NuActiv: recognizing unseen new activities using semantic attribute-based learning. Cheng, H., Sun, F., Griss, M., Davis, P., Li, J., & You, D. In Proceeding of the 11th annual international conference on Mobile systems, applications, and services, of MobiSys '13, pages 361–374, New York, NY, USA, June, 2013. Association for Computing Machinery. ![NuActiv: recognizing unseen new activities using semantic attribute-based learning [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex We study the problem of how to recognize a new human activity when we have never seen any training example of that activity before. Recognizing human activities is an essential element for user-centric and context-aware applications. Previous studies showed promising results using various machine learning algorithms. However, most existing methods can only recognize the activities that were previously seen in the training data. A previously unseen activity class cannot be recognized if there were no training samples in the dataset. Even if all of the activities can be enumerated in advance, labeled samples are often time consuming and expensive to get, as they require huge effort from human annotators or experts. In this paper, we present NuActiv, an activity recognition system that can recognize a human activity even when there are no training data for that activity class. Firstly, we designed a new representation of activities using semantic attributes, where each attribute is a human readable term that describes a basic element or an inherent characteristic of an activity. Secondly, based on this representation, a two-layer zero-shot learning algorithm is developed for activity recognition. Finally, to reinforce recognition accuracy using minimal user feedback, we developed an active learning algorithm for activity recognition. Our approach is evaluated on two datasets, including a 10-exercise-activity dataset we collected, and a public dataset of 34 daily life activities. Experimental results show that using semantic attribute-based learning, NuActiv can generalize knowledge to recognize unseen new activities. Our approach achieved up to 79% accuracy in unseen activity recognition.
@inproceedings{cheng_nuactiv_2013,
address = {New York, NY, USA},
series = {{MobiSys} '13},
title = {{NuActiv}: recognizing unseen new activities using semantic attribute-based learning},
isbn = {978-1-4503-1672-9},
shorttitle = {{NuActiv}},
url = {https://doi.org/10.1145/2462456.2464438},
doi = {10.1145/2462456.2464438},
abstract = {We study the problem of how to recognize a new human activity when we have never seen any training example of that activity before. Recognizing human activities is an essential element for user-centric and context-aware applications. Previous studies showed promising results using various machine learning algorithms. However, most existing methods can only recognize the activities that were previously seen in the training data. A previously unseen activity class cannot be recognized if there were no training samples in the dataset. Even if all of the activities can be enumerated in advance, labeled samples are often time consuming and expensive to get, as they require huge effort from human annotators or experts. In this paper, we present NuActiv, an activity recognition system that can recognize a human activity even when there are no training data for that activity class. Firstly, we designed a new representation of activities using semantic attributes, where each attribute is a human readable term that describes a basic element or an inherent characteristic of an activity. Secondly, based on this representation, a two-layer zero-shot learning algorithm is developed for activity recognition. Finally, to reinforce recognition accuracy using minimal user feedback, we developed an active learning algorithm for activity recognition. Our approach is evaluated on two datasets, including a 10-exercise-activity dataset we collected, and a public dataset of 34 daily life activities. Experimental results show that using semantic attribute-based learning, NuActiv can generalize knowledge to recognize unseen new activities. Our approach achieved up to 79\% accuracy in unseen activity recognition.},
urldate = {2021-09-10},
booktitle = {Proceeding of the 11th annual international conference on {Mobile} systems, applications, and services},
publisher = {Association for Computing Machinery},
author = {Cheng, Heng-Tze and Sun, Feng-Tso and Griss, Martin and Davis, Paul and Li, Jianguo and You, Di},
month = jun,
year = {2013},
keywords = {active learning, activity recognition, context-aware computing, machine learning, mobile sensing, semantic attributes, wearable computing, zero-shot learning},
pages = {361--374},
}
Downloads: 0
