Image-based parameter estimation of 3D object parameters' is a subject that belongs to the science of 'computer vision', which is a variant of human vision. Computer vision intends to develop capabilities within computers, which equal human capabilities without suffering from human shortcomings of getting tired or decreasing concentration, or are even better than human's capabilities. Much progress has been made in this area since the 60's, which is the period, the interest of scientists for this area has been increased rapidly. But to be honest, the ability of humans to recognise e.g. faces is one of the visual tasks that is still being done better by humans then by computers, although progress is still being made in this area. On the other hand, the ability of humans to make accurate estimates of the size of an object or of its position and orientation is a task being done better by computers.This thesis describes a general framework for parameter estimation, which is suitable for computer vision applications. The approach described combines 3D modelling, animation and estimation tools to determine parameters of objects in a scene from 2D grey-level images. The animation tool predicts images using a 3D model of the scene (virtual reality), describing components like cameras, light sources and objects, and their parameters. The 3D modelling, using primitives like quadrics enables the handling of occlusion. A least squares estimator in combination with the modelling tool is used to estimate the selected parameters from real or animated grey-level images. The non-linear relation between the measurements and the set of parameters is coped with by the iterative application of the linear estimator. The least squares estimation paradigm is applied in a standardised way to the grey-level images of objects by considering the pixels as measurements of the object parameters. Two or more images, even when taken from different points of view, can be included simply by extending the measurement vector. Also the inclusion of a Gaussian filter, to increase the estimator performance by improving the image properties, can be carried out in a natural way.The framework is applied to three different applications. The pose of a cube is estimated from monocular and stereo images. A digital elevation map (DEM) is estimated from 2 successive images of an image sequence acquired by a moving camera. The calibration of internal and external camera parameters is the 3 rd application. The results of all the experiments confirm the expected capabilities of model-based parameter estimation from greylevel images. Special attention is given to the convergence properties of the estimator.Experiments, using real images to estimate the pose of a cube, indicate properties of the estimator, which seem to justify the results obtained with animated images. For more definite conclusions it is required to improve the modelling (camera lens) and the experimental setup to allow more accurate determination of external parameters. Real images of a mock-up o...