Can computers learn from humans to see better? Inferring scene semantics from viewers' eye movements

Keyphrases

• Eye Movements 100%
• Scene Semantics 100%
• Saccades 33%
• Scene Understanding 33%
• Human Activities 16%
• Facial Expression 16%
• Strong Correlation 16%
• Nude 16%
• Computer Vision 16%
• Saliency 16%
• Interactive Behavior 16%
• Verbal Description 16%
• Computer Vision Algorithms 16%
• Free Viewing 16%
• Social Scenes 16%
• Complex Scenes 16%
• Linguistic Description 16%
• Eye Saccades 16%
• Vision Understanding 16%
• Viewing Task 16%
• Linguistic Tasks 16%
• Entity Relationship 16%
• Semantic Relations 16%
• Scene Description 16%
• Semantic Gap 16%

Computer Science

• Scene Understanding 100%
• Facial Expression 50%
• Computer Vision 50%
• Computer Vision Algorithms 50%
• semantic gap 50%
• Semantic Relationship 50%
• Scene Description 50%
• Interactive 50%

Psychology

• Eye Movement 100%
• Saccade 50%
• Computer Vision 33%
• Facial Expression 16%