[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-134104-105":59,"doc-detail-134104-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","robust-multi-label-cartoon-character-classification-on-the-novel-kral-sakir-dataset-using-deep-learning-techniques","Robust Multi-Label Cartoon Character Classification on the Novel Kral Sakir Dataset Using Deep Learning Techniques","","Automated cartoon character recognition is essential for content indexing, filtering, and copyright protection, but animated media introduces major difficulties due to high intra-class visual variability, where characters frequently change their appearance. To address this challenge, the paper presents the novel Kral Sakir dataset, a public benchmark with 16,725 curated images for multi-label cartoon character classification under these varied conditions. A comprehensive benchmark evaluates state-of-the-art pretrained CNNs (DenseNet, ResNet, VGG) against a scratch-trained baseline using F1-Score, accuracy, and AUC, showing that fine-tuning pretrained models is highly effective, with DenseNet121 achieving F1-Score 0.9890 and accuracy 0.9898.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/robust-multi-label-cartoon-character-classification-on-the-novel-kral-sakir-dataset-using-deep-learning-techniques/134104/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/robust-multi-label-cartoon-character-classification-on-the-novel-kral-sakir-dataset-using-deep-learning-techniques/134104.png","ImageObject",300,407,{"name":92,"@type":93},"Oliver","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-09-19","2026-08-20",true,{"@type":102,"interactionType":103,"userInteractionCount":52},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What challenge does the paper target in cartoon character recognition?","Question",{"text":112,"@type":113},"The paper targets high intra-class visual variability in animated media, where characters often change appearance due to artistic license and narrative progression.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What is the Kral Sakir dataset and how many images does it include?",{"text":117,"@type":113},"The Kral Sakir dataset is a public benchmark curated for multi-label cartoon character classification, containing 16,725 images.",{"name":119,"@type":110,"acceptedAnswer":120},"Which models and evaluation metrics are used in the benchmark study?",{"text":121,"@type":113},"The study evaluates pretrained CNNs such as DenseNet, ResNet, and VGG, compared with a scratch baseline, using F1-Score, accuracy, and ROC AUC.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},134104,1787234004,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":52,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":26},8796095461610,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Tech Science Press  \nDoi:10.32604/cmc.2025.067840  \nARTICLE  \nRobust Multi-Label Cartoon Character Classification on the Novel Kral Sakir Dataset Using Deep Learning Techniques  \nCandan Tumer1, Erdal Guvenoglu2 and Volkan Tunali3, *  \n1 Graduate School, Maltepe University, Istanbul, 34857, Turkiye  \n2 Department of Computer Programming, Vocational School, Maltepe University, Istanbul, 34857, Turkiye  \n3 Division of Computing, School of Computing, Engineering and Physical Sciences, University of the West of Scotland, London Campus, London, E14 2BE, UK  \n*[Corresponding Author: Volkan Tunali. Email: volkan.tunali@uws.ac.uk](Corresponding Author: Volkan Tunali. Email: volkan.tunali@uws.ac.uk)[ ](Corresponding Author: Volkan Tunali. Email: volkan.tunali@uws.ac.uk)Received: 14 May 2025; Accepted: 19 August 2025; Published: 23 October 2025  \nABSTRACT: Automated cartoon character recognition is crucial for applications in content indexing, filtering, and copyright protection, yet it faces a significant challenge in animated media due to high intra-class visual variability, where characters frequently alter their appearance. To address this problem, we introduce the novel Kral Sakir dataset, a public benchmark of 16,725 images specifically curated for the task of multi-label cartoon character classification under these varied conditions. This paper conducts a comprehensive benchmark study, evaluating the performance of state-of-the-art pretrained Convolutional Neural Networks (CNNs), including DenseNet, ResNet, and VGG, against a custom baseline model trained from scratch. Our experiments, evaluated using metrics of F1-Score, accuracy, and Area Under the ROC Curve (AUC), demonstrate that fine-tuning pretrained models is a highly effective strategy. The best-performing model, DenseNet121, achieved an F1-Score of 0.9890 and an accuracy of 0.9898, significantly outperforming our baseline CNN (F1-Score of 0.9545) . The findings validate the power of transfer learning for this domain and establish a strong performance benchmark. The introduced dataset provides a valuable resource for future research into developing robust and accurate character recognition systems.  \nKEYWORDS: Cartoon character recognition; multi-label classification; deep learning; transfer learning; predictive modelling; artificial intelligence-enhanced (AI-Enhanced) systems; Kral Sakir dataset  \n1 Introduction  \nAnimated content, including cartoons, constitutes a significant and globally popular form of entertainment and communication. The ability to automatically analyse this content, particularly to identify and track characters, offers substantial benefits across various applications, from content indexing and retrieval to copyright protection and interactive media experiences [1] . However, automated cartoon character recognition presents unique and significant challenges that differentiate it from general object recognition tasks.  \nA key challenge in the application of AI to image recognition, and a central focus of our research, stems from the substantial visual variability inherent in domains like cartoon character portrayal. Unlike the more constrained variations typically observed in real-world object classes, cartoon characters often undergo transformations driven by artistic license and narrative progression. These can manifest as considerable changes in attire, colour schemes, accessories, depicted age, and even fundamental drawing style or  \nCopyright © 2025 The Authors. Published by Tech Science Press.  \nThis work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.  \nperspective [2] . This high degree of intra-class variance presents a notable difficulty for current AI models, particularly deep learning architectures that learn features directly from pixel data, as consistent low-level visual patterns may beco","cbCaigmZCfI9bBu2","https://ap.wps.com/l/cbCaigmZCfI9bBu2","pdf",2211498,24,"English","# Introduction\n## Problem: intra-class visual variability in cartoons\n## Proposed solution: Kral Sakir dataset and transfer learning approach","[{\"question\":\"What challenge does the paper target in cartoon character recognition?\",\"answer\":\"The paper targets high intra-class visual variability in animated media, where characters often change appearance due to artistic license and narrative progression.\"},{\"question\":\"What is the Kral Sakir dataset and how many images does it include?\",\"answer\":\"The Kral Sakir dataset is a public benchmark curated for multi-label cartoon character classification, containing 16,725 images.\"},{\"question\":\"Which models and evaluation metrics are used in the benchmark study?\",\"answer\":\"The study evaluates pretrained CNNs such as DenseNet, ResNet, and VGG, compared with a scratch baseline, using F1-Score, accuracy, and ROC AUC.\"}]","Robust Multi-Label Cartoon Character Classification on the Novel Kral Sakir Dataset Using Deep Learning Techniques | PDF"]