What is it about?

Dissecting ab image into the individual objects it is made of is a notoriously difficult task: Colourful surfaces can create the illusion of additional object while similarly coloured objects can be hard to tell a part. Our research uses an entirely different approach: When a light source is moved around the scene, foreground objects will cast shadows onto objects that are further back in the scene. By detecting such shadows, we can detect object contours.

Featured Image

Read the Original

This page is a summary of: Image Segmentation from Shadow-Hints using Minimum Spanning Trees, July 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3641234.3671025.
You can read the full text:

Read

Contributors

The following have contributed to this page