2019/08/30 by Lironne Kurzman, David Vázquez, Kurzman, Lironne +3 · 1 citation
Computer Science · #Advanced Vision and Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Image Enhancement Techniques
paper · pdf · doi:10.48550/arxiv.1908.11525
openalex publication_date 2019/08/30 · openalex created_date 2022/07/28 · openalex updated_date 2026/07/28
We propose a Class-Based Styling method (CBS) that can map different styles\nfor different object classes in real-time. CBS achieves real-time performance\nby carrying out two steps simultaneously. While a semantic segmentation method\nis used to obtain the mask of each object class in a video frame, a styling\nmethod is used to style that frame globally. Then an object class can be styled\nby combining the segmentation mask and the styled image. The user can also\nselect multiple styles so that different object classes can have different\nstyles in a single frame. For semantic segmentation, we leverage DABNet that\nachieves high accuracy, yet only has 0.76 million parameters and runs at 104\nFPS. For the style transfer step, we use a popular real-time method proposed by\nJohnson et al. [7]. We evaluated CBS on a video of the CityScapes dataset and\nobserved high-quality localized style transfer results for different object\nclasses and real-time performance.\n