This research paper proposes a novel methodology for image-to-image style transfer that focuses on applying artistic styles to segmented objects within an image. The approach leverages the YOLOv8 segmentation model and the backbone neural network of YOLOv8 to achieve this in a single deep network.
The key highlights of the proposed approach are:
The authors highlight that this integrated approach advances the state-of-the-art in object-based style transfer by leveraging the latest advancements in segmentation models and style transfer techniques within a single deep network framework.
إلى لغة أخرى
من محتوى المصدر
arxiv.org
الرؤى الأساسية المستخلصة من
by Harshmohan K... في arxiv.org 04-16-2024
https://arxiv.org/pdf/2404.09461.pdfاستفسارات أعمق