Model-based 3D Hand Reconstruction via Self-Supervised Learning

Keyphrases

• Self-supervised Learning 100%
• 3D Hand Reconstruction 100%
• Geometric Cues 66%
• Reconstruction Network 66%
• 3D Annotation 66%
• Training Data 33%
• Neural Network 33%
• State-of-the-art Techniques 33%
• Reconstruction Model 33%
• Single View 33%
• Depth Ambiguity 33%
• Training Stages 33%
• Manual Annotation 33%
• 3D Representation 33%
• Fully-supervised 33%
• Hand Configuration 33%
• Supervised Methods 33%
• RGB Image 33%
• Jointly Estimate 33%
• Hand Reconstruction 33%
• Monocular Image 33%
• 2D Representation 33%
• Self-supervised Method 33%
• Camera Viewpoint 33%

Computer Science

• Annotation 100%
• Self-Supervised Learning 100%
• Supervised Method 66%
• Training Data 33%
• Neural Network 33%
• Hand Configuration 33%
• RGB Image 33%