Hubbry Logo
WikiArtWikiArtMain
Open search
WikiArt
Community hub
WikiArt
logo
8 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
WikiArt
WikiArt
WikiArt
View on Wikipedia
from Wikipedia
Home page of WikiArt

WikiArt (formerly known as WikiPaintings) is a visual art wiki, active since 2010.

The developers are based in Ukraine.[1] Since 2010, the Editor in Chief of WikiArt has been the Ukrainian art critic Kseniia Bilash.[2]

In April 2022, access to WikiArt was restricted in Russia, by request of the General Prosecutor’s office, according to Roskomsvoboda.[3]

AI research

[edit]

WikiArt is often used by scientists who study AI. They train AI on WikiArt data trying to discover its ability to recognize, classify, and generate art.

In 2015, computer scientists Babak Saleh and Ahmed Egammal of Rutgers University used images from WikiArt in training an algorithm to look at paintings and detect the works’ genre, style and artist.[4] Later, researchers from Rutgers University, the College of Charleston and Facebook's AI Lab collaborated on a generative adversarial network (GAN), training it on WikiArt data to tell the difference between a piece of art versus a photograph or diagram, and to identify different styles of art.[5] Then, they designed a creative adversarial network (CAN), also trained on WikiArt dataset, to generate new works that does not fit known artistic styles.[6]

In 2016, Chee Seng Chan (Associate Professor at University of Malaya) and his co-researchers trained a convolutional neural network (CNN) on WikiArt datasets and presented their paper "Ceci n’est pas une pipe: A Deep Convolutional Network for Fine-art Paintings Classification".[7] They released ArtGAN to explore the possibilities of AI in its relation to art. In 2017, a new study and improved ArtGAN was published: "Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork".[8]

In 2018, an Edmond de Belamy portrait produced by a GAN was sold for $432,500 at a Christie's auction. The algorithm was trained on a set of 15,000 portraits from WikiArt, spanning the 14th to the 19th century.[9]

In 2019, Eva Cetinic, a researcher at the Rudjer Boskovic Institute in Croatia, and her colleagues, used images from WikiArt in training machine-learning algorithms to explore the relationship between the aesthetics, sentimental value, and memorability of fine art.[10]

In 2020, Panos Achlioptas, a researcher at Stanford University and his co-researchers collected 439,121 affective annotations involving emotional reactions and written explanations of those, for 81 thousand artworks of WikiArt. Their study involved 6,377 human annotators and it resulted in the first neural-based speaker model that showed non-trivial Turing test performance in emotion-explanation tasks.[11]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

WikiArt

View on Grokipedia
from Grokipedia
WikiArt is an online visual art launched in 2010, offering a searchable database of over 250,000 digitized artworks by more than 5,000 artists, accompanied by biographies, classifications by movements, styles, and genres. Developed by a Ukraine-based team with Kseniia Bilash as since , it functions as a collaborative wiki enabling contributions to art-related content while prioritizing accessibility through multilingual support in eight languages. The platform's core mission emphasizes democratizing global resources, making discoverable via intuitive search tools and organizational schemas that span from ancient to contemporary periods. Key features include high-resolution images suitable for educational and purposes, with artworks often displayed in institutional settings like museums and universities, and its datasets powering advancements in computational art analysis, such as style vector representations for applications. Unlike traditional encyclopedias, WikiArt integrates user contributions under editorial oversight to maintain content quality, though it has faced minor user-reported issues regarding completeness of specific works. Its growth reflects a commitment to empirical cataloging of , avoiding narrative-driven curation in favor of comprehensive, data-rich aggregation that supports first-principles exploration of artistic evolution. No major controversies have marred its operations, distinguishing it from more contentious digital archives prone to ideological disputes.

