LatentStyleGAN: One-Shot Face Face Stylization Using GAN Inversion

LatentStyleGAN: One-Shot Face Face Stylization Using GAN Inversion

1.Maya Sai Gopala Krishna 2. Sathi karthik sudheer Reddy 3. Chukkala Phani kumar 4.Mr. Baidyanath Ram

  Maya Sai Gopala Krishna, Amity School of Engineering and Technology, Amity University Chhattisgarh, Raipur, India – 493225

 Sathi Karthik Sudheer Reddy, Amity School of Engineering and Technology, Amity University Chhattisgarh, Raipur, India – 493225

 Chukkala Phani Kumar, Amity School of Engineering and Technology, Amity University Chhattisgarh, Raipur, India – 493225

Mr. Baidyanath Ram, Amity School of Engineering and Technology, Amity University Chhattisgarh, Raipur, India – 493225

Abstract
Image stylization has become a prominent area in computer vision, enabling creative transformations of real-world photos into artistic formats like sketches and cartoons. This paper introduces StyleWeb, a web-based application that allows users to convert ordinary images into stylized outputs in real time. The system integrates two core approaches: sketch generation using traditional image processing techniques and cartoonization powered by a pre-trained AnimeGANv3 model converted into ONNX format for optimized performance. Developed using the Django framework, StyleWeb offers a secure and responsive backend for efficient image processing and delivery. Through a simple interface, users can upload images, choose a desired style, and receive processed results with minimal delay. The platform bridges modern deep learning methods with practical web deployment, providing an accessible solution for stylized image generation suitable for artists, educators, and casual users alike. StyleWeb showcases how AI-driven creativity can be brought to everyday applications through thoughtful system design and implementation.