PanDA: Unsupervised Domain Adaptation for Multimodal 3D Panoptic Segmentation in Autonomous Driving

• URL: http://arxiv.org/abs/2604.19379v1
• Date: Tue, 21 Apr 2026 12:01:29 GMT
• Title: PanDA: Unsupervised Domain Adaptation for Multimodal 3D Panoptic Segmentation in Autonomous Driving
• Abstract summary: UDA aims to improve generalization under domain shifts commonly encountered in real-world autonomous driving.<n>Existing supervised mm-3DPS methods rely heavily on strong cross-modal complementarity between LiDAR and RGB inputs.<n>We propose PanDA, the first UDA framework specifically designed for multimodal 3D panoptic segmentation.
• Score: 25.00498014106047
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper presents the first study on Unsupervised Domain Adaptation (UDA) for multimodal 3D panoptic segmentation (mm-3DPS), aiming to improve generalization under domain shifts commonly encountered in real-world autonomous driving. A straightforward solution is to employ a pseudo-labeling strategy, which is widely used in UDA to generate supervision for unlabeled target data, combined with an mm-3DPS backbone. However, existing supervised mm-3DPS methods rely heavily on strong cross-modal complementarity between LiDAR and RGB inputs, making them fragile under domain shifts where one modality degrades (e.g., poor lighting or adverse weather). Moreover, conventional pseudo-labeling typically retains only high-confidence regions, leading to fragmented masks and incomplete object supervision, which are issues particularly detrimental to panoptic segmentation. To address these challenges, we propose PanDA, the first UDA framework specifically designed for multimodal 3D panoptic segmentation. To improve robustness against single-sensor degradation, we introduce an asymmetric multimodal augmentation that selectively drops regions to simulate domain shifts and improve robust representation learning. To enhance pseudo-label completeness and reliability, we further develop a dual-expert pseudo-label refinement module that extracts domain-invariant priors from both 2D and 3D modalities. Extensive experiments across diverse domain shifts, spanning time, weather, location, and sensor variations, significantly surpass state-of-the-art UDA baselines for 3D semantic segmentation.

Related papers

• CCF: Complementary Collaborative Fusion for Domain Generalized Multi-Modal 3D Object Detection [10.131066023908023]
  Multi-modal fusion has emerged as a promising paradigm for accurate 3D object detection.<n>However, performance degrades substantially when deployed in target domains different from training.<n>In this work, we identify two factors that limit robust cross-domain generalization.<n>We propose three components to address these challenges.
  arXiv  Detail & Related papers  (2026-03-24T14:39:51Z)
• BEVUDA++: Geometric-aware Unsupervised Domain Adaptation for Multi-View 3D Object Detection [56.477525075806966]
  Vision-centric Bird's Eye View (BEV) perception holds considerable promise for autonomous driving.<n>Recent studies have prioritized efficiency or accuracy enhancements, yet the issue of domain shift has been overlooked.<n>We introduce an innovative geometric-aware teacher-student framework, BEVUDA++, to diminish this issue.
  arXiv  Detail & Related papers  (2025-09-17T16:31:40Z)
• Unsupervised Domain Adaptation for 3D LiDAR Semantic Segmentation Using Contrastive Learning and Multi-Model Pseudo Labeling [0.7373617024876725]
  Unsupervised contrastive learning at the segment level is used to pre-train a backbone network.<n>A multi-model pseudo-labeling strategy is introduced, utilizing an ensemble of diverse state-of-the-art architectures.<n>Experiments adapting from Semantic KITTI to unlabeled target datasets demonstrate significant improvements in segmentation accuracy.
  arXiv  Detail & Related papers  (2025-07-24T08:21:43Z)
• Syn-to-Real Unsupervised Domain Adaptation for Indoor 3D Object Detection [50.448520056844885]
  We propose a novel framework for syn-to-real unsupervised domain adaptation in indoor 3D object detection. Our adaptation results from synthetic dataset 3D-FRONT to real-world datasets ScanNetV2 and SUN RGB-D demonstrate remarkable mAP25 improvements of 9.7% and 9.1% over Source-Only baselines.
  arXiv  Detail & Related papers  (2024-06-17T08:18:41Z)
• Decoupled Pseudo-labeling for Semi-Supervised Monocular 3D Object Detection [108.672972439282]
  We introduce a novel decoupled pseudo-labeling (DPL) approach for SSM3OD. Our approach features a Decoupled Pseudo-label Generation (DPG) module, designed to efficiently generate pseudo-labels. We also present a DepthGradient Projection (DGP) module to mitigate optimization conflicts caused by noisy depth supervision of pseudo-labels.
  arXiv  Detail & Related papers  (2024-03-26T05:12:18Z)
• Visual Foundation Models Boost Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation [17.875516787157018]
  We study how to harness the knowledge priors learned by 2D visual foundation models to produce more accurate labels for unlabeled target domains. Our method is evaluated on various autonomous driving datasets and the results demonstrate a significant improvement for 3D segmentation task.
  arXiv  Detail & Related papers  (2024-03-15T03:58:17Z)
• Learning to Adapt SAM for Segmenting Cross-domain Point Clouds [25.66755977728181]
  Unsupervised domain adaptation (UDA) in 3D segmentation tasks presents a formidable challenge. We propose an innovative hybrid feature augmentation methodology, which significantly enhances the alignment between the 3D feature space and SAM's feature space. Our method is evaluated on many widely-recognized datasets and achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-10-13T02:28:40Z)
• SUG: Single-dataset Unified Generalization for 3D Point Cloud Classification [44.27324696068285]
  We propose a Single-dataset Unified Generalization (SUG) framework to alleviate the unforeseen domain differences faced by a well-trained source model. Specifically, we first design a Multi-grained Sub-domain Alignment (MSA) method, which can constrain the learned representations to be domain-agnostic and discriminative. Then, a Sample-level Domain-aware Attention (SDA) strategy is presented, which can selectively enhance easy-to-adapt samples from different sub-domains.
  arXiv  Detail & Related papers  (2023-05-16T04:36:04Z)
• SSDA3D: Semi-supervised Domain Adaptation for 3D Object Detection from Point Cloud [125.9472454212909]
  We present a novel Semi-Supervised Domain Adaptation method for 3D object detection (SSDA3D) SSDA3D includes an Inter-domain Adaptation stage and an Intra-domain Generalization stage. Experiments show that, with only 10% labeled target data, our SSDA3D can surpass the fully-supervised oracle model with 100% target label.
  arXiv  Detail & Related papers  (2022-12-06T09:32:44Z)
• Unsupervised Domain Adaptation for Monocular 3D Object Detection via Self-Training [57.25828870799331]
  We propose STMono3D, a new self-teaching framework for unsupervised domain adaptation on Mono3D. We develop a teacher-student paradigm to generate adaptive pseudo labels on the target domain. STMono3D achieves remarkable performance on all evaluated datasets and even surpasses fully supervised results on the KITTI 3D object detection dataset.
  arXiv  Detail & Related papers  (2022-04-25T12:23:07Z)
• Unsupervised Domain Adaptive 3D Detection with Multi-Level Consistency [90.71745178767203]
  Deep learning-based 3D object detection has achieved unprecedented success with the advent of large-scale autonomous driving datasets. Existing 3D domain adaptive detection methods often assume prior access to the target domain annotations, which is rarely feasible in the real world. We study a more realistic setting, unsupervised 3D domain adaptive detection, which only utilizes source domain annotations.
  arXiv  Detail & Related papers  (2021-07-23T17:19:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.