History

Founding and initial launch

WikiArt originated as WikiPaintings, an online visual art encyclopedia launched in 2010 by developers based in . The platform was established with the primary objective of democratizing access to by providing a searchable, editable database of artworks, biographies, and related metadata, drawing initially from images and licensed reproductions of paintings from museums and collections. Ukrainian art critic Kseniia Bilash has served as since the site's inception, overseeing content curation and editorial standards to ensure comprehensive coverage while enabling community-driven contributions in a wiki-style format. At launch, WikiPaintings emphasized paintings as its core focus, reflecting the name and aiming to fill gaps in traditional resources by aggregating digitized reproductions alongside user-submitted on lesser-known works and artists. This approach positioned the site as an open alternative to gatekept institutional databases, prioritizing broad over controls and incorporating tools for multilingual support from early stages to reach global audiences. The initial aggregation relied on existing encyclopedic sources for seed data, but the wiki model encouraged expansions through verified user edits, fostering a collaborative for documentation.

Expansion and rebranding

Originally launched as WikiPaintings with a focus on paintings, the platform rebranded to WikiArt to accommodate an expanded scope encompassing photographs, sculptures, and other visual art forms beyond traditional paintings. This shift broadened content categorization to include diverse media, enabling comprehensive coverage of visual arts history and styles. By the mid-2010s, the database had scaled to over 100,000 artworks, reflecting rapid growth driven by community contributions and curatorial additions; this expanded to approximately 250,000 artworks from around 5,000 artists by the early 2020s. Multilingual localization was implemented to support up to 8 languages, facilitating global user access and navigation across cultural contexts. Integration of metadata for genres, styles, and artistic movements further enhanced content organization, adapting to demands for structured exploration without relying on advanced search interfaces. To address scalability challenges from rising user traffic, WikiArt collaborated with Implex, a development firm specializing in high-load systems, which re-engineered the backend for improved performance, including a 30% increase in and enhanced handling of large-scale image delivery. These technical upgrades ensured reliability amid growing demands, prioritizing content accessibility over commercial features.

Recent developments

In May 2017, WikiArt launched its mobile application, allowing users to browse the art database on smartphones with capabilities for offline access to selected high-resolution images and customized feeds based on user preferences for genres, periods, or artists. An Android counterpart became available around the same period, further broadening mobile engagement by integrating similar search and discovery tools adapted for touch interfaces. These apps emphasized portability for art education and casual exploration, syncing with the web platform's metadata for seamless cross-device continuity. By 2025, WikiArt's database had grown to encompass over 250,000 artworks by more than 3,000 artists across over 100 countries, maintaining a chronological scope from prehistoric cave paintings to cutting-edge contemporary pieces while prioritizing high-fidelity reproductions and detailed data. This expansion included regular additions of newly digitized or crowdsourced entries, with recent incorporations such as works by artists like Jacob Mącznik and Mykhailo Vainshtein, reflecting ongoing curation efforts to enhance global representation. WikiArt, operated by a Ukraine-based development team, has sustained full operational continuity since the escalation of the in 2022, with no reported service disruptions despite regional geopolitical challenges; the platform continued publishing art-related and updates, such as analyses of thematic paintings, as late as October 2025.

Features and

Web platform capabilities

WikiArt's facilitates browsing of its database through categorized galleries emphasizing visual immersion, allowing users to navigate artworks by artist nationality, such as American or French painters, and by art movements including and Naturalism. Additional filters enable exploration by artistic styles and genres like , with each category displaying image-heavy grids of high-resolution reproductions to support direct visual assessment over textual summaries. Individual artist pages aggregate biographical details, including key life events and influences, alongside interactive timelines that sequence works chronologically for contextual understanding, and sections linking to similar or related pieces within the collection. These elements draw from crowdsourced contributions cross-verified against sources, with registered users able to propose edits for metadata or additional verified information, subject to moderation to maintain factual integrity. A dedicated "Popular Artworks" feature ranks entries empirically by user engagement metrics, such as the number of views over the preceding 30 days, generating dynamic lists of the top 600 paintings visited site-wide. Similarly, "Popular Artists" tallies the most accessed profiles, highlighting approximately 240 figures based on comparable view data, thereby surfacing content through observed user interest rather than predefined curatorial hierarchies. This approach leverages aggregate behavioral data to prioritize and relevance as determined by platform traffic patterns.

