CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition
Existing activity tracker datasets for human activity recognition are typically obtained by having participants perform predefined activities in an enclosed environment under supervision. This results in small datasets with a limited number of activities and heterogeneity, lacking the mixed and nuan...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Nature Research
2024
|
| _version_ | 1817931000491016192 |
|---|---|
| author | Chan, S Hang, Y Tong, C Acquah, A Schonfeldt, A Gershuny, J Doherty, A |
| author_facet | Chan, S Hang, Y Tong, C Acquah, A Schonfeldt, A Gershuny, J Doherty, A |
| author_sort | Chan, S |
| collection | OXFORD |
| description | Existing activity tracker datasets for human activity recognition are typically obtained by having participants perform predefined activities in an enclosed environment under supervision. This results in small datasets with a limited number of activities and heterogeneity, lacking the mixed and nuanced movements normally found in free-living scenarios. As such, models trained on laboratory-style datasets may not generalise out of sample. To address this problem, we introduce a new dataset involving wrist-worn accelerometers, wearable cameras, and sleep diaries, enabling data collection for over 24 hours in a free-living setting. The result is CAPTURE-24, a large activity tracker dataset collected in the wild from 151 participants, amounting to 3883 hours of accelerometer data, of which 2562 hours are annotated. CAPTURE-24 is two to three orders of magnitude larger than existing publicly available datasets, which is critical to developing accurate human activity recognition models. |
| first_indexed | 2024-12-09T03:15:04Z |
| format | Journal article |
| id | oxford-uuid:b93ae246-a5bf-4be8-97bb-5af394d8a3d6 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-12-09T03:15:04Z |
| publishDate | 2024 |
| publisher | Nature Research |
| record_format | dspace |
| spelling | oxford-uuid:b93ae246-a5bf-4be8-97bb-5af394d8a3d62024-10-16T20:10:27ZCAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognitionJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:b93ae246-a5bf-4be8-97bb-5af394d8a3d6EnglishJisc Publications RouterNature Research2024Chan, SHang, YTong, CAcquah, ASchonfeldt, AGershuny, JDoherty, AExisting activity tracker datasets for human activity recognition are typically obtained by having participants perform predefined activities in an enclosed environment under supervision. This results in small datasets with a limited number of activities and heterogeneity, lacking the mixed and nuanced movements normally found in free-living scenarios. As such, models trained on laboratory-style datasets may not generalise out of sample. To address this problem, we introduce a new dataset involving wrist-worn accelerometers, wearable cameras, and sleep diaries, enabling data collection for over 24 hours in a free-living setting. The result is CAPTURE-24, a large activity tracker dataset collected in the wild from 151 participants, amounting to 3883 hours of accelerometer data, of which 2562 hours are annotated. CAPTURE-24 is two to three orders of magnitude larger than existing publicly available datasets, which is critical to developing accurate human activity recognition models. |
| spellingShingle | Chan, S Hang, Y Tong, C Acquah, A Schonfeldt, A Gershuny, J Doherty, A CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition |
| title | CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition |
| title_full | CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition |
| title_fullStr | CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition |
| title_full_unstemmed | CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition |
| title_short | CAPTURE-24: A large dataset of wrist-worn activity tracker data collected in the wild for human activity recognition |
| title_sort | capture 24 a large dataset of wrist worn activity tracker data collected in the wild for human activity recognition |
| work_keys_str_mv | AT chans capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition AT hangy capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition AT tongc capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition AT acquaha capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition AT schonfeldta capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition AT gershunyj capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition AT dohertya capture24alargedatasetofwristwornactivitytrackerdatacollectedinthewildforhumanactivityrecognition |