The perception of translucency from surface gloss

Area of Science:

• Visual perception
• Computational neuroscience
• Computer vision

Background:

• Translucent objects like fruit and wax create complex images due to light reflection and transmission.
• Previous research highlights the importance of perceived shape and shading in visual translucency perception.
• The role of 3D shape inferred from surface gloss (shape from specular highlights) in translucency perception remains less understood.

Purpose of the Study:

• To investigate how interactions between specular and non-specular image properties, derived from different 3D shape cues, influence the perception of translucency.
• To develop and validate a computational model that predicts perceived translucency based on image features.

Main Methods:

• Experiments were conducted to examine the influence of incongruent 3D shape information in specular and non-specular image components on perceived translucency.
• A novel computational model was developed using measurable image features related to shading and specular highlights.
• Model performance was evaluated using 10-fold cross-validation.

Main Results:

• Perceived translucency can be explained by the incongruence between 3D shape cues used for specular and non-specular image components.
• The proposed computational model, based on shading relative to specular highlights, explained 59% of the variability in perceived translucency judgments.
• The model outperformed alternative models relying on explicit subjective measures of perceived surface shape.

Conclusions:

• The visual system infers translucency from the interplay of specular and non-specular shading in glossy, semi-opaque materials.
• The developed computational model implicitly captures relevant geometric information for predicting observer judgments of translucency.
• Understanding the integration of different visual cues is crucial for explaining the perception of material properties like translucency.