Text-to-image synthesis with self-supervised learning

Computer Science

• Image Synthesis 100%
• Self-Supervised Learning 100%
• Discriminator 20%
• Generative Adversarial Networks 10%
• Image Content 10%
• Structural Similarity 10%
• Distance Function 10%
• Graphic Design 10%
• Synthesis Model 10%

Keyphrases

• Label Smoothing 20%
• Training Instability 10%
• Penalty Value 10%
• Content Diversity 10%
• Matching Distance 10%