Active vision is a framework where the observer actively controls the geometric parameters of the sensory apparatus to improve perceptual results. This approach contrasts with passive vision, where the observer does not actively manipulate the sensory input. Active vision allows for solving basic vision problems more efficiently, as ill-posed and nonlinear problems become well-posed and linear. Examples include shape from shading, shape from contour, shape from texture, and structure from motion. These problems are shown to be easier for an active observer, as they do not require correspondence between features from multiple viewpoints. Active vision does not imply active sensing but rather a general methodology for addressing low-level vision problems. Human perception is active, involving exploratory and searching behavior. This active behavior allows humans to efficiently perform visual tasks. In contrast, passive vision has been shown to be problematic, as many basic problems are ill-posed and require regularization. Regularization, while useful, has limitations in handling the full complexity of vision, especially in the presence of discontinuities. Passive vision problems can also be unstable, where small errors in input can lead to catastrophic results. Active vision addresses these issues by allowing for unique solutions and stable algorithms. For example, shape from shading can be solved uniquely with an active observer, without requiring assumptions about surface smoothness or albedo. Shape from contour can be solved using linear equations, and shape from texture can be solved without assumptions. Structure from motion can be solved using linear equations and closed-form solutions, without relying on local correspondence. Table 1 compares the performance of passive and active observers in solving basic vision problems. The results show that active vision provides significant advantages in terms of computational theory, algorithms, and implementation. The study highlights the potential of active vision in addressing the limitations of passive vision.Active vision is a framework where the observer actively controls the geometric parameters of the sensory apparatus to improve perceptual results. This approach contrasts with passive vision, where the observer does not actively manipulate the sensory input. Active vision allows for solving basic vision problems more efficiently, as ill-posed and nonlinear problems become well-posed and linear. Examples include shape from shading, shape from contour, shape from texture, and structure from motion. These problems are shown to be easier for an active observer, as they do not require correspondence between features from multiple viewpoints. Active vision does not imply active sensing but rather a general methodology for addressing low-level vision problems. Human perception is active, involving exploratory and searching behavior. This active behavior allows humans to efficiently perform visual tasks. In contrast, passive vision has been shown to be problematic, as many basic problems are ill-posed and require regularization. Regularization, while useful, has limitations in handling the full complexity of vision, especially in the presence of discontinuities. Passive vision problems can also be unstable, where small errors in input can lead to catastrophic results. Active vision addresses these issues by allowing for unique solutions and stable algorithms. For example, shape from shading can be solved uniquely with an active observer, without requiring assumptions about surface smoothness or albedo. Shape from contour can be solved using linear equations, and shape from texture can be solved without assumptions. Structure from motion can be solved using linear equations and closed-form solutions, without relying on local correspondence. Table 1 compares the performance of passive and active observers in solving basic vision problems. The results show that active vision provides significant advantages in terms of computational theory, algorithms, and implementation. The study highlights the potential of active vision in addressing the limitations of passive vision.