{"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/syn2real-a-new-benchmark-forsynthetic-to-real","title":"Syn2Real: A New Benchmark forSynthetic-to-Real Visual Domain Adaptation","arxiv_id":"1806.09755","date":"2018-06-26","proceeding":null,"authors":["Xingchao Peng","Ben Usman","Kuniaki Saito","Neela Kaushik","Judy Hoffman","Kate Saenko"],"abstract":"Unsupervised transfer of object recognition models from synthetic to real\ndata is an important problem with many potential applications. The challenge is\nhow to \"adapt\" a model trained on simulated images so that it performs well on\nreal-world data without any additional supervision. Unfortunately, current\nbenchmarks for this problem are limited in size and task diversity. In this\npaper, we present a new large-scale benchmark called Syn2Real, which consists\nof a synthetic domain rendered from 3D object models and two real-image domains\ncontaining the same object categories. We define three related tasks on this\nbenchmark: closed-set object classification, open-set object classification,\nand object detection. Our evaluation of multiple state-of-the-art methods\nreveals a large gap in adaptation performance between the easier closed-set\nclassification task and the more difficult open-set and detection tasks. We\nconclude that developing adaptation methods that work well across all three\ntasks presents a significant future challenge for syn2real domain transfer.","url_abs":"http://arxiv.org/abs/1806.09755v1","url_pdf":"http://arxiv.org/pdf/1806.09755v1.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":[],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-recognition","task_name":"Object Recognition"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"syn2real","name":"Syn2Real","full_name":"Syn2Real"}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.09755","atlas_url":"https://app.syntology.ai/?focus=1806.09755","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}