Comparing Shape Representations for the Aesthetic Judgment of 3-D Shape Pairs

Area of Science:

• Computer Vision
• Perception Psychology
• Human-Computer Interaction

Background:

• Visual aesthetics are crucial for 3-D shape perception.
• Understanding how shape representation impacts aesthetic judgment is vital for 3-D modeling and data collection.

Purpose of the Study:

• To investigate the influence of different 3-D shape representations on human aesthetic judgments.
• To compare aesthetic evaluations across various representations like voxels, points, wireframe, and polygons.

Main Methods:

• Human participants evaluated the aesthetic appeal of 3-D shape pairs.
• Shapes were presented using diverse representations: voxels, points, wireframe, and polygon meshes.
• A larger and more diverse set of shape classes was analyzed compared to previous studies.

Main Results:

• Aesthetic judgments using low-resolution voxel and point representations were comparable to those using polygon meshes.
• Humans can reliably assess 3-D shape aesthetics even from coarse visual data.
• The findings hold across a wider range of shape classes.

Conclusions:

• The choice of 3-D shape representation has a limited impact on human aesthetic judgments.
• Coarse shape representations are sufficient for many aesthetic evaluation tasks.
• Implications for efficient data collection and improved 3-D modeling algorithms.