Rotation-Adaptive Point Cloud Domain Generalization via Intricate Orientation Learning
- 28 January 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (5) , 4232-4239
Abstract
The vulnerability of 3D point cloud analysis to unpredictable rotations poses an open yet challenging problem: orientation-aware 3D domain generalization. Cross-domain robustness and adaptability of 3D representations are crucial but not easily achieved through rotation augmentation. Motivated by the inherent advantages of intricate orientations in enhancing generalizability, we propose an innovative rotation-adaptive domain generalization framework for 3D point cloud analysis. Our approach aims to alleviate orientational shifts by leveraging intricate samples in an iterative learning process. Specifically, we identify the most challenging rotation for each point cloud and construct an intricate orientation set by optimizing intricate orientations. Subsequently, we employ an orientation-aware contrastive learning framework that incorporates an orientation consistency loss and a margin separation loss, enabling effective learning of categorically discriminative and generalizable features with rotation consistency. Extensive experiments and ablations conducted on 3D cross-domain benchmarks firmly establish the state-of-the-art performance of our proposed approach in the context of orientation-aware 3D domain generalization.Keywords
All Related Versions
Funding Information
- China National Key R&D Program (2023YFE0202700, 2024YFB4709200)
- Key-Area Research and Development Program of Guangzhou City (2023B01J0022)
- Guangdong Provincial Natural Science Foundation for Outstanding Youth Team (2024B1515040010)
- NSFC Key project (U23A20391)
- Guangdong Natural Science Funds for Distinguished Young Scholars (2023B1515020097)
- AI Singapore Programme
- National Research Foundation Singapore (AISG3-GV-2023-011)
- Lee Kong Chian Fellowships
This publication has 31 references indexed in Scilit:
- Point Cloud Domain Adaptation via Masked Local 3D Structure Prediction2022
- SVNet: Where SO(3) Equivariance Meets Binarization on Point Cloud Representation2022
- RIConv++: Effective Rotation Invariant Convolutions for 3D Point Clouds Deep LearningInternational Journal of Computer Vision, 2022
- Global Context Aware Convolutions for 3D Point Cloud Understanding2020
- Unsupervised Person Re-Identification by Soft Multilabel Learning2019
- ScanNet: Richly-Annotated 3D Reconstructions of Indoor Scenes2017
- FaceNet: A unified embedding for face recognition and clustering2015
- 3D ShapeNets: A deep representation for volumetric shapes2015
- Dynamic 3D Scene Analysis from a Moving Vehicle2007
- Integrating structured biological data by Kernel Maximum Mean DiscrepancyBioinformatics, 2006