Philippe SCHYNS
Diagnostic Recognition: Task Constraints, Object Information, and their Interactions
Abstract:Object recognition and categorization are both concerned with the understanding of how input information matches with memorized object information. It is therefore surprising that these two fields have evolved independently, without much cross-fertilization between their research. It is the main objective of this paper to lay out the basis of a dialogue between object recognition and categorization research, with the hope of raising issues the could cross-fertilize both domains. To this end, this paper develops diagnostic recognition, a framework which formulates recognition performance as an interaction of task constraints and object information. We argue and present examples suggesting that diagnostic recognition could be applied to the understanding of everyday object recognition.