Find Me a Sky: A Data-Driven Method for Color-Consistent Sky Search and Replacement

Rawat, Saumya; Gairola, Siddhartha; Shah, Rajvi; Narayanan, P. J.

doi:10.1007/978-3-319-73603-7_18

Cited by 5 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To further enhance cloud classification accuracy, we employ semisupervised learning to leverage a large set of 9,883 unlabeled cloud photos from the SkyFinder dataset [32] and FindMeASky dataset [33]. We also apply data augmentation following the schemes of FixMatch [17].…”

Section: Semi-supervised Learningmentioning

confidence: 99%

“…The CCSN dataset contains 2,543 cloud images, in which cloud photographs were labeled into 10 cloud categories, thus we formulated cloud-type classification as a 10-class problem. For semi-supervised learning, we leveraged the SkyFinder [32] and FindMeASky [33] datasets, which came with the sky segmentation masks but no cloud-type labels. After eliminating duplicate images, our unlabeled dataset comprised 9,883 photos.…”

Section: Painting and Photo Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

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

Zhang,

Mansfield,

et al. 2024

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

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.

show abstract

Section: Semi-supervised Learningmentioning

confidence: 99%

Section: Painting and Photo Datasetsmentioning

confidence: 99%

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

Zhang,

Mansfield,

et al. 2024

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

show abstract

“…To further enhance cloud classification accuracy, we employ semisupervised learning to leverage a large set of unlabeled cloud photos. We selected 9883 images from SkyFinder dataset [49] and FindMeASky dataset [50] as unlabeled data. Specifically, we follow the approach of FixMatch [51].…”

Section: Semi-supervised Learningmentioning

confidence: 99%

“…To evaluate the accuracy of sky segmentation and the classification performance on the sky photos, we adopt the CCSN dataset for training and testing. To realize semi-supervised learning, we leverage the SkyFinder [49] and FindMeASky [50] datasets to boost the classification performance. The SkyFinder dataset contains over 90,000 outdoor sky photos in different weather situations with associated detailed weather data and annotated sky pixels.…”

Section: Cloud Photo Datasetmentioning

confidence: 99%

A Machine Learning Paradigm for Studying Pictorial Realism: Are Constable's Clouds More Real than His Contemporaries?

Zhang¹,

Mansfield²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

European artists have sought to create life-like images since the Renaissance. The techniques used by artists to impart realism to their paintings often rely on approaches based in mathematics, like linear perspective; yet the means used to assess the verisimilitude of realist paintings have remained subjective, even intuitive. An exploration of alternative and relatively objective methods for evaluating pictorial realism could enhance existing art historical research. We propose a machine-learning-based paradigm for studying pictorial realism in an explainable way. Unlike subjective evaluations made by art historians or computer-based painting analysis exploiting inexplicable learned features, our framework assesses realism by measuring the similarity between clouds painted by exceptionally skillful 19th-century landscape painters like John 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. Our analyses suggest that artists working in the decades leading up to the invention of photography worked in a mode that anticipated some of the stylistic features of photography. The study is a springboard for deeper analyses of pictorial realism using computer vision and machine learning.

show abstract

An experimental evaluation of visual similarity for HDR images

Aydinlilar

Akyüz

2021

Multimed Tools Appl

View full text Add to dashboard Cite

Find Me a Sky: A Data-Driven Method for Color-Consistent Sky Search and Replacement

Cited by 5 publications

References 16 publications

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

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

A Machine Learning Paradigm for Studying Pictorial Realism: Are Constable's Clouds More Real than His Contemporaries?

An experimental evaluation of visual similarity for HDR images

Contact Info

Product

Resources

About