Semi-supervised Ranking for Object Image Blur Assessment

TL;DR

This paper introduces a semi-supervised ranking approach for object image blur assessment, utilizing pairwise rank labels and a self-supervised quadruplet ranking consistency to improve blur scoring accuracy.

Contribution

It proposes a novel semi-supervised framework combining pairwise ranking labels and self-supervised learning for more effective image blur assessment.

Findings

Effective blur score prediction using pairwise rank labels.

Improved performance with the self-supervised quadruplet ranking method.

Demonstrated superiority over existing methods on a new large-scale dataset.

Abstract

Assessing the blurriness of an object image is fundamentally important to improve the performance for object recognition and retrieval. The main challenge lies in the lack of abundant images with reliable labels and effective learning strategies. Current datasets are labeled with limited and confused quality levels. To overcome this limitation, we propose to label the rank relationships between pairwise images rather their quality levels, since it is much easier for humans to label, and establish a large-scale realistic face image blur assessment dataset with reliable labels. Based on this dataset, we propose a method to obtain the blur scores only with the pairwise rank labels as supervision. Moreover, to further improve the performance, we propose a self-supervised method based on quadruplet ranking consistency to leverage the unlabeled data more effectively. The supervised and self-supervised methods constitute a final semi-supervised learning framework, which can be trained end-to-end. Experimental results demonstrate the effectiveness of our method.