Object Detection using Semantic Decomposition for Energy-Efficient Neural Computing

Machine-learning algorithms are used to solve classification problems across a broad range of computing domains, from data centers to wearable technology, and place significant demand on their computing abilities.In this paper, we present a new approach to optimize energy efficiency of machine-learning using semantic decomposition to build a hierarchical framework of classifiers. We observe that certain semantic information like color or texture are common across various images in real-world data-sets for object detection applications.We exploit these common semantic features to distinguish the objects of interest from the remaining inputs in a data-set at a lower computational effort. We build a hierarchical framework of classifiers, with increasing levels of complexity, trained to recognize the broad representative semantic features of the input. The degree of confidence at the output of a classifier is used to decide whether classification can be terminated at the current stage. Our methodology thus allows us to transform a given classification algorithm into a semantically decomposed hierarchical framework. We use color and texture as distinctive traits to carry out several experiments for object detection from the Caltech data-set. Our experiments show that the proposed method yields 1.93 times improvement in average energy over the traditional single classifier model.