Aperture Problem in Computer Vision
- Without distinct features to track, the true visual motion is ambiguous
- Locally, one can compute only the component of the visual motion in the direction perpendicular to the contour
In relation to the Internet: our browsing is also very static and in a small window. By default, you can only browse on a page-level scale. What about subdomains? Apex domains (see more in DNS)? How do we expand the aperture of the web browser to encompass more information scales? (potential idea)