A CNN-based Feature Space for Semi-supervised Incremental Learning in Assisted Living Applications

TL;DR

This paper introduces a semi-supervised incremental learning method using CNN-derived feature space to enhance object recognition in assisted living, achieving a 40% accuracy improvement on new instances.

Contribution

It proposes leveraging CNN feature space for semi-supervised labeling to enable incremental learning in assisted living applications, which is a novel approach.

Findings

Improved classification accuracy of new instances by 40%.

Effective semi-supervised labeling using feature space.

Enhanced CNN performance in changing environments.

Abstract

A Convolutional Neural Network (CNN) is sometimes confronted with objects of changing appearance ( new instances) that exceed its generalization capability. This requires the CNN to incorporate new knowledge, i.e., to learn incrementally. In this paper, we are concerned with this problem in the context of assisted living. We propose using the feature space that results from the training dataset to automatically label problematic images that could not be properly recognized by the CNN. The idea is to exploit the extra information in the feature space for a semi-supervised labeling and to employ problematic images to improve the CNN's classification model. Among other benefits, the resulting semi-supervised incremental learning process allows improving the classification accuracy of new instances by 40% as illustrated by extensive experiments.