Representations for Cognitive Vision:A Review of Appearance-Based, Spatio-Temporal, and Graph-Based Approaches

Abstract

The emerging discipline of cognitive vision requires a proper representation of visual information including spatial and temporal relationships, scenes, events, semantics and context. This review article summarizes existing representational schemes in computer vision which might be useful for cognitive vision, and discusses promising future research directions. The various approaches are categorized according to appearance-based, spatio-temporal, and graph-based representations for cognitive vision. While the representation of objects has been covered extensively in computer vision research, both from a reconstruction as well as from a recognition point of view, cognitive vision will also require new ideas how to represent scenes. We introduce new concepts for scene representations and discuss how these might be efficiently implemented in future cognitive vision systems.