P2-Net: Joint Description and Detection of Local Features for Pixel and Point Matching

Keyphrases

• Point Matching 100%
• Point Cloud 100%
• Local Features 100%
• Pixel Matching 100%
• 2D Image 50%
• Learning-based 50%
• Source Code 50%
• Local Region 50%
• Competitive Performance 50%
• Plethora 50%
• 2D to 3D 50%
• 3D Point Cloud 50%
• Ultra-wide 50%
• Visual Localization 50%
• Image-based Point Cloud 50%
• Intrinsic Information 50%
• Fully Convolutional 50%
• Representation Space 50%
• Novel Loss Function 50%
• Shared Latent Representation 50%
• Keypoint Detector 50%
• 3D Keypoints 50%
• 3D Local Descriptor 50%
• 3D Input 50%
• Fine-grained Matching 50%
• Local Feature Detector 50%
• 2D Keypoint 50%

Computer Science

• Point Cloud 100%
• local feature 100%
• Feature Detector 33%
• Intrinsic Information 33%
• Representation Space 33%
• feature descriptor 33%
• Experimental Result 33%

Biochemistry, Genetics and Molecular Biology

• Writing 100%
• Visual Orientation 50%