Mobile application

The WikiArt mobile application, developed by Little Marvel Software Inc., is available for purchase on the iOS for $4.99, with optional in-app purchases. It delivers portable access to the platform's database of over 250,000 artworks by more than 3,000 artists from over 100 countries, covering periods from prehistoric cave paintings to contemporary pieces. Designed for and compatible iOS devices, the app enables on-the-go art discovery through a mobile-optimized interface that mirrors core web functionalities while prioritizing touch-based navigation for browsing collections. Users can explore artworks categorized by , style, , period, or associated , with integrated biographical details and historical context for each entry to support educational and inspirational use. The app incorporates device-native sharing options for exporting images or details to other applications, facilitating quick dissemination in learning or social contexts. Multilingual support in seven languages, bolstered by community translations, extends its reach for global portability. Version 6.3.1, updated as of September 23, 2025, includes stability improvements to sustain seamless mobile access amid growing content demands. The application earns a 4.8 out of 5 rating from approximately 1,900 reviews, reflecting user appreciation for its role in extending access beyond desktop constraints. Android availability remains limited to unofficial APK distributions, lacking the verified integration and updates seen in the iOS counterpart.

Search and discovery tools

WikiArt's search functionality supports keyword-based queries for artwork titles, artist names, and descriptive terms, enabling users to locate specific pieces within its database of over 2 million images spanning various periods and mediums. This query-driven approach contrasts with categorical browsing by prioritizing user-initiated exploration, such as entering "" to retrieve Vincent van Gogh's 1889 alongside related metadata on its Post-Impressionist context. Advanced search tools extend discovery through filters including year of creation (facilitating era-specific selections, e.g., artworks from the ), genres, styles, art movements, nationalities, and fields like or . Users can refine results by these parameters—for instance, isolating works by Italian artists—to conduct systematic analyses grounded in empirical classifications rather than subjective curation. No algorithmic recommendation engine based on user history or view counts is implemented, avoiding potential biases in suggestion algorithms; instead, empirical popularity metrics highlight frequently accessed pieces, such as Claude Monet's Water Lilies series from 1919. Metadata integration enhances cross-referencing, with each artwork linked to artist biographies, movement affiliations, and chronological details, allowing users to trace influences like Naturalism's alignment with 19th-century scientific observation in artists such as . This structure supports causal inquiry into art historical developments without reliance on interpretive overlays, as filters aggregate data directly from verified entries rather than external narratives.

Content and database

Scope of artworks and artists

WikiArt's database includes over 250,000 high-resolution images of artworks spanning from prehistoric eras, such as representations, to 21st-century contemporary pieces. These images prioritize verifiable works housed in museums, universities, and public collections, ensuring coverage of historically documented art forms across , , and other media. The platform catalogs contributions from more than 5,000 artists representing diverse nationalities, with dedicated listings for figures from regions including Europe, Asia, Africa, and the Americas, as well as underrepresented cultural groups. This breadth extends to over 3,000 artists in subsets analyzed for stylistic research, reflecting a focus on empirical inclusion of influential creators rather than exhaustive universality. Gender representation features a distinct category for female artists across periods, though comprehensive distributional metrics remain limited in public documentation, underscoring the male-dominated historical art canon while incorporating verified women contributors. Exclusion criteria emphasize historically significant or empirically popular works, omitting purely commercial productions or those lacking institutional verification to sustain scholarly . This selective scope differentiates WikiArt from broader image repositories by anchoring content in established art historical precedents, with ongoing additions favoring pieces from public-domain or licensed institutional sources.

Categorization and metadata

WikiArt employs a hierarchical categorization system that organizes artworks primarily by art movements, styles, and genres, reflecting established art historical frameworks rather than subjective modern reinterpretations. Art movements are structured chronologically and culturally, spanning from medieval periods like Gothic to contemporary ones such as , with Western art emphasizing aesthetic and cultural shifts (e.g., , ) and Eastern art divided into historical periods. Styles denote specific visual techniques and elements associated with movements or schools, such as Realism or , while genres classify content by thematic depictions, following a classical originating in 17th-century European theory that prioritizes historical, religious, and portrait subjects over or . Each artwork entry includes structured metadata fields, including creation dates, genres, styles, and locations (often denoting or current holding ), which facilitate chronological and geographical grounded in verifiable . Additional contextual details, such as medium and artist nationality, support causal inquiries into influences like or regional traditions, though dimensions are inconsistently recorded and not universally standardized across entries. This metadata enables users to query works by empirical attributes, such as timeframe or thematic content, promoting analyses aligned with historical evidence over interpretive bias. Categorization and metadata undergo periodic revisions based on scholarly research and consensus among contributors, exemplified by reclassifications of attribution or stylistic alignment when new empirical emerges, ensuring alignment with art historical verifiability rather than fixed narratives. Such updates prioritize factual corrections, as seen in adjustments to movement tags for ambiguous transitional works, maintaining the database's utility for rigorous, evidence-based study.

