{"id":20058644,"url":"https://github.com/roboflow/neuralhash-collisions","last_synced_at":"2025-05-05T14:31:30.983Z","repository":{"id":98131696,"uuid":"397943518","full_name":"roboflow/neuralhash-collisions","owner":"roboflow","description":"A catalog of naturally occurring images whose Apple NeuralHash is identical.","archived":false,"fork":false,"pushed_at":"2022-12-26T13:50:58.000Z","size":28398,"stargazers_count":149,"open_issues_count":0,"forks_count":6,"subscribers_count":24,"default_branch":"main","last_synced_at":"2025-04-30T20:50:52.607Z","etag":null,"topics":["apple","csam","neuralhash","perceptual-hashing"],"latest_commit_sha":null,"homepage":"https://blog.roboflow.com/nerualhash-collision/","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/roboflow.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-08-19T12:57:24.000Z","updated_at":"2025-03-26T14:23:57.000Z","dependencies_parsed_at":null,"dependency_job_id":"482ad8ea-398c-4f15-bafa-8c38f42102ef","html_url":"https://github.com/roboflow/neuralhash-collisions","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/roboflow%2Fneuralhash-collisions","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/roboflow%2Fneuralhash-collisions/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/roboflow%2Fneuralhash-collisions/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/roboflow%2Fneuralhash-collisions/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/roboflow","download_url":"https://codeload.github.com/roboflow/neuralhash-collisions/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":252514976,"owners_count":21760479,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["apple","csam","neuralhash","perceptual-hashing"],"created_at":"2024-11-13T13:02:42.291Z","updated_at":"2025-05-05T14:31:30.977Z","avatar_url":"https://github.com/roboflow.png","language":"JavaScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# neuralhash-collisions\nA catalog of naturally occurring images whose Apple NeuralHash is identical.\n\nSee this [NeuralHash collision blog post](https://blog.roboflow.com/nerualhash-collision/) with full details and background.\n\n## Contributing\n\nPRs are welcome and encouraged! You can find the code to run NeuralHash on images in \n[this repo](https://github.com/AsuharietYgvar/AppleNeuralHash2ONNX). \nPlease do not submit artificially generated adversarial examples. This repo is meant to document \nnatural collisions found in the wild.\n\nPlease follow the example format (add a folder in `collisions` named with the colliding hash and\ncontaining the example images with a README cataloging information about their provenance). Ideally\nprovide evidence that these images existed prior to the public divulgence of the NeuralHash weights\nto prove that they are not artificial.\n\n## Suggested Methodology\n\nThere are many [open source image datasets](https://universe.roboflow.com)\nwhich contain a wide variety of images.\n\nIf you generate a NeuralHash for each one, store the image path / hash pairs in\na text file, sort them by hash, and compare adjacent lines' hashes, you\ncan quickly check n^2 image pairs (where `n` is the size of the dataset).\n\nThere are some example scripts you may adapt in the `utils` directory.\n\n## Hash Lists\n\nYou may also compare your hashes against the ImageNet hashes using the txt\nfile archived in the `imagenet` directory.\n\nIf you check a large, public corpus please PR the hash list so others can\neasily compare them for collisions across sources.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Froboflow%2Fneuralhash-collisions","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Froboflow%2Fneuralhash-collisions","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Froboflow%2Fneuralhash-collisions/lists"}