{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/high-quality-facial-photo-sketch-synthesis","title":"High-Quality Facial Photo-Sketch Synthesis Using Multi-Adversarial Networks","arxiv_id":"1710.10182","date":"2017-10-27","proceeding":null,"authors":["Lidan Wang","Vishwanath A. Sindagi","Vishal M. Patel"],"abstract":"Synthesizing face sketches from real photos and its inverse have many\napplications. However, photo/sketch synthesis remains a challenging problem due\nto the fact that photo and sketch have different characteristics. In this work,\nwe consider this task as an image-to-image translation problem and explore the\nrecently popular generative models (GANs) to generate high-quality realistic\nphotos from sketches and sketches from photos. Recent GAN-based methods have\nshown promising results on image-to-image translation problems and\nphoto-to-sketch synthesis in particular, however, they are known to have\nlimited abilities in generating high-resolution realistic images. To this end,\nwe propose a novel synthesis framework called Photo-Sketch Synthesis using\nMulti-Adversarial Networks, (PS2-MAN) that iteratively generates low resolution\nto high resolution images in an adversarial way. The hidden layers of the\ngenerator are supervised to first generate lower resolution images followed by\nimplicit refinement in the network to generate higher resolution images.\nFurthermore, since photo-sketch synthesis is a coupled/paired translation\nproblem, we leverage the pair information using CycleGAN framework. Both Image\nQuality Assessment (IQA) and Photo-Sketch Matching experiments are conducted to\ndemonstrate the superior performance of our framework in comparison to existing\nstate-of-the-art solutions. Code available at:\nhttps://github.com/lidan1/PhotoSketchMAN.","url_abs":"http://arxiv.org/abs/1710.10182v2","url_pdf":"http://arxiv.org/pdf/1710.10182v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"high-quality-facial-photo-sketch-synthesis","repo_url":"https://github.com/lidan1/PhotoSketchMAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"face-sketch-synthesis","task_name":"Face Sketch Synthesis"},{"task_slug":"image-quality-assessment","task_name":"Image Quality Assessment"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"high","task_name":"Vocal Bursts Intensity Prediction"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"cycle-consistency-loss","method_name":"Cycle Consistency Loss"},{"method_slug":"gan-least-squares-loss","method_name":"GAN Least Squares Loss"},{"method_slug":"instance-normalization","method_name":"Instance Normalization"},{"method_slug":"patchgan","method_name":"PatchGAN"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-sketch-synthesis-on-cuhk","task":"Face Sketch Synthesis","dataset":"CUHK","model":"PS2-MAN","rank_in_archive_order":2,"of":2,"metrics":{"FSIM":"73.61%","SSIM":"61.56%"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}