Learning Modality-Invariant Features by Cross-Modality Adversarial Network for Visual Question Answering

Keyphrases

• Cross-modal 100%
• Adversarial Network 100%
• Visual Question Answering 100%
• Modality-invariant Feature 100%
• Learning Modalities 100%
• Prediction Accuracy 20%
• Web Application 20%
• Model Accuracy 20%
• Fusion Method 20%
• Attention Mechanism 20%
• Large Gap 20%
• Attention Learning 20%
• Multimodal Fusion 20%
• Adversarial Learning 20%
• Question Answering Model 20%
• Multimodal Information 20%
• Multimodal Features 20%
• Alignment Performance 20%
• Answer Prediction 20%
• Semantic Alignment 20%
• Multimodal Task 20%

Computer Science

• Visual Question Answering 100%
• Multimodal Information 20%
• Prediction Accuracy 20%
• Web Application 20%
• Attention (Machine Learning) 20%
• Invariant 20%