Model-based recognition of 2D objects in perspective images
Stefan Wirtz, Dietrich W. R. Paulus · Pattern Recognition and Image Analysis · 2011
We report on a case study showing that using symbolic descriptions to recognize objects in perspective images delivers similar results as heuristic or statistical methods. The knowledge is modeled in TGraphs which are typed, attributed, and ordered directed graphs. We combined the search in the state space with a maximum weight bipartite graph-matching and in consequence we were able reduce the numerous amount of hypotheses. Furthermore we used hash tables to increase the runtime efficiency. As a result we are able to show that model based object recognition using symbolic descriptions is on a competitive basis.