Deep learning has substantially improved the accuracy of<\/p>\n
Human Activity Recognition (HAR) from wearable sensors, but existing<\/p>\n
pipelines mostly rely on fixed, high-frequency sampling strategies for the<\/p>\n
inertial sensors. Such designs mostly lead to unnecessary sensing, com-<\/p>\n
putation, and energy consumption, for instance during low-motion or<\/p>\n
repetitive activities. We show in this work that many activities in cur-<\/p>\n
rent benchmarks can be recognized accurately at much lower sampling<\/p>\n
rates, while others do benefit from higher temporal resolution. We then<\/p>\n
introduce a frequency-adaptive HAR framework that supports runtime<\/p>\n
selection of the sampling rate most appropriate for each activity. The<\/p>\n
framework is enabled by a single-checkpoint multirate ContinuousCon-<\/p>\n
vLSTM model, whose continuous convolution kernels adjust their effec-<\/p>\n
tive temporal support as a function of sampling frequency. We evaluate<\/p>\n
ContinuousConvLSTM and quantify the achievable accuracy\u2013efficiency<\/p>\n
trade-offs relative to fixed-frequency baselines.<\/p>\n","protected":false},"excerpt":{"rendered":"
Deep learning has substantially improved the accuracy of Human Activity Recognition (HAR) from wearable sensors, but existing pipelines mostly rely on fixed, high-frequency sampling strategies for the inertial sensors. Such designs mostly lead to unnecessary sensing, com- putation, and energy consumption, for instance during low-motion or repetitive activities. We show in this work that many activities in cur- rent benchmarks can be recognized accurately at much lower sampling rates, while […]<\/p>\n","protected":false},"author":14,"featured_media":0,"template":"","meta":{"_acf_changed":false,"footnotes":""},"publication-type":[32],"publication-year":[1680],"class_list":["post-43799","publication","type-publication","status-publish","hentry","publication-type-inproceedings","publication-year-1680"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/43799","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication"}],"about":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/types\/publication"}],"author":[{"embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/users\/14"}],"version-history":[{"count":0,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/43799\/revisions"}],"wp:attachment":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/media?parent=43799"}],"wp:term":[{"taxonomy":"publication-type","embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication-type?post=43799"},{"taxonomy":"publication-year","embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication-year?post=43799"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}