A Game of Boxes

As humans (and animals) are moving around, we can keep track of objects in our environment even if some of those objects are disappearing out of our field of view. To this day, there is no detailed computational model in literature that suggests how humans are doing this. We aim to find a 3D-reconstruction-based or a view-based model (no explicit definition of a 3D coordinate frame) that help us find an explanation on how humans are forming a representation of a scene.