Outperforming Image Segmentation by Exploiting Approximate K-Means Algorithms

摘要: Recently emerged as an effective approach, Approximate Computing introduces a new design paradigm for trade system overhead off result quality. Indeed, by relaxing the need fully precise outcome, techniques allow to gain performance parameters, such computational time or area of integrated circuits, executing inexact operations. In this work, we propose approximate version K-means algorithm be used image segmentation, with aim reduce needed synthesize it on hardware target. particular, detail methodology find variants and some experimental evidences proof-of-concept.