Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{180342,
        author = {Shantanu Kharade and Pranav Asane and Shubham Tapale and Neeraj Kalambe and Priti Malkhede},
        title = {Face Sketch to Image Generation using Hybrid GAN},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {1139-1146},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180342},
        abstract = {This research introduces a novel approach to face sketch-to-image synthesis using a hybrid Generative Adversarial Network (GAN) architecture that incorporates UNet-style skip connections. Our model effectively bridges the domain gap between sketches and photorealistic facial images by leveraging both the generative capabilities of a bidirectional GAN and the structural preservation advantages of skip connections. By implementing CycleGAN principles with custom-designed generator and discriminator architectures, we achieve notable improvements in both perceptual quality and structural fidelity. Experimental results demonstrate that our approach outperforms several state-of-the-art methods, achieving an SSIM score of 70 and a PSNR of 16. The model also demonstrates robustness across diverse sketch styles and facial attributes, making it suitable for real-world applications in law enforcement, entertainment, and digital art. The web application built on this model provides a user-friendly interface that enables users to easily upload sketches and generate high-quality face images, with features including real-time processing indicators and image download options.},
        keywords = {Face sketch to image generation, Generative Adversarial Networks (GANs), Deep learning, Web application, DCGAN, CycleGAN, User-friendly interface, Image synthesis, Computer vision.},
        month = {June},
        }