CymbaDiff: Structured Spatial Diffusion for Sketch-based 3D Semantic Urban Scene Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liang, Li, Miao, Bo, Wang, Xinyu, Akhtar, Naveed, Vice, Jordan, Mian, Ajmal
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909993914073088
author Liang, Li
Miao, Bo
Wang, Xinyu
Akhtar, Naveed
Vice, Jordan
Mian, Ajmal
author_facet Liang, Li
Miao, Bo
Wang, Xinyu
Akhtar, Naveed
Vice, Jordan
Mian, Ajmal
contents Outdoor 3D semantic scene generation produces realistic and semantically rich environments for applications such as urban simulation and autonomous driving. However, advances in this direction are constrained by the absence of publicly available, well-annotated datasets. We introduce SketchSem3D, the first large-scale benchmark for generating 3D outdoor semantic scenes from abstract freehand sketches and pseudo-labeled annotations of satellite images. SketchSem3D includes two subsets, Sketch-based SemanticKITTI and Sketch-based KITTI-360 (containing LiDAR voxels along with their corresponding sketches and annotated satellite images), to enable standardized, rigorous, and diverse evaluations. We also propose Cylinder Mamba Diffusion (CymbaDiff) that significantly enhances spatial coherence in outdoor 3D scene generation. CymbaDiff imposes structured spatial ordering, explicitly captures cylindrical continuity and vertical hierarchy, and preserves both physical neighborhood relationships and global context within the generated scenes. Extensive experiments on SketchSem3D demonstrate that CymbaDiff achieves superior semantic consistency, spatial realism, and cross-dataset generalization. The code and dataset will be available at https://github.com/Lillian-research-hub/CymbaDiff
format Preprint
id arxiv_https___arxiv_org_abs_2510_13245
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CymbaDiff: Structured Spatial Diffusion for Sketch-based 3D Semantic Urban Scene Generation
Liang, Li
Miao, Bo
Wang, Xinyu
Akhtar, Naveed
Vice, Jordan
Mian, Ajmal
Computer Vision and Pattern Recognition
Outdoor 3D semantic scene generation produces realistic and semantically rich environments for applications such as urban simulation and autonomous driving. However, advances in this direction are constrained by the absence of publicly available, well-annotated datasets. We introduce SketchSem3D, the first large-scale benchmark for generating 3D outdoor semantic scenes from abstract freehand sketches and pseudo-labeled annotations of satellite images. SketchSem3D includes two subsets, Sketch-based SemanticKITTI and Sketch-based KITTI-360 (containing LiDAR voxels along with their corresponding sketches and annotated satellite images), to enable standardized, rigorous, and diverse evaluations. We also propose Cylinder Mamba Diffusion (CymbaDiff) that significantly enhances spatial coherence in outdoor 3D scene generation. CymbaDiff imposes structured spatial ordering, explicitly captures cylindrical continuity and vertical hierarchy, and preserves both physical neighborhood relationships and global context within the generated scenes. Extensive experiments on SketchSem3D demonstrate that CymbaDiff achieves superior semantic consistency, spatial realism, and cross-dataset generalization. The code and dataset will be available at https://github.com/Lillian-research-hub/CymbaDiff
title CymbaDiff: Structured Spatial Diffusion for Sketch-based 3D Semantic Urban Scene Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.13245