Force histograms and neural networks for human-based spatial relationship generalization

The aim of this paper is the design of a system that can learn any individual user's perception of four subjective and key spatial relationships between image objects: "to the right of " "above, " "to the left of " and "below." The proposed approach is based on the utilization of artificial Neural Networks (NNs) and the modeling of relative positions between 2D objects through histograms of forces. The NNs are fed by features extracted from the histograms and are trained to numerically assess the four relationships according to the target perception.