Object recognition and articulated object learning by accumulative Hopfield matching

✍ Scribed by Wen-Jing Li; Tong Lee

• Publisher: Elsevier Science
• Year: 2002
• Tongue: English
• Weight: 560 KB
• Volume: 35
• Category: Article
• ISSN: 0031-3203
• DOI: 10.1016/s0031-3203(01)00158-3

✦ Synopsis

In this paper, a novel object recognition method based on attributed relational graph matching is proposed, which is called accumulative Hopÿeld matching. We ÿrst divide the scene graph into many sub-graphs, and a modiÿed Hopÿeld network is then constructed to obtain the sub-graph isomorphism between each sub-scene graph and model graph. The ÿnal result is deduced by accumulating the solutions of all small sub-networks. Comparing to the traditional Hopÿeld network, the proposed system has the advantage of ÿnding homomorphic mappings between two graphs. Furthermore, the system can be applied for articulated object recognition and visual model learning, which is considered as a di cult topic till now. The proposed method has been evaluated with real images.