Stereo Correspondence with Discrete-Time Cellular Neural Networks

In this paper, we propose a new approach of solving the stereopsis problem with a discrete-time cellular neural network(DTCNN) where each node has connectionsonly with its local neighbors.Because the matching process of stereo correspondencedepends on its geometricallylocal characteristics, the DTCNN is suitable for the stereo correspondence.Moreover, it can be easily implemented in VLSI. Therefore, we employed a two-layer DTCNN with dual templates, which are determinedwith the back propagation learning rule. Based on evaluation of the proposed approach for several random dot stereograms, its performance is better than that of the Marr-Poggio algorithm.

• Type: Conference
• Year: 1994
• Where: ISCAS
• Authors: Sungjun Park, Seung-Jai Min, Soo-Ik Chae
• Keywords: Hardware, ISCAS 1994, Propagation Learning Rule, Stereo Correspondencedepends, Stereopsis Problem