Paintings and Drawings Aesthetics Assessment with Rich Attributes for Various Artistic Categories

TL;DR

This work addresses the lack of multi-attribute, multi-category aesthetics data for paintings by introducing the APDD dataset, which contains 4,985 paintings across 24 categories and 10 aesthetic attributes with over 31,100 annotations from 51 experts. It also presents AANSPS, a multi-branch network based on EfficientNet-B4 with Efficient Channel Attention to predict total aesthetic scores and per-attribute scores, trained via a two-stage scheme and pretraining on a related model. The authors demonstrate that AANSPS outperforms strong baselines on APDD in terms of MSE, MAE, and SROCC, indicating improved interpretability and reliability for painting aesthetics. Together, APDD and AANSPS provide a valuable resource and methodological framework to advance computational aesthetics for paintings and drawings, enabling more nuanced, category-aware evaluations with potential applications in art analysis and curation. Additionally, the work discusses a practical annotation platform and outlines directions for expanding the dataset and attributes to better capture the diversity of painting traditions, styles, and subjects $k=\psi(c)=| \frac{\log_{2}{C}}{\gamma} + \frac{b}{\gamma} |_{odd}$ with $\gamma=2$ and $b=1$, yielding $k=7$ for $C=1792$.$

Abstract

Image aesthetic evaluation is a highly prominent research domain in the field of computer vision. In recent years, there has been a proliferation of datasets and corresponding evaluation methodologies for assessing the aesthetic quality of photographic works, leading to the establishment of a relatively mature research environment. However, in contrast to the extensive research in photographic aesthetics, the field of aesthetic evaluation for paintings and Drawings has seen limited attention until the introduction of the BAID dataset in March 2023. This dataset solely comprises overall scores for high-quality artistic images. Our research marks the pioneering introduction of a multi-attribute, multi-category dataset specifically tailored to the field of painting: Aesthetics of Paintings and Drawings Dataset (APDD). The construction of APDD received active participation from 28 professional artists worldwide, along with dozens of students specializing in the field of art. This dataset encompasses 24 distinct artistic categories and 10 different aesthetic attributes. Each image in APDD has been evaluated by six professionally trained experts in the field of art, including assessments for both total aesthetic scores and aesthetic attribute scores. The final APDD dataset comprises a total of 4985 images, with an annotation count exceeding 31100 entries. Concurrently, we propose an innovative approach: Art Assessment Network for Specific Painting Styles (AANSPS), designed for the assessment of aesthetic attributes in mixed-attribute art datasets. Through this research, our goal is to catalyze advancements in the field of aesthetic evaluation for paintings and drawings, while enriching the available resources and methodologies for its further development and application.

Figures (7)

• Figure 1: Samples from the APDD dataset. APDD covers 24 artistic categories and 10 aesthetic attributes. Different artistic categories correspond to different sets of attributes.
• Figure 2: 24 Artistic Categories in the APDD Dataset.
• Figure 3: Correspondence between artistic categories and aesthetic attributes
• Figure 4: Representative Images for the category of oil painting - symbolism - portraiture
• Figure 5: The score distribution of APDD. This distribution suggests a consensus on the aesthetic appeal of images, with most perceived as having a moderate level of appeal, while extreme scores are less frequent.