{"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/comparing-deep-neural-networks-against-humans","title":"Comparing deep neural networks against humans: object recognition when the signal gets weaker","arxiv_id":"1706.06969","date":"2017-06-21","proceeding":null,"authors":["Robert Geirhos","David H. J. Janssen","Heiko H. Schütt","Jonas Rauber","Matthias Bethge","Felix A. Wichmann"],"abstract":"Human visual object recognition is typically rapid and seemingly effortless,\nas well as largely independent of viewpoint and object orientation. Until very\nrecently, animate visual systems were the only ones capable of this remarkable\ncomputational feat. This has changed with the rise of a class of computer\nvision algorithms called deep neural networks (DNNs) that achieve human-level\nclassification performance on object recognition tasks. Furthermore, a growing\nnumber of studies report similarities in the way DNNs and the human visual\nsystem process objects, suggesting that current DNNs may be good models of\nhuman visual object recognition. Yet there clearly exist important\narchitectural and processing differences between state-of-the-art DNNs and the\nprimate visual system. The potential behavioural consequences of these\ndifferences are not well understood. We aim to address this issue by comparing\nhuman and DNN generalisation abilities towards image degradations. We find the\nhuman visual system to be more robust to image manipulations like contrast\nreduction, additive noise or novel eidolon-distortions. In addition, we find\nprogressively diverging classification error-patterns between humans and DNNs\nwhen the signal gets weaker, indicating that there may still be marked\ndifferences in the way humans and current DNNs perform visual object\nrecognition. We envision that our findings as well as our carefully measured\nand freely available behavioural datasets provide a new useful benchmark for\nthe computer vision community to improve the robustness of DNNs and a\nmotivation for neuroscientists to search for mechanisms in the brain that could\nfacilitate this robustness.","url_abs":"http://arxiv.org/abs/1706.06969v2","url_pdf":"http://arxiv.org/pdf/1706.06969v2.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":"comparing-deep-neural-networks-against-humans","repo_url":"https://github.com/rgeirhos/object-recognition","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-recognition","task_name":"Object Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.06969","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1706.06969"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rgeirhos/object-recognition","reach":{"status":"unanswered"}}],"summary":{"ran_fixture":1},"by_repo_kind":{},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"e92a4458fd3a9644","entry":"adjust_contrast","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"e92a4458fd3a9644"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}