biostudies-literatureUnknown11Wang FWe propose a self-supervised point cloud representation learning framework CrossAlignNet based on cross-modal mask alignment strategy, to solve the problems of imbalance between global semantic and local geometric feature learning, as well as cross-modal information asymmetry in existing methods. A geometrically consistent mask region is established between the point cloud patches and the corresponding image patches through a synchronized mask alignment strategy to ensure cross-modal information symmetry. A dual-task learning framework is designed: the global semantic alignment task enhances the cross-modal semantic consistency through contrastive learning, and the local mask reconstruction task fuses the image cues using the cross-attention mechanism to recover the local geometric structuPeerJ. Computer sciencee3194https://www.ebi.ac.uk/biostudies/studies/S-EPMC12453720biostudies-literatureCrossAlignNet: a self-supervised feature learning framework for 3D point cloud understanding.PMC12453720Zhang WWu JDong XWang FZhou TfalseCrossAlignNet: a self-supervised feature learning framework for 3D point cloud understanding.We propose a self-supervised point cloud representation learning framework CrossAlignNet based on cross-modal mask alignment strategy, to solve the problems of imbalance between global semantic and local geometric feature learning, as well as cross-modal information asymmetry in existing methods. A geometrically consistent mask region is established between the point cloud patches and the corresponding image patches through a synchronized mask alignment strategy to ensure cross-modal information symmetry. A dual-task learning framework is designed: the global semantic alignment task enhances the cross-modal semantic consistency through contrastive learning, and the local mask reconstruction task fuses the image cues using the cross-attention mechanism to recover the local geometric structu2025-01-01T00:00:00Z20252026-04-30T03:15:41.994Z2026-04-30T03:11:13.519ZS-EPMC124537204098941310.7717/peerj-cs.3194