{"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/applying-domain-randomization-to-synthetic","title":"Applying Domain Randomization to Synthetic Data for Object Category Detection","arxiv_id":"1807.09834","date":"2018-07-16","proceeding":null,"authors":["João Borrego","Atabak Dehban","Rui Figueiredo","Plinio Moreno","Alexandre Bernardino","José Santos-Victor"],"abstract":"Recent advances in deep learning-based object detection techniques have\nrevolutionized their applicability in several fields. However, since these\nmethods rely on unwieldy and large amounts of data, a common practice is to\ndownload models pre-trained on standard datasets and fine-tune them for\nspecific application domains with a small set of domain relevant images. In\nthis work, we show that using synthetic datasets that are not necessarily\nphoto-realistic can be a better alternative to simply fine-tune pre-trained\nnetworks. Specifically, our results show an impressive 25% improvement in the\nmAP metric over a fine-tuning baseline when only about 200 labelled images are\navailable to train. Finally, an ablation study of our results is presented to\ndelineate the individual contribution of different components in the\nrandomization pipeline.","url_abs":"http://arxiv.org/abs/1807.09834v1","url_pdf":"http://arxiv.org/pdf/1807.09834v1.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":"applying-domain-randomization-to-synthetic","repo_url":"https://github.com/jsbruglie/tf-shape-detection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.09834","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}