A Machine Learning Paradigm for Studying Pictorial Realism: How Accurate are Constable's Clouds?

Abstract

The British landscape painter John Constable is considered foundational for the Realist movement in 19 th -century European painting. Constable's painted skies, in particular, were seen as remarkably accurate by his contemporaries, an impression shared by many viewers today. Yet, assessing the accuracy of realist paintings like Constable's is subjective or intuitive, even for professional art historians, making it difficult to say with certainty what set Constable's skies apart from those of his contemporaries. Our goal is to contribute to a more objective understanding of Constable's realism. We propose a new machine-learning-based paradigm for studying pictorial realism in an explainable way. Our framework assesses realism by measuring the similarity between clouds painted by artists noted for their skies, like Constable, and photographs of clouds. The experimental results of cloud classification show that Constable approximates more consistently than his contemporaries the formal features of actual clouds in his paintings. The study, as a novel interdisciplinary approach that combines computer vision and machine learning, meteorology, and art history, is a springboard for broader and deeper analyses of pictorial realism.

A Multi-Stage Feature Aggregation and Structure Awareness Network for Concrete Bridge Crack Detection

“…The network proposed in this paper was trained and evaluated using the publicly available bridge crack dataset BlurredCrack [ 2 ]. To further validate the adaptability of the model, two publicly available pavement crack datasets, CrackLS315 [ 20 ] and CFD [ 9 ], were used to verify the generalization ability of the network in this paper. The BlurredCrack, CrackLS315, and CFD datasets contain 2350, 315, and 118 crack images, respectively.…”

“…In previous studies, many traditional methods have been proposed for crack detection [ 3 , 4 , 5 , 6 , 7 , 8 ]. The traditional methods mainly include edge detection [ 3 , 4 ], threshold segmentation [ 5 , 6 ], and machine learning [ 7 , 8 , 9 , 10 , 11 , 12 ] methods. The methods based on edge detection and threshold segmentation are sensitive to background noise interference and reduce the precision of crack detection under complex backgrounds.…”

“…In recent years, deep learning techniques have gained widespread popularity and demonstrated powerful performance in image classification, object detection, and image segmentation [ 9 , 13 ] and have been widely applied to crack detection. Many previous studies [ 14 , 15 , 16 , 17 , 18 ] mainly used segmentation networks (e.g., FCN [ 15 ], UNet [ 16 ], SegNet [ 17 ], etc.)…”