{"id":20730837,"url":"https://github.com/bharathsudharsan/covid-away","last_synced_at":"2025-06-14T11:05:09.884Z","repository":{"id":48745697,"uuid":"274868750","full_name":"bharathsudharsan/COVID-away","owner":"bharathsudharsan","description":"Code for paper 'Avoid touching your face: A hand-to-face 3d motion dataset (covid-away) and trained models for smartwatches'","archived":false,"fork":false,"pushed_at":"2022-07-23T09:52:33.000Z","size":107177,"stargazers_count":18,"open_issues_count":0,"forks_count":4,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-04-23T22:04:00.526Z","etag":null,"topics":["3d-motion","accelerometer","barometric-pressure","cnn-model","covid-19","dataset-creation","features-extraction","gyroscope","iforest","local-outlier-factor","one-class-classification","one-class-svm","optimization","pickle","rotation-vectors","sensor-fusion","smartwatch","tflite","visualization"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bharathsudharsan.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-06-25T08:45:05.000Z","updated_at":"2024-02-23T06:43:12.000Z","dependencies_parsed_at":"2022-07-26T00:46:59.065Z","dependency_job_id":null,"html_url":"https://github.com/bharathsudharsan/COVID-away","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/bharathsudharsan/COVID-away","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bharathsudharsan%2FCOVID-away","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bharathsudharsan%2FCOVID-away/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bharathsudharsan%2FCOVID-away/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bharathsudharsan%2FCOVID-away/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bharathsudharsan","download_url":"https://codeload.github.com/bharathsudharsan/COVID-away/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bharathsudharsan%2FCOVID-away/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":259804839,"owners_count":22913901,"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":["3d-motion","accelerometer","barometric-pressure","cnn-model","covid-19","dataset-creation","features-extraction","gyroscope","iforest","local-outlier-factor","one-class-classification","one-class-svm","optimization","pickle","rotation-vectors","sensor-fusion","smartwatch","tflite","visualization"],"created_at":"2024-11-17T05:12:52.598Z","updated_at":"2025-06-14T11:05:09.868Z","avatar_url":"https://github.com/bharathsudharsan.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"## COVID-away: Hand-to-face 3D Motion Dataset and Models for Smartwatches\n\n### Overview\n\nWe humans on average touch our face (eye, nose and mouth) 10-20 times an hour, which is often the primary source of getting infected by a variety of viral infections including seasonal Influenza, Coronavirus, Swine flu, Ebola virus, etc. \n\nIn this work, we have collected a hand-to-face multi-sensor 3D motion dataset and named it COVID-away dataset.\n\nUsing our dataset, we trained models that can continuously monitor human arm/hand movement using a wearable device and trigger a timely notification (e.g. vibration) to warn the device users when their hands are moved (unintentionally) towards their face. \n\nThe trained COVID-away models can be easily integrated into an app for smartwatches or fitness bands. \n\nEvaluation shows that the Minimum Covariance Determinant (MCD) model produces the highest F1-score (0.93) using just the smartwatch’s accelerometer data (39 features).\n\n**Paper:** [https://dl.acm.org/doi/10.1145/3423423.3423433](https://dl.acm.org/doi/10.1145/3423423.3423433)\n\n**Video:** [https://confirm.ie/covid_away/](https://confirm.ie/covid_away/)\n\n**WHO page** [https://search.bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov/resource/en/covidwho-901451](https://search.bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov/resource/en/covidwho-901451)\n\n### COVID-away Dataset\n\nAs shown below, we recorded the accelerometer, gyroscope, barometric pressure \\\u0026 rotation vector data for 2071 dynamic hand-to-face movements, performed with various postures (standing, leaning, slouching, etc.) and wrist orientations (variations in Roll, Pitch, and Yaw).\n\n![alt text](https://github.com/bharathsudharsan/COVID-away/blob/master/Covid-away_dataset_building.png)\n\n### Features Extractor\n\nWe provide a generic feature extractor for enabling users to extract 10 essential features (shown in below Table) from a single data field (dataset row) in any sensor-based motion dataset. Using this, we compute 102 features for each recorded hand-to-face motion data pattern.\n\n![alt text](https://github.com/bharathsudharsan/COVID-away/blob/master/Table1_feature_vectors.PNG)\n\n### COVID-away Models\n\nWe provide the beloy type models trained using the features extracted from our COVID-away Dataset. These models when deployed on smartwatches, instantly warn the users when their hands are moved (un-intentionally) to the face.\n\n- COVID-away One-Class Classification Models include:\n  -  One-Class SupportVector Machines (OC-SVM)\n  -  Isolation Forest (iForest)\n  -  Minimum Covariance Determinant (MCD)\n  -  Local Outlier Factor (LOF).\n- COVID-away CNNs and their model size \u0026 latency optimized versions\n\nIf the code is useful, please consider citing Covid-away paper using the below BibTex entry:\n\n```\n@inproceedings{Bharathcovidaway,\n  author    = {Bharath Sudharsan and John G. Breslin and Muhammad Intizar Ali},\n  title     = {Avoid Touching Your Face: A Hand-to-face 3D Motion Dataset (COVID-away) and Trained Models for Smartwatches},\n  booktitle = {In 10th International Conference on the Internet of Things Companion (IoT ’20 Companion)},\n  publisher = {ACM},\n  year      = {2020},\n  doi       = {10.1145/3423423.3423433},\n}\n```\n\nFor any clarification/further information please don't hesitate to contact me. Email: bharathsudharsan023@gmail.com\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbharathsudharsan%2Fcovid-away","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbharathsudharsan%2Fcovid-away","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbharathsudharsan%2Fcovid-away/lists"}