Transfer learning using Generative Artificial Intelligence for Object Detection

DOI

10.2991/978-94-6463-858-5_255How to use a DOI?

Keywords

Object detection; transfer learning; generative artificial intelligence; convolutional neural networks; generative adversarial networks; synthetic data; small datasets; feature extraction; real-time detection; computer vision

Abstract

Object detection is the very backbone of computer vision. Applications surface in autonomous vehicles, healthcare diagnostics, and surveillance systems. All the conventional methods of object detection typically require massive amounts of annotated data and use heavy computation, making it hard to use them in conditions where data is scarce or hardware is limited. Thus, the work proposes an entirely new approach combining transfer learning and generative artificial intelligence for such situations. Transfer learning would use previously learned model inputs, while generative artificial intelligence-generate new realistic synthetic samples using the technology of generative adversarial networks. These two were being integrated into one. A new, robust Object Detection System would thus be created. Fine-tuning pre-trained convolutional neural networks to extract domain-specific features involved employing GANs at varying degrees to solve data imbalance issues and increase levels of generalization. A setting between the server and the client with a web-based interface developed using Flask was also part of the system. Essentially, users must upload images and videos that would then be submitted into the real-time detection system with accuracy metrics underneath them. Finally and most importantly, experiments conduct the task of validating the framework that proves increased detection accuracy and adaptability with respect to limited datasets. Therefore, this research follows hand in hand between scenarios of small data and the high-performance platform outcomes, hence, show possibilities where transfer learning meets generative AI in archiving the next stroke in object detection technologies. It offers an extremely scalable, efficient, as well as resource kind scheme to practice the solution in the real world, making a significant contribution to the field of computer vision.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S. Prayla Shyry
AU  - Sahil
AU  - Saranu Akhil
PY  - 2025
DA  - 2025/11/04
TI  - Transfer learning using Generative Artificial Intelligence for Object Detection
BT  - Proceedings of International Conference on Computer Science and Communication Engineering (ICCSCE 2025)
PB  - Atlantis Press
SP  - 3052
EP  - 3064
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-858-5_255
DO  - 10.2991/978-94-6463-858-5_255
ID  - Shyry2025
ER  -