Litcius/Paper detail

FG-Conv: Large-Scale LiDAR Point Clouds Understanding Leveraging Feature Correlation Mining and Geometric-Aware Modeling

Kangcheng Liu, Zhi Gao, Feng Lin, Ben M. Chen

202133 citationsDOI

Abstract

This work presents a general deep learning framework for large-scale point clouds understanding without voxelizations, called FG-Conv, which achieves an accurate and real-time understanding of point clouds. Through our novel design combining feature level correlation mining and deformable convolutions based geometric aware modeling, the local feature relationships and geometric patterns can be captured. The attention mechanism is also adopted to enhance the global long-range feature correlations. Finally, the feature pyramid residual learning network is proposed to combine patterns at different resolutions in a memory-efficient way. Extensive experiments on real-world challenging datasets demonstrated that our approaches outperform state-of-the-art methods in terms of accuracy and efficiency. Weakly supervised transfer learning demonstrates the generalization capacity of our methods.

Topics & Concepts

Computer sciencePoint cloudFeature (linguistics)Pyramid (geometry)Artificial intelligenceGeneralizationLidarScale (ratio)Pattern recognition (psychology)ResidualFeature extractionPoint (geometry)Machine learningData miningAlgorithmRemote sensingMathematicsGeographyMathematical analysisGeometryLinguisticsPhilosophyCartography3D Shape Modeling and Analysis3D Surveying and Cultural HeritageRemote Sensing and LiDAR Applications
FG-Conv: Large-Scale LiDAR Point Clouds Understanding Leveraging Feature Correlation Mining and Geometric-Aware Modeling | Litcius