{"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/impact-of-ground-truth-annotation-quality-on","title":"Impact of Ground Truth Annotation Quality on Performance of Semantic Image Segmentation of Traffic Conditions","arxiv_id":"1901.00001","date":"2018-12-30","proceeding":null,"authors":["Vlad Taran","Yuri Gordienko","Alexandr Rokovyi","Oleg Alienin","Sergii Stirenko"],"abstract":"Preparation of high-quality datasets for the urban scene understanding is a\nlabor-intensive task, especially, for datasets designed for the autonomous\ndriving applications. The application of the coarse ground truth (GT)\nannotations of these datasets without detriment to the accuracy of semantic\nimage segmentation (by the mean intersection over union - mIoU) could simplify\nand speedup the dataset preparation and model fine tuning before its practical\napplication. Here the results of the comparative analysis for semantic\nsegmentation accuracy obtained by PSPNet deep learning architecture are\npresented for fine and coarse annotated images from Cityscapes dataset. Two\nscenarios were investigated: scenario 1 - the fine GT images for training and\nprediction, and scenario 2 - the fine GT images for training and the coarse GT\nimages for prediction. The obtained results demonstrated that for the most\nimportant classes the mean accuracy values of semantic image segmentation for\ncoarse GT annotations are higher than for the fine GT ones, and the standard\ndeviation values are vice versa. It means that for some applications some\nunimportant classes can be excluded and the model can be tuned further for some\nclasses and specific regions on the coarse GT dataset without loss of the\naccuracy even. Moreover, this opens the perspectives to use deep neural\nnetworks for the preparation of such coarse GT datasets.","url_abs":"http://arxiv.org/abs/1901.00001v1","url_pdf":"http://arxiv.org/pdf/1901.00001v1.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":"impact-of-ground-truth-annotation-quality-on","repo_url":"https://github.com/Lawhy/Searchive","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"auxiliary-classifier","method_name":"Auxiliary Classifier"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dilated-convolution","method_name":"Dilated Convolution"},{"method_slug":"pspnet","method_name":"PSPNet"},{"method_slug":"pyramid-pooling-module","method_name":"Pyramid Pooling Module"},{"method_slug":"relu","method_name":"ReLU"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}