Blurring the boundary between models and reality: Visual perception of scale assessed by performance

Area of Science:

• Visual Perception
• Computational Neuroscience
• Image Processing

Background:

• Visual perception constructs 3D representations from 2D retinal images.
• Depth cues in images lack inherent scale information.
• Image blur, arising from limited depth of field, offers potential scale cues.

Purpose of the Study:

• Investigate the role of image blur gradients in visual scale perception.
• Provide performance-based evidence for human use of blur gradients in scale judgments.
• Determine the influence of blur gradient orientation and rate of change on scale perception.

Main Methods:

• Artificial manipulation of image blur to simulate "fake tilt shift miniaturization."
• Forced-choice tasks comparing real scenes and scale models.
• Analysis of performance based on blur gradient orientation and rate.

Main Results:

• Human vision utilizes image blur gradients for scale estimation.
• Performance-based evidence confirms the use of blur cues in distinguishing real scenes from scale models.
• Blur gradient orientation is critical, while the rate of change is less important.

Conclusions:

• Image blur gradients are a significant cue for visual scale perception.
• Human visual system employs a coarse analysis of blur orientation for scale judgments.
• Findings contribute to understanding how visual perception estimates absolute depth and size.