Editing and community contributions

WikiArt permits registered users to edit artist biographies and associated metadata, including corrections to factual inaccuracies, additions to descriptive content, and suggestions for image replacements or identifications of unnamed works. Users may also contribute translations of site content and report errors such as grammatical issues or low-quality images, facilitating iterative improvements to the encyclopedia's entries. These participatory features distinguish WikiArt from purely static databases, enabling crowdsourced enhancements while relying on user-submitted inputs for expansion. To mitigate risks of or erroneous additions, reported issues and proposed edits are reviewed by the site's editorial staff, ensuring that contributions align with the platform's goal of accurate representation. The core development and oversight team, operational since the site's , maintains administrative control over content integrity, with edits processed through designated channels rather than open, real-time wiki-style revisions. This moderated approach balances community input against potential inaccuracies, prioritizing reliability in a domain prone to subjective interpretations of artistic history. Although WikiArt lacks explicit public guidelines mandating citations for biographical or tagging edits, the emphasis on correcting verifiable inaccuracies implies a preference for empirically grounded submissions over , as unverified inputs risk rejection during editorial review. This framework has supported gradual database growth through volunteer-like contributions, though without formalized roles such as dedicated curators, expansions remain under centralized Ukrainian-based administration.

Technical implementation

Architecture and scalability

WikiArt's backend architecture, developed and maintained by the software studio Implex since 2016, utilizes C# for the core engine and as the database to support high-load operations. This stack facilitates efficient data management for over 250,000 artworks across 3,000 artists, enabling scalability to serve global users in eight languages and over 100 countries. Re-engineering efforts by Implex, conducted in two phases, optimized the platform for intensive traffic, resulting in a 30% increase in engine performance and the ability to handle peak loads from sources such as Reddit-driven surges and persistent bot activity. These improvements supported a doubling of monthly visits from 1.1 million to 2.3 million, alongside approximately 10 million monthly page views, without reported downtime. In the second development phase following changes in ownership, enhancements included AI-assisted tagging of more than 120,000 artworks using tools and SEO optimizations that elevated visibility on search engines, further demonstrating resource allocation aligned with demand patterns during educational and viral usage spikes. The architecture prioritizes backend efficiency to sustain ad-supported hosting costs amid growing query volumes for art-related searches.

Data sourcing and licensing

WikiArt aggregates images of artworks primarily from public domain sources, user contributions, and openly accessible online collections, including reproductions from museums and galleries worldwide. The platform hosts approximately 250,000 artworks by over 3,000 artists, with many images derived from historical works where copyrights have expired, though photographic reproductions may retain separate protections under applicable laws. Licensing policies prioritize compliance with copyright notices, requiring the removal or of disputed images upon verified infringement claims, as outlined in the site's terms of use. This approach addresses provenance challenges by relying on empirical verification of ownership and public availability rather than assuming perpetual , while avoiding inclusion of privately held copyrighted materials without permission. Derived datasets, such as the WikiArt collection mirrored on platforms like , explicitly restrict usage to non-commercial research, with original authors disclaiming liability for image copyrights to encourage ethical data handling. To maintain integrity against institutional tendencies to hoard digital reproductions, WikiArt promotes for educational and analytical purposes, updating its database periodically to excise non-compliant entries based on legal challenges or updated . This framework counters barriers posed by APIs and restricted archives, fostering broader empirical access to historical records without endorsing unlicensed scraping of protected content.

