Shape-Graph Based Object Recognition Using Node Context Embedding

Authors: Marton Szemenyei, Ferenc Vajda

Abstract: Graphical object representation is fequently used for visual object recognition and detection methods. Since most machine learning methods requira vectorial input, significant research has been done on assigning feature vectors to graphs - a process known as graph embedding. However, when one wishes to detect objects in a larger scene, it is a more viable strategy to assign feature vectors to graph nodes, and classify them individually. In this paper, we present a graph node embedding algorithm for 3D object detection based on primitive shape graphs. Our embedding algorithm encodes the local context of the selected node into the feature vector, thus improving the classification accuracy of nodes. The method also imposes no restriction on the structure of the graphs or the weights on the nodes and edges. The method presented here will be used as part of an intelligent object pairing algorithm for Tangible Augmented Reality.

WSEAS Transactions on Information Science and Applications, ISSN / E-ISSN: 1790-0832 / 2224-3402, Volume 15, 2018, Art. #10

PDF
DOI
XML
Cite
×
Citation Tools
WSEAS AMA
Marton Szemenyei, Ferenc Vajda, "Shape-Graph Based Object Recognition Using Node Context Embedding," WSEAS Transactions on Information Science and Applications, vol. 15, pp. 91-99, 2018, DOI:
Marton Szemenyei, Ferenc Vajda. Shape-Graph Based Object Recognition Using Node Context Embedding. WSEAS Transactions on Information Science and Applications. 2018;15:91-99.
Copy to Clipboard
Citation copied to Clipboard

Certification

---