{"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/audio-based-activities-of-daily-living-adl","title":"Audio-Based Activities of Daily Living (ADL) Recognition with Large-Scale Acoustic Embeddings from Online Videos","arxiv_id":"1810.08691","date":"2018-10-19","proceeding":null,"authors":["Dawei Liang","Edison Thomaz"],"abstract":"Over the years, activity sensing and recognition has been shown to play a key\nenabling role in a wide range of applications, from sustainability and\nhuman-computer interaction to health care. While many recognition tasks have\ntraditionally employed inertial sensors, acoustic-based methods offer the\nbenefit of capturing rich contextual information, which can be useful when\ndiscriminating complex activities. Given the emergence of deep learning\ntechniques and leveraging new, large-scaled multi-media datasets, this paper\nrevisits the opportunity of training audio-based classifiers without the\nonerous and time-consuming task of annotating audio data. We propose a\nframework for audio-based activity recognition that makes use of millions of\nembedding features from public online video sound clips. Based on the\ncombination of oversampling and deep learning approaches, our framework does\nnot require further feature processing or outliers filtering as in prior work.\nWe evaluated our approach in the context of Activities of Daily Living (ADL) by\nrecognizing 15 everyday activities with 14 participants in their own homes,\nachieving 64.2% and 83.6% averaged within-subject accuracy in terms of top-1\nand top-3 classification respectively. Individual class performance was also\nexamined in the paper to further study the co-occurrence characteristics of the\nactivities and the robustness of the framework.","url_abs":"http://arxiv.org/abs/1810.08691v2","url_pdf":"http://arxiv.org/pdf/1810.08691v2.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":"audio-based-activities-of-daily-living-adl","repo_url":"https://github.com/dawei-liang/AudioAR_Research_Codes","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"activity-recognition","task_name":"Activity Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}