Object-level Scene Deocclusion

Keyphrases

• Reocclusion 100%
• Object-level 100%
• Feature Map 50%
• Real Scene 33%
• Training Model 16%
• Training Set 16%
• 3D Reconstruction 16%
• Large-scale Dataset 16%
• Single View 16%
• Diffusion Framework 16%
• Efficiency Improvement 16%
• Variational Autoencoder 16%
• Recomposition 16%
• Cross-domain 16%
• Large Margin 16%
• Partial View 16%
• Amodal 16%
• Occluded Region 16%
• Self-supervised Learning 16%
• Cocoa 16%
• Text Prompt 16%
• Foundation Models 16%

Computer Science

• Feature Map 100%
• Annotation 33%
• 3d Scenes 33%
• Self-Supervised Learning 33%
• Pre-Trained Model 33%
• Variational Autoencoder 33%