Revisiting Image Aesthetic Assessment via Self-Supervised Feature Learning

Abstract

Visual aesthetic assessment has been an active research field for decades. Although latest methods have achieved promising performance on benchmark datasets, they typically rely on a large number of manual annotations including both aesthetic labels and related image attributes. In this paper, we revisit the problem of image aesthetic assessment from the self-supervised feature learning perspective. Our motivation is that a suitable feature representation for image aesthetic assessment should be able to distinguish different expert-designed image manipulations, which have close relationships with negative aesthetic effects. To this end, we design two novel pretext tasks to identify the types and parameters of editing operations applied to synthetic instances. The features from our pretext tasks are then adapted for a one-layer linear classifier to evaluate the performance in terms of binary aesthetic classification. We conduct extensive quantitative experiments on three benchmark datasets and demonstrate that our approach can faithfully extract aesthetics-aware features and outperform alternative pretext schemes. Moreover, we achieve comparable results to state-of-the-art supervised methods that use 10 million labels from ImageNet.

AAAI Conference on Artificial Intelligence, 2020, accepted

References in corpus (7)

• Unsupervised Person Re-identification: Clustering and Fine-tuning
• High-Fidelity Image Generation With Fewer Labels
• PatchShuffle Regularization
• Photo Aesthetics Ranking Network with Attributes and Content Adaptation
• Deep Cropping via Attention Box Prediction and Aesthetics Assessment
• RankIQA: Learning from Rankings for No-reference Image Quality Assessment
• Effective Aesthetics Prediction with Multi-level Spatially Pooled Features