Lightness perception for matte and glossy complex shapes.

Area of Science:

• Visual Perception
• Computer Graphics
• Material Science

Background:

• Human visual perception can estimate surface albedo despite variations from shading, illumination, and specular reflections.
• Understanding how luminance distributions influence albedo perception is crucial for realistic rendering and visual science.

Purpose of the Study:

• To investigate how luminance statistics predict surface albedo for matte and glossy objects.
• To compare human judgments of albedo with physical predictions, particularly for glossy surfaces.
• To explore the impact of luminance distribution manipulation on lightness perception.

Main Methods:

• Physically based rendering simulations to analyze luminance distributions and albedo.
• Human observer experiments involving sorting objects by perceived albedo in virtual scenes.
• Selective manipulation of luminance distribution bands for matte and glossy objects.

Main Results:

• For matte surfaces, the brightest luminance areas accurately predict albedo.
• For glossy surfaces, darker luminance percentiles correlate best with albedo and human judgments.
• Human observers tend to perceive glossy objects as darker and judge their lightness less precisely, ignoring specular highlights.

Conclusions:

• Human albedo perception effectively discounts specular reflections from glossy surfaces.
• This discounting leads to a systematic bias, making glossy objects appear darker than their actual albedo.
• Luminance distribution analysis provides insights into the mechanisms of lightness perception for different surface types.