The human visual system and CNNs can both support robust online translation tolerance following extreme displacements

Keyphrases

• Accuracy Rate 33%
• Computational Vision 33%
• Convolutional Neural Network 100%
• Deep Convolutional Neural Network (deep CNN) 33%
• Degree of Match 33%
• Eye-tracking Study 33%
• Global Average Pooling Layer 33%
• Human Performance 33%
• Human Vision 33%
• Human Visual System 100%
• Invariance 66%
• Large Receptive Field 33%
• Model Performance 33%
• Near-complete 33%
• Network Support 33%
• Novel Object 33%
• Online Translation 100%
• Recognize Objects 33%
• Spatial Transform 33%
• Strong Constraint 33%
• Translation Tolerance 100%
• Unseen 33%
• Visual Angle 33%
• Visual System 33%
• Visual Translation 33%

Computer Science

• Average Pooling 50%
• Computational Modeling 50%
• Convolutional Neural Network 100%
• Deep Convolutional Neural Networks 50%
• Human Performance 50%
• Human Visual System 100%
• Receptive Field 50%
• Tracking (Position) 50%

Neuroscience

• Neural Network 100%
• Receptive Field 33%