Product Visibility in Advertising: A Deep Learning-based Analysis of Food and Cosmetic Advertisements
| International Journal of Electrical and Electronics Engineering |
| © 2026 by SSRG - IJEEE Journal |
| Volume 13 Issue 3 |
| Year of Publication : 2026 |
| Authors : Madhura Joshi, Poojaa Krishnan, Utkarsha Savkare, Sheel Dongre, Shilpa Deshpande, Rashmi Apte, Mangesh Bedekar |
10.14445/23488379/IJEEE-V13I3P116
How to Cite?
Madhura Joshi, Poojaa Krishnan, Utkarsha Savkare, Sheel Dongre, Shilpa Deshpande, Rashmi Apte, Mangesh Bedekar, "Product Visibility in Advertising: A Deep Learning-based Analysis of Food and Cosmetic Advertisements," SSRG International Journal of Electrical and Electronics Engineering, vol. 13, no. 3, pp. 210-229, 2026. Crossref, https://doi.org/10.14445/23488379/IJEEE-V13I3P116
Abstract:
Advertising today relies a lot on visuals. When a product is shown clearly in the advertisement, it increases attention and helps people remember and purchase the brand. Product detection in advertisements is hence the crucial need of businesses in today’s world. Most existing systems mainly look for logos, advertisement sections, or general categories. They do not explore how clearly and how often the product appears throughout the whole video. Manual review of an advertisement video takes a lot of time, and it depends on personal judgment. Older computer vision techniques fail to detect products in real-world conditions like blur, objects blocking the view, changing brightness, and fast scene changes. With the intent of addressing the earlier-mentioned issues, this paper offers a system called a Product Visibility Analysis for Advertisements (ProVis-Ad) to automatically detect the product and measure its visibility in the real advertisements. The proposed work utilizes Deep Learning models, namely, You Only Look Once version 8 (YOLOv8), version 5 (YOLOv5), Faster Region-based Convolutional Neural Network (R-CNN), and Single Shot Multibox Detector (SSD300) to detect the product in an advertisement. The research includes eight datasets made from 155 real food and cosmetics advertisements, containing 10,810 labelled frames made using Roboflow and the Computer Vision Annotating Tool (CVAT). The ProVis-Ad system calculates product visibility for each video frame to identify the total duration of the presence of the product on screen in the advertisement. The results show a clear difference in the performance across models and advertisement domains. Food advertisements show more consistent product visibility than cosmetics advertisements. The system also introduces a new brand-level price measure to check if the products with a higher price get more on-screen attention. Experimental results demonstrate that the performance of YOLOv8 is superior to that of other models with respect to accuracy and F1-score for the detection of products in advertisements.
Keywords:
Advertisement analysis, Computer vision, Deep learning, Frame-level detection, Product visibility.
References:
[1] S. Shunmuga Krishnan, and Ramesh K. Sitaraman, “Understanding the Effectiveness of Video Ads: A Measurement Study,” IMC '13: Proceedings of the 2013 Conference on Internet Measurement Conference, Barcelona Spain, pp. 149-162, 2013.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Michele Bittencourt Rodrigues et al., “Revolutionising Food Advertising Monitoring: A Machine Learning-based Method for Automated Classification of Food Videos,” Public Health Nutrition, vol. 26, no. 12, pp. 2717-2727, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Nuri Paşa Özer, and Ali Erkam Yarar, “An Analysis of Cosmetic Advertisements on Instagram,” Journal of Civilization and Society, vol. 9, no. 1, pp. 40-56, 2025.
[Google Scholar] [Publisher Link]
[4] Steven C.H. Hoi et al., “Logo-Net: Large-Scale Deep Logo Detection and Brand Recognition with Deep Region-based Convolutional Networks,” arXiv Preprint, pp. 1-15, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Shaoqing Ren et al., “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” Advances in Neural Information Processing Systems (NeurIPS), vol. 28, pp. 1-9, 2015.
[Google Scholar] [Publisher Link]
[6] Rahma Dania, and Rosi Kumala Sari, “A Multimodal Analysis of Food Advertisement,” iNELTAL Conference Proceedings: The International English Language Teachers and Lecturers Conference, pp. 86-92, 2020.
[Google Scholar] [Publisher Link]
[7] Zaeem Hussain et al., “Automatic Understanding of Image and Video Advertisements,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, pp. 1100-1110, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Areeg Fahad Rasheed, and M. Zarkoosh, “Optimized YOLOv8 for Multi-Scale Object Detection,” Journal of Real-Time Image Processing, vol. 22, no. 1, pp. 1-14, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Hieu Duong-Trung, and Nghia Duong-Trung, “Integrating YOLOv8-agri and DeepSORT for Advanced Motion Detection in Agriculture and Fisheries,” EAI Endorsed Transactions: on Industrial Networks and Intelligent Systems, vol. 11, no. 1, pp. 1-11, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Wei Liu et al., “SSD: Single Shot MultiBox Detector,” Computer Vision - ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, vol. 9905, pp. 21-37, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Ángel Morera et al., “SSD vs. YOLO for Detection of Outdoor Urban Advertising Panels Under Multiple Variabilities,” Sensors, vol. 20, no. 16, pp. 1-23, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Hairong Li, and Nan Zhang, “Computer Vision Models for Image Analysis in Advertising Research,” Journal of Advertising, vol. 53, no. 5, pp. 771-790, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Patrick Jonsson, “A Deep Learning Approach to Advertisement Detection in Newspapers,” Master’s Thesis, KTH Royal Institute of Technology, Stockholm, Sweden, 2022.
