{"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/semantic-adversarial-examples","title":"Semantic Adversarial Examples","arxiv_id":"1804.00499","date":"2018-03-16","proceeding":null,"authors":["Hossein Hosseini","Radha Poovendran"],"abstract":"Deep neural networks are known to be vulnerable to adversarial examples,\ni.e., images that are maliciously perturbed to fool the model. Generating\nadversarial examples has been mostly limited to finding small perturbations\nthat maximize the model prediction error. Such images, however, contain\nartificial perturbations that make them somewhat distinguishable from natural\nimages. This property is used by several defense methods to counter adversarial\nexamples by applying denoising filters or training the model to be robust to\nsmall perturbations.\n  In this paper, we introduce a new class of adversarial examples, namely\n\"Semantic Adversarial Examples,\" as images that are arbitrarily perturbed to\nfool the model, but in such a way that the modified image semantically\nrepresents the same object as the original image. We formulate the problem of\ngenerating such images as a constrained optimization problem and develop an\nadversarial transformation based on the shape bias property of human cognitive\nsystem. In our method, we generate adversarial images by first converting the\nRGB image into the HSV (Hue, Saturation and Value) color space and then\nrandomly shifting the Hue and Saturation components, while keeping the Value\ncomponent the same. Our experimental results on CIFAR10 dataset show that the\naccuracy of VGG16 network on adversarial color-shifted images is 5.7%.","url_abs":"http://arxiv.org/abs/1804.00499v1","url_pdf":"http://arxiv.org/pdf/1804.00499v1.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":"semantic-adversarial-examples","repo_url":"https://github.com/HosseinHosseini/Semantic-Adversarial-Examples","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.00499","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1804.00499"}},"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/HosseinHosseini/Semantic-Adversarial-Examples","reach":null}],"summary":{"ran_honours":1,"unverified":1},"by_repo_kind":{"listed":{"samples":2,"ran":1,"repositories":1}},"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":2,"samples":[{"code_sha256_prefix":"1a06bde28c9caa07","entry":"data_normalize","repo":"HosseinHosseini/Semantic-Adversarial-Examples","repo_kind":"listed","path":"utils_cifar10.py","file_url":"https://github.com/HosseinHosseini/Semantic-Adversarial-Examples/blob/HEAD/utils_cifar10.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1a06bde28c9caa07"}},{"code_sha256_prefix":"5173abd0f030602d","entry":"color_shift_attack","repo":"HosseinHosseini/Semantic-Adversarial-Examples","repo_kind":"listed","path":"color_shift_attack.py","file_url":"https://github.com/HosseinHosseini/Semantic-Adversarial-Examples/blob/HEAD/color_shift_attack.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5173abd0f030602d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}