Texture properties affecting the accuracy of surface attitude judgements

Area of Science:

• Visual perception
• Computational neuroscience
• Computer vision

Background:

• Previous research focused on geometric deformations in texture perception.
• Fewer studies investigated texture's intrinsic spatial properties influencing shape perception.
• Existing studies often confound multiple texture properties.

Purpose of the Study:

• To investigate how intrinsic texture properties influence surface attitude perception.
• To isolate the roles of texels, spatial scale, discrete symmetries, and regularity.
• To analyze perception of slant and tilt for synthetic naturalistic textures.

Main Methods:

• Utilized a broad range of synthetic textures.
• Isolated key texture properties: texels, spatial scale, discrete symmetries, and regularity.
• Assessed judgments of surface slant and tilt.

Main Results:

• Texels, spatial scale structure, and discrete symmetries significantly impact attitude perception.
• Discrete rotational symmetries are crucial for accurate tilt estimation, potentially via skew symmetry or linear perspective.
• Multi-scale texture distribution enhances accuracy across a wider range of viewing distances.
• Observed minor biases in symmetric textures, aligning with Bayesian cue combination models.

Conclusions:

• Intrinsic texture properties are vital for visual surface attitude perception.
• Discrete symmetries, particularly rotational, play a significant role in depth and orientation judgments.
• Multi-scale texture processing expands the operational range for accurate 3D shape perception from texture.