# Overlap-Aware Feature Learning for Robust Unsupervised Domain Adaptation for 3D Semantic Segmentation

arXiv: https://arxiv.org/abs/2504.01668 · HTML: https://mlsanity.com/paper/2504.01668/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 37 references that are in the map.
- First posted on arXiv 2 Apr 2025.

## Builds on (most cited first, 24 of 24)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distances](https://mlsanity.com/paper/1306.0895/) (2013)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [Contrastive Adaptation Network for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1901.00976/) (2019)
- [SemanticPOSS: A Point Cloud Dataset with Large Quantity of Dynamic Instances](https://mlsanity.com/paper/2002.09147/) (2020)
- [Improving Black-box Adversarial Attacks with a Transfer-based Prior](https://mlsanity.com/paper/1906.06919/) (2019)
- [Complete & Label: A Domain Adaptation Approach to Semantic Segmentation of LiDAR Point Clouds](https://mlsanity.com/paper/2007.08488/) (2020)
- [Extending Adversarial Attacks and Defenses to Deep 3D Point Cloud Classifiers](https://mlsanity.com/paper/1901.03006/) (2019)
- [3D Semantic Segmentation in the Wild: Learning Generalized Models for Adverse-Condition Point Clouds](https://mlsanity.com/paper/2304.00690/) (2023)
- [Statistical, Robustness, and Computational Guarantees for Sliced Wasserstein Distances](https://mlsanity.com/paper/2210.09160/) (2022)
- [Energy-Based Sliced Wasserstein Distance](https://mlsanity.com/paper/2304.13586/) (2023)
- [Imperceptible Transfer Attack and Defense on 3D Point Cloud Classification](https://mlsanity.com/paper/2111.10990/) (2021)
- [On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Driving](https://mlsanity.com/paper/2201.01850/) (2022)
- [Adversarial Robustness for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2109.00946/) (2021)
- [Exploring Robustness of Unsupervised Domain Adaptation in Semantic Segmentation](https://mlsanity.com/paper/2105.10843/) (2021)
- [Compositional Semantic Mix for Domain Adaptation in Point Cloud Segmentation](https://mlsanity.com/paper/2308.14619/) (2023)
- [Unsupervised Domain Adaptation for LiDAR Panoptic Segmentation](https://mlsanity.com/paper/2109.15286/) (2021)
- [DIMAL: Deep Isometric Manifold Learning Using Sparse Geodesic Sampling](https://mlsanity.com/paper/1711.06011/) (2017)
- [Adversarial Feature Desensitization](https://mlsanity.com/paper/2006.04621/) (2020)
- [Exploring Adversarially Robust Training for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2202.09300/) (2022)
- [Unsupervised Domain Adaptation for Point Cloud Semantic Segmentation via Graph Matching](https://mlsanity.com/paper/2208.04510/) (2022)
- [An Unsupervised Domain Adaptive Approach for Multimodal 2D Object Detection in Adverse Weather Conditions](https://mlsanity.com/paper/2203.03568/) (2022)
- [T-UDA: Temporal Unsupervised Domain Adaptation in Sequential Point Clouds](https://mlsanity.com/paper/2309.08302/) (2023)
- [On Adversarial Robustness of Point Cloud Semantic Segmentation](https://mlsanity.com/paper/2112.05871/) (2021)

## Abstract (arXiv)

3D point cloud semantic segmentation (PCSS) is a cornerstone for environmental perception in robotic systems and autonomous driving, enabling precise scene understanding through point-wise classification. While unsupervised domain adaptation (UDA) mitigates label scarcity in PCSS, existing methods critically overlook the inherent vulnerability to real-world perturbations (e.g., snow, fog, rain) and adversarial distortions. This work first identifies two intrinsic limitations that undermine current PCSS-UDA robustness: (a) unsupervised features overlap from unaligned boundaries in shared-class regions and (b) feature structure erosion caused by domain-invariant learning that suppresses target-specific patterns. To address the proposed problems, we propose a tripartite framework consisting of: 1) a robustness evaluation model quantifying resilience against adversarial attack/corruption types through robustness metrics; 2) an invertible attention alignment module (IAAM) enabling bidirectional domain mapping while preserving discriminative structure via attention-guided overlap suppression; and 3) a contrastive memory bank with quality-aware contrastive learning that progressively refines pseudo-labels with feature quality for more discriminative representations. Extensive experiments on SynLiDAR-to-SemanticPOSS adaptation demonstrate a maximum mIoU improvement of 14.3% under adversarial attack.