Impact and applications

Usage statistics and accessibility

WikiArt receives up to 2.5 million monthly visits and approximately 10 million page views, reflecting substantial global engagement with its visual art database. These figures indicate peaks associated with both educational inquiries, such as searches for specific artists or movements, and casual browsing of high-resolution images and biographies. Publicly available site analytics underscore this usage pattern, with traffic driven primarily by organic searches for artworks rather than institutional referrals. The platform ensures broad accessibility through its free, web-based interface available at wikiart.org, requiring no subscription or for core features like viewing, profiles, and search functionalities. Localization in eight languages supports users worldwide, extending reach beyond English-speaking audiences and facilitating ungatekept access to art historical content. While specific low-bandwidth optimizations are not explicitly documented, the site's responsive design accommodates diverse devices and connection speeds, aligning with its mission to democratize art knowledge for non-elite demographics including casual enthusiasts and self-learners. User logs and interaction studies reveal a diverse , encompassing hobbyists, students, and general users rather than solely art professionals or academics, evidenced by exploratory analyses of browsing behaviors on the platform. This broad appeal is further supported by the encyclopedia's editable nature, which invites contributions from varied global participants, though traffic demographics remain aggregated without detailed public breakdowns beyond language-based localization metrics.

Role in art research and education

WikiArt provides researchers with structured metadata on over 100,000 artworks, enabling quantitative analyses of artistic styles and movements through filters for genres, techniques, and historical periods. Scholars have leveraged this data to create vector-based representations of style and color distributions, facilitating empirical comparisons across eras and influences such as material innovations in pigment technology during the . The database's chronological artist timelines, spanning from antiquity to contemporary works, support causal inquiries into art historical developments, including how events like the correlated with shifts toward Realism and by aggregating dated entries without interpretive overlay. In pedagogical contexts, WikiArt functions as an accessible repository of public-domain and high-resolution images, recommended by university libraries for instruction due to its non-commercial research licensing and searchable attributes that aid in curating lesson-specific selections. Educators utilize its categorization by nationality, century, and style to develop curricula tracing verifiable lineages, such as the evolution from Academicism to , bypassing reliance on potentially selective institutional archives. Annotated subsets, like the WikiArt Emotions dataset encompassing 4,000 paintings labeled for evoked sentiments, are distributed for classroom explorations of psychological responses to visual elements, promoting hands-on analysis over rote memorization. The platform's exportable metadata, including biographies and artwork dates verified through curation, underpins academic publications by supplying baseline datasets for testing, as evidenced in studies on painter attribution via stylistic markers. This utility extends to interdisciplinary , where filters enable unbiased sequencing of movements—prioritizing chronological over narrative-driven histories—to examine causal factors like systems' impact on output volumes in Baroque-era .

Utilization in machine learning and AI

WikiArt serves as a primary data source for datasets focused on artistic image analysis and generation, with collections such as the WikiArt dataset comprising over 80,000 images scraped from the site and restricted explicitly to non-commercial research applications. These datasets have enabled training of generative adversarial networks (GANs) for tasks including style transfer and synthetic artwork creation, as demonstrated in frameworks like ArtGAN, which leverage categorical labels from WikiArt metadata to condition outputs on specific artistic movements or genres, yielding visually coherent imitations of historical styles. Similarly, conditional GAN variants have utilized WikiArt subsets for emotion-guided art generation, mapping textual or labeled inputs to stylistic outputs derived from the dataset's diverse tagged images. Empirical advancements include feature extraction from WikiArt's annotated images to derive quantitative representations, such as in the 2022 WikiArtVectors initiative, which processed 81,000+ artworks into style embeddings and color histograms using information-theoretic metrics like Jensen-Shannon divergence for pairwise comparisons. This vector dataset facilitates models for computational , enabling tasks like clustering artworks by stylistic proximity or tracking diachronic shifts in color usage across eras, thereby supporting scalable analyses unattainable through manual curation. Such applications have empirically improved model performance in style classification, with convolutional neural networks achieving accuracies exceeding 70% on WikiArt-derived benchmarks for multi-class prediction. Debates persist regarding potential biases in AI outcomes from WikiArt-sourced data, which empirically overrepresents Western European canons—comprising roughly 80% of entries from 500 BCE to present—potentially embedding Eurocentric stylistic priors into trained models despite inclusions of non-Western artists. This skew has been quantified in analyses showing GAN-generated outputs favoring canonical motifs over underrepresented global traditions, though proponents argue that WikiArt's ongoing expansions mitigate such effects by broadening metadata tags for comparisons. Empirical tests on debiased subsets confirm that while full-dataset training amplifies Western biases in feature vectors, selective filtering can yield more balanced representations without substantial loss in model fidelity.

