Christopher DiMattina

The paper finds that texture images are less comfortable to view than other natural images, supporting the Cortical Hyper-metabolism Hypothesis.

This study investigates how humans classify image edges as surfaces, object boundaries, or illumination changes using synthetic stimuli and finds that boundary sharpness, luminance modulation, and texture modulation interact to influence classification.

The paper tests a computational model predicting that density-defined boundaries are easier to segment than feature-defined boundaries, finding lower segmentation thresholds for density boundaries.

This paper investigates how local achromatic cues like contrast, texture, and penumbral blur help distinguish shadow edges from surface boundaries in grayscale images.

The paper investigates whether luminance texture boundaries and luminance step boundaries are segmented using different mechanisms, finding that texture boundaries are more robust to masking by step boundaries.

This study investigates how humans segment luminance texture boundaries and proposes a two-stage model to explain performance in the presence of interfering luminance step boundaries.

This paper introduces a machine learning framework to model human perception of second-order visual stimuli and finds humans use region-based and edge-based strategies for orientation tasks.