{"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/on-zero-shot-recognition-of-generic-objects","title":"On zero-shot recognition of generic objects","arxiv_id":"1904.04957","date":"2019-04-10","proceeding":"CVPR 2019 6","authors":["Tristan Hascoet","Yasuo Ariki","Tetsuya Takiguchi"],"abstract":"Many recent advances in computer vision are the result of a healthy\ncompetition among researchers on high quality, task-specific, benchmarks. After\na decade of active research, zero-shot learning (ZSL) models accuracy on the\nImagenet benchmark remains far too low to be considered for practical object\nrecognition applications. In this paper, we argue that the main reason behind\nthis apparent lack of progress is the poor quality of this benchmark. We\nhighlight major structural flaws of the current benchmark and analyze different\nfactors impacting the accuracy of ZSL models. We show that the actual\nclassification accuracy of existing ZSL models is significantly higher than was\npreviously thought as we account for these flaws. We then introduce the notion\nof structural bias specific to ZSL datasets. We discuss how the presence of\nthis new form of bias allows for a trivial solution to the standard benchmark\nand conclude on the need for a new benchmark. We then detail the semi-automated\nconstruction of a new benchmark to address these flaws.","url_abs":"http://arxiv.org/abs/1904.04957v1","url_pdf":"http://arxiv.org/pdf/1904.04957v1.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":"on-zero-shot-recognition-of-generic-objects","repo_url":"https://github.com/TristHas/GOZ","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"object-recognition","task_name":"Object Recognition"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[],"datasets_introduced":[{"slug":"goz","name":"GOZ","full_name":"Generic Object ZSL Dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}