{"_id":"WYxQmTPyxTfnFb3gK","bibbaseid":"saleh-zhang-calvozaragoza-vigliensoni-fujinaga-pixeljswebbasedpixelclassificationcorrectionplatformforgroundtruthcreation-2017","author_short":["Saleh, Z.","Zhang, K.","Calvo-Zaragoza, J.","Vigliensoni, G.","Fujinaga, I."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Pixel.js: Web-Based Pixel Classification Correction Platform for Ground Truth Creation","isbn":"978-1-5386-3586-5","shorttitle":"Pixel.js","url":"https://www.computer.org/csdl/proceedings-article/icdar/2017/3586c039/12OmNylboHt","doi":"10.1109/ICDAR.2017.267","abstract":"Image segmentation plays a key role in document recognition and analysis. However, segmentation algorithms output a non-negligible amount of misclassified pixels. We introduce Pixel.js, an open-source, web-based pixel-level classification correction platform to correct the output of inaccurate heuristic and trained image segmentation algorithms. The corrected output can be used as ground truth for training or evaluating the performance of such algorithms. Our goal is to provide an accessible platform that can be integrated in complex workflows to reduce the time and resources spent on the manual creation of the aforementioned ground truth data.","language":"English","urldate":"2023-11-09","publisher":"IEEE Computer Society","author":[{"propositions":[],"lastnames":["Saleh"],"firstnames":["Zeyad"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Ké"],"suffixes":[]},{"propositions":[],"lastnames":["Calvo-Zaragoza"],"firstnames":["Jorge"],"suffixes":[]},{"propositions":[],"lastnames":["Vigliensoni"],"firstnames":["Gabriel"],"suffixes":[]},{"propositions":[],"lastnames":["Fujinaga"],"firstnames":["Ichiro"],"suffixes":[]}],"month":"November","year":"2017","note":"ISSN: 2379-2140","keywords":"#nosource","pages":"39–40","bibtex":"@inproceedings{saleh_pixeljs_2017,\n\ttitle = {Pixel.js: {Web}-{Based} {Pixel} {Classification} {Correction} {Platform} for {Ground} {Truth} {Creation}},\n\tisbn = {978-1-5386-3586-5},\n\tshorttitle = {Pixel.js},\n\turl = {https://www.computer.org/csdl/proceedings-article/icdar/2017/3586c039/12OmNylboHt},\n\tdoi = {10.1109/ICDAR.2017.267},\n\tabstract = {Image segmentation plays a key role in document recognition and analysis. However, segmentation algorithms output a non-negligible amount of misclassified pixels. We introduce Pixel.js, an open-source, web-based pixel-level classification correction platform to correct the output of inaccurate heuristic and trained image segmentation algorithms. The corrected output can be used as ground truth for training or evaluating the performance of such algorithms. Our goal is to provide an accessible platform that can be integrated in complex workflows to reduce the time and resources spent on the manual creation of the aforementioned ground truth data.},\n\tlanguage = {English},\n\turldate = {2023-11-09},\n\tpublisher = {IEEE Computer Society},\n\tauthor = {Saleh, Zeyad and Zhang, Ké and Calvo-Zaragoza, Jorge and Vigliensoni, Gabriel and Fujinaga, Ichiro},\n\tmonth = nov,\n\tyear = {2017},\n\tnote = {ISSN: 2379-2140},\n\tkeywords = {\\#nosource},\n\tpages = {39--40},\n}\n\n\n\n","author_short":["Saleh, Z.","Zhang, K.","Calvo-Zaragoza, J.","Vigliensoni, G.","Fujinaga, I."],"key":"saleh_pixeljs_2017","id":"saleh_pixeljs_2017","bibbaseid":"saleh-zhang-calvozaragoza-vigliensoni-fujinaga-pixeljswebbasedpixelclassificationcorrectionplatformforgroundtruthcreation-2017","role":"author","urls":{"Paper":"https://www.computer.org/csdl/proceedings-article/icdar/2017/3586c039/12OmNylboHt"},"keyword":["#nosource"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/fsimonetta","dataSources":["pzyFFGWvxG2bs63zP"],"keywords":["#nosource"],"search_terms":["pixel","web","based","pixel","classification","correction","platform","ground","truth","creation","saleh","zhang","calvo-zaragoza","vigliensoni","fujinaga"],"title":"Pixel.js: Web-Based Pixel Classification Correction Platform for Ground Truth Creation","year":2017}