Abstract:Domain Generalization (DG) has been recently explored to improve the generalizability of point cloud classification (PCC) models toward unseen domains. However, they often suffer from limited receptive fields or quadratic complexity due to the use of convolution neural networks or vision Transformers. In this paper, we present the first work that studies the generalizability of state space models (SSMs) in DG PCC and find that directly applying SSMs into DG PCC will encounter several challenges: the inherent topology of the point cloud tends to be disrupted and leads to noise accumulation during the serialization stage. Besides, the lack of designs in domain-agnostic feature learning and data scanning will introduce unanticipated domain-specific information into the 3D sequence data. To this end, we propose a novel framework, PointDGMamba, that excels in strong generalizability toward unseen domains and has the advantages of global receptive fields and efficient linear complexity. PointDGMamba consists of three innovative components: Masked Sequence Denoising (MSD), Sequence-wise Cross-domain Feature Aggregation (SCFA), and Dual-level Domain Scanning (DDS). In particular, MSD selectively masks out the noised point tokens of the point cloud sequences, SCFA introduces cross-domain but same-class point cloud features to encourage the model to learn how to extract more generalized features. DDS includes intra-domain scanning and cross-domain scanning to facilitate information exchange between features. In addition, we propose a new and more challenging benchmark PointDG-3to1 for multi-domain generalization. Extensive experiments demonstrate the effectiveness and state-of-the-art performance of our presented PointDGMamba.

Domain Adaptive Sampling for Cross-Domain Point Cloud Recognition

Domain Adaptation on Point Clouds Via Geometry-Aware Implicits

Self-Supervised Global-Local Structure Modeling for Point Cloud Domain Adaptation with Reliable Voted Pseudo Labels

Point Cloud Domain Adaptation Via Masked Local 3D Structure Prediction

Multiscale Attention-Based Subdomain Dynamic Adaptation for Cross-Domain Scene Classification

Cross-Dataset Point Cloud Recognition Using Deep-Shallow Domain Adaptation Network.

Point-Based Multilevel Domain Adaptation for Point Cloud Segmentation

PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representation

Masked Local-Global Representation Learning for 3D Point Cloud Domain Adaptation

Unsupervised Domain Adaptive Point Cloud Semantic Segmentation.

Learning to Adapt SAM for Segmenting Cross-domain Point Clouds

Self-Supervised Boundary Point Prediction Task for Point Cloud Domain Adaptation

Unsupervised Spatial-Awareness Attention-Based and Multi-Scale Domain Adaption Network for Point Cloud Classification.

DACS: Domain Adaptation via Cross-domain Mixed Sampling

A Learnable Self-supervised Task for Unsupervised Domain Adaptation on Point Clouds

Bridging Domain Gap of Point Cloud Representations via Self-Supervised Geometric Augmentation

Domain Adaptation on Point Clouds for 6D Pose Estimation in Bin-Picking Scenarios

PointDGMamba: Domain Generalization of Point Cloud Classification via Generalized State Space Model

Epointda: an End-to-End Simulation-to-Real Domain Adaptation Framework for LiDAR Point Cloud Segmentation

Compositional Semantic Mix for Domain Adaptation in Point Cloud Segmentation