Data-Side Efficiencies for Lightweight Convolutional Neural Networks

TL;DR

This paper investigates how data attributes like class count, color, resolution, and scale influence the design of lightweight CNNs, proposing metrics to efficiently evaluate their impact and improve model efficiency.

Contribution

It introduces metric learning-based intra- and inter-class similarity metrics to guide data attribute selection for lightweight CNNs, reducing computation by up to 30x.

Findings

Metrics require 30x less computation than full inference.

Applying metrics reduced computation by 66% in a robot path planning task.

Achieved a 3.5% accuracy increase using the proposed approach.

Abstract

We examine how the choice of data-side attributes for two important visual tasks of image classification and object detection can aid in the choice or design of lightweight convolutional neural networks. We show by experimentation how four data attributes - number of classes, object color, image resolution, and object scale affect neural network model size and efficiency. Intra- and inter-class similarity metrics, based on metric learning, are defined to guide the evaluation of these attributes toward achieving lightweight models. Evaluations made using these metrics are shown to require 30x less computation than running full inference tests. We provide, as an example, applying the metrics and methods to choose a lightweight model for a robot path planning application and achieve computation reduction of 66% and accuracy gain of 3.5% over the pre-method model.