art 3D scene generation quality with lower memory requirements, faster inference, and greater editability than voxel-based methods. Beyond generation, PrITTI supports a range of downstream applications, including scene editing, inpainting, outpainting, and photo-realistic street-view synthesis. The source code and more results can be found at https://raniatze.github.io/pritti/.
title PrITTI: Primitive-based Generation of Controllable and Editable 3D Semantic Urban Scenes
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.19117