An apparatus for generating a model for pose estimation of a system obtains training data for a multiple locations. The training data includes one or more images captured by an image capturing device and the respective poses of the captured images. At least one data sample is generated from the training data for each of the captured images, where a data sample for an image is an assignment of the image and at least one other image selected from the training data to respective poses. A neural network is trained with a data set made up of the data samples to estimate a respective pose of a localization image from: the localization image; at least one additional image from the training data; and a respective pose of each of additional image.