Object Recognition Based on Amounts of Unlabeled Data

TL;DR

This paper introduces a semi-supervised object recognition method that leverages unlabeled data using Boost Picking Teaching, achieving high accuracy with minimal labeled data on CIFAR datasets.

Contribution

It presents a novel semi-supervised approach combining Boost Picking Teaching with an ensemble strategy for improved object recognition.

Findings

Achieved 78.39% accuracy on CIFAR-10 with only 2% labeled data.

Achieved 50.77% accuracy on CIFAR-100 with only 2% labeled data.

Validated the effectiveness of the ensemble strategy theoretically.

Abstract

This paper proposes a novel semi-supervised method on object recognition. First, based on Boost Picking, a universal algorithm, Boost Picking Teaching (BPT), is proposed to train an effective binary-classifier just using a few labeled data and amounts of unlabeled data. Then, an ensemble strategy is detailed to synthesize multiple BPT-trained binary-classifiers to be a high-performance multi-classifier. The rationality of the strategy is also analyzed in theory. Finally, the proposed method is tested on two databases, CIFAR-10 and CIFAR-100. Using 2% labeled data and 98% unlabeled data, the accuracies of the proposed method on the two data sets are 78.39% and 50.77% respectively.