Creatism: A deep-learning photographer capable of creating professional work. Fang, H. & Zhang, M. arXiv:1707.03491 [cs], July, 2017. arXiv: 1707.03491Paper abstract bibtex Machine-learning excels in many areas with well-defined goals. However, a clear goal is usually not available in art forms, such as photography. The success of a photograph is measured by its aesthetic value, a very subjective concept. This adds to the challenge for a machine learning approach. We introduce Creatism, a deep-learning system for artistic content creation. In our system, we break down aesthetics into multiple aspects, each can be learned individually from a shared dataset of professional examples. Each aspect corresponds to an image operation that can be optimized efficiently. A novel editing tool, dramatic mask, is introduced as one operation that improves dramatic lighting for a photo. Our training does not require a dataset with before/after image pairs, or any additional labels to indicate different aspects in aesthetics. Using our system, we mimic the workflow of a landscape photographer, from framing for the best composition to carrying out various post-processing operations. The environment for our virtual photographer is simulated by a collection of panorama images from Google Street View. We design a "Turing-test"-like experiment to objectively measure quality of its creations, where professional photographers rate a mixture of photographs from different sources blindly. Experiments show that a portion of our robot's creation can be confused with professional work.
@article{fang_creatism:_2017,
title = {Creatism: {A} deep-learning photographer capable of creating professional work},
shorttitle = {Creatism},
url = {http://arxiv.org/abs/1707.03491},
abstract = {Machine-learning excels in many areas with well-defined goals. However, a clear goal is usually not available in art forms, such as photography. The success of a photograph is measured by its aesthetic value, a very subjective concept. This adds to the challenge for a machine learning approach. We introduce Creatism, a deep-learning system for artistic content creation. In our system, we break down aesthetics into multiple aspects, each can be learned individually from a shared dataset of professional examples. Each aspect corresponds to an image operation that can be optimized efficiently. A novel editing tool, dramatic mask, is introduced as one operation that improves dramatic lighting for a photo. Our training does not require a dataset with before/after image pairs, or any additional labels to indicate different aspects in aesthetics. Using our system, we mimic the workflow of a landscape photographer, from framing for the best composition to carrying out various post-processing operations. The environment for our virtual photographer is simulated by a collection of panorama images from Google Street View. We design a "Turing-test"-like experiment to objectively measure quality of its creations, where professional photographers rate a mixture of photographs from different sources blindly. Experiments show that a portion of our robot's creation can be confused with professional work.},
urldate = {2018-01-17TZ},
journal = {arXiv:1707.03491 [cs]},
author = {Fang, Hui and Zhang, Meng},
month = jul,
year = {2017},
note = {arXiv: 1707.03491},
keywords = {Computer Science - Computer Vision and Pattern Recognition}
}
Downloads: 0
{"_id":"zop7HYDBtiMkri2JS","bibbaseid":"fang-zhang-creatismadeeplearningphotographercapableofcreatingprofessionalwork-2017","downloads":0,"creationDate":"2018-04-06T04:26:07.174Z","title":"Creatism: A deep-learning photographer capable of creating professional work","author_short":["Fang, H.","Zhang, M."],"year":2017,"bibtype":"article","biburl":"https://bibbase.org/zotero/alwynmathew","bibdata":{"bibtype":"article","type":"article","title":"Creatism: A deep-learning photographer capable of creating professional work","shorttitle":"Creatism","url":"http://arxiv.org/abs/1707.03491","abstract":"Machine-learning excels in many areas with well-defined goals. However, a clear goal is usually not available in art forms, such as photography. The success of a photograph is measured by its aesthetic value, a very subjective concept. This adds to the challenge for a machine learning approach. We introduce Creatism, a deep-learning system for artistic content creation. In our system, we break down aesthetics into multiple aspects, each can be learned individually from a shared dataset of professional examples. Each aspect corresponds to an image operation that can be optimized efficiently. A novel editing tool, dramatic mask, is introduced as one operation that improves dramatic lighting for a photo. Our training does not require a dataset with before/after image pairs, or any additional labels to indicate different aspects in aesthetics. Using our system, we mimic the workflow of a landscape photographer, from framing for the best composition to carrying out various post-processing operations. The environment for our virtual photographer is simulated by a collection of panorama images from Google Street View. We design a \"Turing-test\"-like experiment to objectively measure quality of its creations, where professional photographers rate a mixture of photographs from different sources blindly. Experiments show that a portion of our robot's creation can be confused with professional work.","urldate":"2018-01-17TZ","journal":"arXiv:1707.03491 [cs]","author":[{"propositions":[],"lastnames":["Fang"],"firstnames":["Hui"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Meng"],"suffixes":[]}],"month":"July","year":"2017","note":"arXiv: 1707.03491","keywords":"Computer Science - Computer Vision and Pattern Recognition","bibtex":"@article{fang_creatism:_2017,\n\ttitle = {Creatism: {A} deep-learning photographer capable of creating professional work},\n\tshorttitle = {Creatism},\n\turl = {http://arxiv.org/abs/1707.03491},\n\tabstract = {Machine-learning excels in many areas with well-defined goals. However, a clear goal is usually not available in art forms, such as photography. The success of a photograph is measured by its aesthetic value, a very subjective concept. This adds to the challenge for a machine learning approach. We introduce Creatism, a deep-learning system for artistic content creation. In our system, we break down aesthetics into multiple aspects, each can be learned individually from a shared dataset of professional examples. Each aspect corresponds to an image operation that can be optimized efficiently. A novel editing tool, dramatic mask, is introduced as one operation that improves dramatic lighting for a photo. Our training does not require a dataset with before/after image pairs, or any additional labels to indicate different aspects in aesthetics. Using our system, we mimic the workflow of a landscape photographer, from framing for the best composition to carrying out various post-processing operations. The environment for our virtual photographer is simulated by a collection of panorama images from Google Street View. We design a \"Turing-test\"-like experiment to objectively measure quality of its creations, where professional photographers rate a mixture of photographs from different sources blindly. Experiments show that a portion of our robot's creation can be confused with professional work.},\n\turldate = {2018-01-17TZ},\n\tjournal = {arXiv:1707.03491 [cs]},\n\tauthor = {Fang, Hui and Zhang, Meng},\n\tmonth = jul,\n\tyear = {2017},\n\tnote = {arXiv: 1707.03491},\n\tkeywords = {Computer Science - Computer Vision and Pattern Recognition}\n}\n\n","author_short":["Fang, H.","Zhang, M."],"key":"fang_creatism:_2017","id":"fang_creatism:_2017","bibbaseid":"fang-zhang-creatismadeeplearningphotographercapableofcreatingprofessionalwork-2017","role":"author","urls":{"Paper":"http://arxiv.org/abs/1707.03491"},"keyword":["Computer Science - Computer Vision and Pattern Recognition"],"downloads":0,"html":""},"search_terms":["creatism","deep","learning","photographer","capable","creating","professional","work","fang","zhang"],"keywords":["computer science - computer vision and pattern recognition"],"authorIDs":[],"dataSources":["p3JdPh89hHfoARFkn"]}