Multi-view KPConv for enhanced 3D point cloud semantic segmentation using multi-modal fusion with 2D images

Keyphrases

• 2D Image 100%
• Multimodal Fusion 100%
• 3D Point Cloud Semantic Segmentation 100%
• 3D Point Cloud 66%
• Fusion Algorithm 66%
• Performance Improvement 33%
• Geometric Information 33%
• Segmentation Problem 33%
• 2D to 3D 33%
• Deep Learning Algorithm 33%
• 3D Scene 33%
• 3D Semantic Segmentation 33%
• Lifting Technique 33%
• Performance Advantage 33%
• 2D Image Feature 33%
• Custom Dataset 33%
• COLMAP 33%
• Fusion Structure 33%
• Multimodal Algorithms 33%

Computer Science

• Point Cloud 100%
• Image Segmentation 100%
• Open Source 33%
• Geometric Information 33%
• Segmentation Task 33%
• image feature 33%
• 3d Scenes 33%
• Fusion Structure 33%
• Deep Learning Method 33%

Engineering

• Point Cloud 100%
• Multi-Modal Fusion 100%
• Supplementary Information 33%
• Fusion Structure 33%
• Deep Learning Method 33%