Low-Memory Cost Belief Propagation Architecture for Disparity Estimation

Abstract

In disparity estimation, belief propagation can deliver better disparity quality than other algorithms but suffer from large storage cost, especially at the message update processing. To reduce the storage cost, this paper proposes low-memory cost architectures for the message update PE to satisfy the real-time application. We propose four architectures which are post-normalization, shadow buffer, no memory, and no memory+double PE architectures. Compared to the previous design, the proposed no memory+double PE architecture can save 28% of the hardware cost at most for 320x240@30fps and 64 disparity levels.