SF-City: A Source-Free Domain Adaptation Method for City-scale Point Cloud Semantic Segmentation

SF-City: A Source-Free Domain Adaptation Method for City-scale Point Cloud Semantic Segmentation

IEEE Transactions on Multimedia

10.1109/TMM.2025.3590934

https://doi.org/10.1109/tmm.2025.3590934

Yan Liu, Hongyuan Zhu, Yinjie Lei, Hao Liu, Yun Pei, Yulan Guo

23 July 2025

Liu, Y., Zhu, H., Lei, Y., Liu, H., Pei, Y., & Guo, Y. (2025). SF-City: A Source-Free Domain Adaptation Method for City-scale Point Cloud Semantic Segmentation. IEEE Transactions on Multimedia, 1–15. https://doi.org/10.1109/tmm.2025.3590934

City-scale point cloud semantic segmentation is an im- portant yet challenging task. Despite progress, existing methods rely heavily on point-wise annotations. An alternative solution is to apply the Unsupervised Domain Adaptation (UDA) approach. Recently, the 2D foundation model has achieved significant progress with training with internet-scale images. Therefore, adapting 2D foundation models to 3D City-scale point clouds is an attempting idea. Due to the data protection and storage issue, 2D source domain data is typically unavailable. Thus, we focus on Source-Free Domain Adaptation (SFDA) and propose a Source-Free City-scale point cloud semantic segmentation method, namely SF-City. Our method leverages knowledge from 2D pre-trained models to generate point-wise pseudo labels for training a 3D semantic segmentation network. We convert point clouds into remote-sensing-like images using Bird’s-Eye-View (BEV) projection. However, directly using source models for pseudo label generation is hindered by domain gaps such as viewpoint variations, concept divergences, and geometry loss. To tackle these problems, we propose a Multi-scale Content Feature Extractor (MCFE) to extract holistic and contextual feature rep- resentations. Then, an Uncertainty-guided Inter-Model Feature Integrator (UIFI) is introduced to integrate inherent knowledge across source models. Furthermore, the Geometric-guided Pseudo Label Generator (GPLG) is leveraged to introduce geometric information to regulate pseudo labels. Through extensive experi- ments on two public benchmarks, SF-City demonstrates superior performance, achieving an mIoU of 28.8% on the SensatUrban dataset, outperforming recent state-of-the-art methods CLIP- FO3D by about 6.3%

Publisher Copyright

This research / project is supported by the Economic Development Board - Space Technology Development Grant Programme
Grant Reference no. : S22-19016- STDP

© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

https://oar.a-star.edu.sg/communities-collections/articles/22332

1520-9210
1941-0077

Institute for Infocomm Research