Reception and criticisms

Positive assessments

WikiArt has been praised for its extensive collection, encompassing over 250,000 artworks by more than 5,000 artists spanning various periods and styles, which surpasses the scope of many traditional references by offering searchable, high-resolution images alongside biographical and contextual details. This comprehensiveness enables users to explore movements, genres, and individual works with advanced filters for attributes like color, medium, and date, facilitating deeper engagement than static encyclopedias. The platform's open-editing model, moderated for accuracy, aligns with wiki principles to foster collaborative knowledge-building, earning recognition as one of the 15 best free reference websites by the for democratizing access to visual art resources. Localized in eight languages and optimized for global reach, it supports diverse users, including those in non-English-speaking regions, and has been integrated into public displays in museums, universities, and town halls across over 100 countries. In educational contexts, WikiArt serves as a valuable tool for instruction, with its intuitive interface and vast database making complex topics approachable for students and educators worldwide, as evidenced by its adoption in classrooms to enhance and research. Academic studies frequently leverage its datasets for analyses in style recognition and emotion annotation, underscoring its utility in advancing art-related . Usage metrics reflect this impact, with monthly visits doubling to 2.3 million following technical enhancements that improved performance by 30 percent, indicating sustained user value and scalability.

Accuracy and reliability concerns

As a crowdsourced platform allowing registered users to edit artist pages, add artworks, and correct information, WikiArt relies on community contributions for content updates, which are subject to moderation by a volunteer team to maintain quality. This process enables rapid corrections of factual errors, such as misattributions or inaccurate metadata, through user reports of inaccuracies, but the absence of mandatory citations or expert verification for submissions can permit unverified edits to enter the database temporarily. Critiques highlight that WikiArt's crowd-sourced metadata, despite moderation, lacks transparency in verification methods, potentially resulting in unreliable information and categorization errors, such as conflating self-taught naïve artists with trained painters imitating primitivist styles under a single "Naïve Art (Primitivism)" label, which prioritizes contextual assumptions over consistent visual or historical criteria. Unlike static, expert-curated databases that minimize changes post-validation, WikiArt's dynamic editing fosters ongoing dynamism—facilitating updates to reflect new scholarship—but introduces risks of empirical inaccuracies from non-specialist inputs, with reliability varying relative to professionally vetted collections. Instances of , such as deliberate false edits or spam, remain possible in open-editable systems like WikiArt, though the site's consistent pre-publication moderation of updates aims to detect and revert such disruptions, similar to mechanisms in comparable wiki platforms. Users are encouraged to report detected errors, including misattributions, enabling community-driven rectification, though the effectiveness depends on volunteer oversight rather than automated or expert-only controls. WikiArt's presentation of copyrighted artworks relies on the fair use doctrine under copyright law, which permits limited reproduction for purposes such as criticism, comment, news reporting, teaching, , or . The site's terms of use explicitly state that copyrighted images are displayed in accordance with this principle, typically as thumbnails or low-resolution previews to facilitate educational and informational access without substituting for the original works. No major legal challenges from museums or rights holders against WikiArt for image scraping or display have been documented, though the platform implements measures to deter unauthorized bulk scraping, such as IP restrictions associated with public cloud providers. Museums have occasionally asserted rights over digitized works or imposed usage guidelines, but these claims often lack enforceability under for reproductions not claiming new authorship, allowing platforms like WikiArt to argue transformative, non-commercial use. The utilization of WikiArt's in training generative AI models has elicited ethical concerns regarding and attribution, as scraped images enable algorithms to learn and replicate artistic styles, potentially leading to outputs that mimic specific artists without compensation or credit. Critics contend this undermines artists' and , viewing unrestricted data access as diluting incentives for original creation, particularly when models generate works indistinguishable from human efforts. Proponents of , however, maintain that such training constitutes , akin to transformative technologies like search engines or , fostering in art analysis and without direct infringement, as the process extracts abstract features rather than copying exact images. Empirical analyses of AI outputs trained on datasets including WikiArt suggest limited verbatim reproduction risks, supporting arguments for permissive licensing to accelerate causal advancements in computational over stringent restrictions that could stifle .

