Active Vision

Active Vision

1988 | JOHN (YIANNIS) ALOIMONOS and ISAAC WEISS, AMIT BANDYOPADHYAY
The paper by Aloimonos, Weiss, and Bandypadhyay explores the advantages of active vision over passive vision. Active vision assumes that the observer is engaged in activities that control the geometric parameters of the sensory apparatus, such as moving or rotating to improve perceptual results. The authors demonstrate that active observers can solve basic vision problems more efficiently than passive ones, making ill-posed and nonlinear problems well-posed and linear. Specifically, they show that shape from shading, depth computation, shape from contour, shape from texture, and structure from motion are easier for active observers. The paper emphasizes that active vision does not rely on feature correspondence from multiple viewpoints and introduces a general framework for addressing low-level vision problems. The authors also highlight the limitations of passive vision, which often requires additional constraints and can be unstable, while active vision can handle complex shapes and discontinuities more effectively.The paper by Aloimonos, Weiss, and Bandypadhyay explores the advantages of active vision over passive vision. Active vision assumes that the observer is engaged in activities that control the geometric parameters of the sensory apparatus, such as moving or rotating to improve perceptual results. The authors demonstrate that active observers can solve basic vision problems more efficiently than passive ones, making ill-posed and nonlinear problems well-posed and linear. Specifically, they show that shape from shading, depth computation, shape from contour, shape from texture, and structure from motion are easier for active observers. The paper emphasizes that active vision does not rely on feature correspondence from multiple viewpoints and introduces a general framework for addressing low-level vision problems. The authors also highlight the limitations of passive vision, which often requires additional constraints and can be unstable, while active vision can handle complex shapes and discontinuities more effectively.
Reach us at info@study.space
[slides] Active vision | StudySpace