[Google Scholar]
[14] Petar Ristoski et al., “A Machine Learning Approach for Product Matching and Categorization: Use Case: Enriching Product Ads with Semantic Structured Data,” Semantic Web: - Interoperability, Usability, Applicability, vol. 9, no. 5, pp. 707-728, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[15] James Hahn, and Adriana Kovashka, “Measuring Effectiveness of Video Advertisements,” arxiv Preprint, pp. 1-11, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Dongfang Li et al., “On Detection of Advertising Images,” 2007 IEEE International Conference on Multimedia and Expo, Beijing, China, pp. 1758-1761, 2007.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Hao Li, and Hui-Yi Lo, “Do you Recognize its Brand? The Effectiveness of Online In-Stream Video Advertisements,” Journal of Advertising, vol. 44, no. 3, pp. 208-218, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Edward Ratner, Schuyler Cullen, and James Quigley, “Object Recognition in Complex Video Scenes for Advertising Applications,” 2015 49th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, USA, pp. 1387-1392, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Oluwaseyi Ezekiel Olorunshola, Martins Ekata Irhebhude, and Abraham Eseoghene Evwiekpaefe, “A Comparative Study of YOLOv5 and YOLOv7 Object Detection Algorithms,” Journal of Computing and Social Informatics, vol. 2, no. 1, pp. 1-12, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Dillon Reis et al., “Real-Time Flying Object Detection with YOLOv8,” arXiv Preprint, pp. 1-10, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Zuzana Berger Haladova, Michal Zrubec, and Zuzana Cernekova, “A Method for Estimating Roadway Billboard Salience,” SAP '24: ACM Symposium on Applied Perception 2024, Dublin, Ireland, pp. 1-5, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Shervin Minaee et al., “Ad-Net: Audio-Visual Convolutional Neural Network for Advertisement Detection in Videos,” arXiv Preprint, pp. 1-5, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Jinqiao Wang et al., “ActiveAd: A Novel Framework of Linking ad Videos to Online Products,” Neurocomputing, vol. 185, pp. 82-92, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Faeze Zakaryapour Sayyad et al., “AdVision: An Efficient and Effective Deep Learning-based Advertisement Detector for Printed Media,” Machine Learning with Applications, vol. 21, pp. 1-12, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Haidi Zhu et al., “A Review of Video Object Detection: Datasets, Metrics and Methods,” Applied Sciences, vol. 10, no. 21, pp. 1-24, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Vivi Afifah, and Surni Erniwati, “Yolov8 for Object Detection: A Comprehensive Review of Advances, Techniques, and Applications,” IJACI: International Journal of Advanced Computing and Informatics, vol. 2, no. 1, pp. 53-61, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Haijun Zhang et al., “Object-Level Video Advertising: An Optimization Framework,” IEEE Transactions on Industrial Informatics, vol. 13, no. 2, pp. 520-531, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Tomohiko Takahashi et al., “Arbitrary Product Detection From Advertisement Video by using Object Independent Features,” 2011 IEEE International Conference on Multimedia and Expo, Barcelona, Spain, pp. 1-6, 2011.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Khinsa Fairuz Zahirah et al., “Evaluating the Effectiveness of Digital Product Advertisement Type using Machine Learning and Shapley Additive Explanations Analysis,” Journal of Information and Communication Technology (JICT), vol. 25, no. 1, pp. 79-106, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[30] Rafaela Cajado Magalhães de Alencar et al., “Impact of Visual Appeals and Brand Ambassador in Online Food Advertising on Consumer Purchase Behaviour,” International Journal of Social Economics, vol. 53, no. 1, pp. 149-162, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[31] Sukriti Dhang, Mimi Zhang, and Soumyabrata Dev, “AdSegNet: A Deep Network to Localize Billboard in Outdoor Scenes,” Signal, Image and Video Processing, vol. 18, no. 10, pp. 7221-7235, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[32] Wonkyung Kim et al., “A Deep Learning Approach for Identifying User Interest from Targeted Advertising,” Journal of Information Processing Systems, vol. 18, no. 2, pp. 245-257, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[33] Divya Nimma et al., “Object Detection in Real-Time Video Surveillance using Attention based Transformer-YOLOv8 Model,” Alexandria Engineering Journal, vol. 118, pp. 482-495, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[34] Martin Magdin, and Zoltán Balogh, “Comparison Classification Algorithms and the YOLO Method for Video Analysis and Object Detection,” Scientific Reports, vol. 15, no. 1, pp. 1-13, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[35] Marco A. Moreno-Armendáriz et al., “Deep-Learning-based Adaptive Advertising with Augmented Reality,” Sensors, vol. 22, no. 1, pp. 1-20, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[36] Chien-Yao Wang, Alexey Bochkovskiy, and Hong-Yuan Mark Liao, “YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors,” 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, pp. 7464-7475, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[37] Hangyue Zhao, Hongpu Zhang, and Yanyun Zhao, “YOLOv7-sea: Object Detection of Maritime UAV Images based on Improved YOLOv7,” 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), Waikoloa, HI, USA, pp. 233-238, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[38] Frouke Hermens, “Automatic Object Detection for Behavioural Research using YOLOv8,” Behavior Research Methods, vol. 56, no. 7, pp. 7307-7330, 2024.
[CrossRef] [Google Scholar] [Publisher Link]