References

  1. https://implex.dev/case-study/wikiart
  2. https://www.wikiart.org/en/about
  3. https://pmc.ncbi.nlm.nih.gov/articles/PMC9497612/
  4. https://www.wikiart.org/en/How-to-Contribute
  5. https://scholarworks.aub.edu.lb/bitstream/handle/10938/24097/AntounMaya_2023.pdf?sequence=1
  6. https://ua.linkedin.com/in/kseniia-bilash-426a7451
  7. https://library.oapen.org/bitstream/20.500.12657/95990/3/SID-0000004867631_optimized.pdf
  8. https://appadvice.com/app/wikiart/1235995167
  9. https://apkpure.com/wikiart/com.wikitude.nativesdksampleapp
  10. https://apps.apple.com/ua/app/wikiart/id1235995167
  11. https://www.wikiart.org/en/recently-added-artists
  12. https://www.wikiart.org/news/
  13. https://www.wikiart.org/
  14. https://www.wikiart.org/en
  15. https://www.wikiart.org/en/artists-by-art-movement
  16. https://www.wikiart.org/en/mary-cassatt
  17. https://www.wikiart.org/en/popular-paintings/alltime
  18. https://www.wikiart.org/en/popular-artists
  19. https://www.wikiart.org/en/popular-artists/text-list
  20. https://apps.apple.com/us/app/wikiart/id1235995167
  21. https://download.cnet.com/wikiart/3000-20414_4-78339135.html
  22. https://apkpure.com/wikiart/wikiart.appkao
  23. https://www.wikiart.org/en/paintingadvancedsearch
  24. https://wikiart.updatestar.com/en
  25. https://www.wikiart.org/en/artists-by-century
  26. https://www.wikiart.org/en/artists-by-nation
  27. https://www.wikiart.org/en/female-artists
  28. https://www.wikiart.org/en/paintings-by-style
  29. https://www.wikiart.org/en/paintings-by-genre
  30. https://norma.ncirl.ie/4438/1/mrinmoyduttachoudhury.pdf
  31. https://guides.osu.edu/art/images
  32. https://www.wikiart.org/en/terms-of-use
  33. https://www.kaggle.com/datasets/steubk/wikiart
  34. https://www.wikiart.org/store/about-us
  35. http://eecs.qmul.ac.uk/~gyorgyf/files/papers/marengo2017hci.pdf
  36. https://www.wikiart.org/en/privacy-policy
  37. https://www.mdpi.com/1099-4300/24/9/1175
  38. https://www.wikiart.org/en/chronological-artists
  39. https://www.svkir.com/papers/Mohammad-Kiritchenko-WikiArt-LREC-2018.pdf
  40. https://syncedreview.com/2017/03/26/artgan-artwork-synthesis-with-conditional-categorical-gans/
  41. https://towardsdatascience.com/emotion-based-art-generation-using-c-gan-de9abed6fa2f/
  42. https://santafe.edu/research/results/papers/31-wikiartvectors-style-and-color-representations-
  43. https://link.springer.com/article/10.1007/s10462-024-10727-0
  44. https://www.researchgate.net/publication/346196272_Dissecting_landscape_art_history_with_information_theory
  45. https://visarts.eu/images/22Papers/Wasielewski-Beyond-the-Universal-Art-Dataset.pdf
  46. https://tripscholars.com/best-art-history-apps/
  47. https://www.saashub.com/compare-wikiart-vs-art-genius
  48. https://www.techdirt.com/2014/05/22/metropolitan-museum-art-claims-copyright-over-massive-trove-public-domain-works-it-has-released/
  49. https://arxiv.org/html/2404.08030v1
  50. https://medium.com/voicehq/the-ethical-dilemma-of-ai-art-e0915103745a
  51. https://www.arl.org/blog/training-generative-ai-models-on-copyrighted-works-is-fair-use/
  52. https://arxiv.org/html/2404.13964v1
Add your contribution
Related Hubs
User Avatar
No comments yet.