Heterogeneous Prototype Learning From Contaminated Faces Across Domains via Disentangling Latent Factors

Keyphrases

• Adversarial Games 11%
• Cross-domain Learning 11%
• Domain-specific Features 11%
• Encoder-decoder 11%
• Face Image 11%
• Face Synthesis 22%
• Facial Variation 11%
• Feature Space 22%
• Heterogeneous Data 11%
• Identity Characteristics 11%
• Identity Information 11%
• Latent Factors 100%
• Latent Feature Space 11%
• Learning Framework 11%
• Learning Objectives 11%
• Prototype Learning 100%
• Source Domain 22%
• Synthesis Methods 11%
• Synthesis Problem 11%
• Target Domain 11%

Computer Science

• Discriminator 22%
• Feature Space 33%
• Identity Information 11%
• Learning Framework 11%
• Prototype Learning 100%
• Synthesis Problem 11%