fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings

Abstract The notion of visual similarity is essential for computer vision, and in applications and studies revolving around vector embeddings of images. However, the scarcity of benchmark datasets poses a significant hurdle in exploring how these models perceive similarity. Here we introduce Style A...

Full description

Saved in:
Bibliographic Details
Main Authors: Tillmann Ohm, Andres Karjus, Mikhail V. Tamm, Maximilian Schich
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Data
Online Access:https://doi.org/10.1038/s41597-025-04529-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850067048891154432
author Tillmann Ohm
Andres Karjus
Mikhail V. Tamm
Maximilian Schich
author_facet Tillmann Ohm
Andres Karjus
Mikhail V. Tamm
Maximilian Schich
author_sort Tillmann Ohm
collection DOAJ
description Abstract The notion of visual similarity is essential for computer vision, and in applications and studies revolving around vector embeddings of images. However, the scarcity of benchmark datasets poses a significant hurdle in exploring how these models perceive similarity. Here we introduce Style Aligned Artwork Datasets (SALAD), and an example of fruit-SALAD with 10,000 images of fruit depictions. This combined semantic category and style benchmark comprises 100 instances each of 10 easy-to-recognize fruit categories, across 10 easy distinguishable styles. Leveraging a systematic pipeline of generative image synthesis, this visually diverse yet balanced benchmark demonstrates salient differences in semantic category and style similarity weights across various computational models, including machine learning models, feature extraction algorithms, and complexity measures, as well as conceptual models for reference. This meticulously designed dataset offers a controlled and balanced platform for the comparative analysis of similarity perception. The SALAD framework allows the comparison of how these models perform semantic category and style recognition task to go beyond the level of anecdotal knowledge, making it robustly quantifiable and qualitatively interpretable.
format Article
id doaj-art-64b5c75f48234193b3d0dcc3ae2bce99
institution DOAJ
issn 2052-4463
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Data
spelling doaj-art-64b5c75f48234193b3d0dcc3ae2bce992025-08-20T02:48:30ZengNature PortfolioScientific Data2052-44632025-02-0112111010.1038/s41597-025-04529-4fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddingsTillmann Ohm0Andres Karjus1Mikhail V. Tamm2Maximilian Schich3Tallinn University, School of Digital TechnologiesEstonian Business SchoolTallinn University, School of Digital TechnologiesTallinn University, ERA Chair of Cultural Data AnalyticsAbstract The notion of visual similarity is essential for computer vision, and in applications and studies revolving around vector embeddings of images. However, the scarcity of benchmark datasets poses a significant hurdle in exploring how these models perceive similarity. Here we introduce Style Aligned Artwork Datasets (SALAD), and an example of fruit-SALAD with 10,000 images of fruit depictions. This combined semantic category and style benchmark comprises 100 instances each of 10 easy-to-recognize fruit categories, across 10 easy distinguishable styles. Leveraging a systematic pipeline of generative image synthesis, this visually diverse yet balanced benchmark demonstrates salient differences in semantic category and style similarity weights across various computational models, including machine learning models, feature extraction algorithms, and complexity measures, as well as conceptual models for reference. This meticulously designed dataset offers a controlled and balanced platform for the comparative analysis of similarity perception. The SALAD framework allows the comparison of how these models perform semantic category and style recognition task to go beyond the level of anecdotal knowledge, making it robustly quantifiable and qualitatively interpretable.https://doi.org/10.1038/s41597-025-04529-4
spellingShingle Tillmann Ohm
Andres Karjus
Mikhail V. Tamm
Maximilian Schich
fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings
Scientific Data
title fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings
title_full fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings
title_fullStr fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings
title_full_unstemmed fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings
title_short fruit-SALAD: A Style Aligned Artwork Dataset to reveal similarity perception in image embeddings
title_sort fruit salad a style aligned artwork dataset to reveal similarity perception in image embeddings
url https://doi.org/10.1038/s41597-025-04529-4
work_keys_str_mv AT tillmannohm fruitsaladastylealignedartworkdatasettorevealsimilarityperceptioninimageembeddings
AT andreskarjus fruitsaladastylealignedartworkdatasettorevealsimilarityperceptioninimageembeddings
AT mikhailvtamm fruitsaladastylealignedartworkdatasettorevealsimilarityperceptioninimageembeddings
AT maximilianschich fruitsaladastylealignedartworkdatasettorevealsimilarityperceptioninimageembeddings