{"_id":"uPxToaX7j8vXDxYog","bibbaseid":"cheng-sun-griss-davis-li-you-nuactivrecognizingunseennewactivitiesusingsemanticattributebasedlearning-2013","author_short":["Cheng, H.","Sun, F.","Griss, M.","Davis, P.","Li, J.","You, D."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"New York, NY, USA","series":"MobiSys '13","title":"NuActiv: recognizing unseen new activities using semantic attribute-based learning","isbn":"978-1-4503-1672-9","shorttitle":"NuActiv","url":"https://doi.org/10.1145/2462456.2464438","doi":"10.1145/2462456.2464438","abstract":"We study the problem of how to recognize a new human activity when we have never seen any training example of that activity before. Recognizing human activities is an essential element for user-centric and context-aware applications. Previous studies showed promising results using various machine learning algorithms. However, most existing methods can only recognize the activities that were previously seen in the training data. A previously unseen activity class cannot be recognized if there were no training samples in the dataset. Even if all of the activities can be enumerated in advance, labeled samples are often time consuming and expensive to get, as they require huge effort from human annotators or experts. In this paper, we present NuActiv, an activity recognition system that can recognize a human activity even when there are no training data for that activity class. Firstly, we designed a new representation of activities using semantic attributes, where each attribute is a human readable term that describes a basic element or an inherent characteristic of an activity. Secondly, based on this representation, a two-layer zero-shot learning algorithm is developed for activity recognition. Finally, to reinforce recognition accuracy using minimal user feedback, we developed an active learning algorithm for activity recognition. Our approach is evaluated on two datasets, including a 10-exercise-activity dataset we collected, and a public dataset of 34 daily life activities. Experimental results show that using semantic attribute-based learning, NuActiv can generalize knowledge to recognize unseen new activities. Our approach achieved up to 79% accuracy in unseen activity recognition.","urldate":"2021-09-10","booktitle":"Proceeding of the 11th annual international conference on Mobile systems, applications, and services","publisher":"Association for Computing Machinery","author":[{"propositions":[],"lastnames":["Cheng"],"firstnames":["Heng-Tze"],"suffixes":[]},{"propositions":[],"lastnames":["Sun"],"firstnames":["Feng-Tso"],"suffixes":[]},{"propositions":[],"lastnames":["Griss"],"firstnames":["Martin"],"suffixes":[]},{"propositions":[],"lastnames":["Davis"],"firstnames":["Paul"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Jianguo"],"suffixes":[]},{"propositions":[],"lastnames":["You"],"firstnames":["Di"],"suffixes":[]}],"month":"June","year":"2013","keywords":"active learning, activity recognition, context-aware computing, machine learning, mobile sensing, semantic attributes, wearable computing, zero-shot learning","pages":"361–374","bibtex":"@inproceedings{cheng_nuactiv_2013,\n\taddress = {New York, NY, USA},\n\tseries = {{MobiSys} '13},\n\ttitle = {{NuActiv}: recognizing unseen new activities using semantic attribute-based learning},\n\tisbn = {978-1-4503-1672-9},\n\tshorttitle = {{NuActiv}},\n\turl = {https://doi.org/10.1145/2462456.2464438},\n\tdoi = {10.1145/2462456.2464438},\n\tabstract = {We study the problem of how to recognize a new human activity when we have never seen any training example of that activity before. Recognizing human activities is an essential element for user-centric and context-aware applications. Previous studies showed promising results using various machine learning algorithms. However, most existing methods can only recognize the activities that were previously seen in the training data. A previously unseen activity class cannot be recognized if there were no training samples in the dataset. Even if all of the activities can be enumerated in advance, labeled samples are often time consuming and expensive to get, as they require huge effort from human annotators or experts. In this paper, we present NuActiv, an activity recognition system that can recognize a human activity even when there are no training data for that activity class. Firstly, we designed a new representation of activities using semantic attributes, where each attribute is a human readable term that describes a basic element or an inherent characteristic of an activity. Secondly, based on this representation, a two-layer zero-shot learning algorithm is developed for activity recognition. Finally, to reinforce recognition accuracy using minimal user feedback, we developed an active learning algorithm for activity recognition. Our approach is evaluated on two datasets, including a 10-exercise-activity dataset we collected, and a public dataset of 34 daily life activities. Experimental results show that using semantic attribute-based learning, NuActiv can generalize knowledge to recognize unseen new activities. Our approach achieved up to 79\\% accuracy in unseen activity recognition.},\n\turldate = {2021-09-10},\n\tbooktitle = {Proceeding of the 11th annual international conference on {Mobile} systems, applications, and services},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Cheng, Heng-Tze and Sun, Feng-Tso and Griss, Martin and Davis, Paul and Li, Jianguo and You, Di},\n\tmonth = jun,\n\tyear = {2013},\n\tkeywords = {active learning, activity recognition, context-aware computing, machine learning, mobile sensing, semantic attributes, wearable computing, zero-shot learning},\n\tpages = {361--374},\n}\n\n","author_short":["Cheng, H.","Sun, F.","Griss, M.","Davis, P.","Li, J.","You, D."],"key":"cheng_nuactiv_2013","id":"cheng_nuactiv_2013","bibbaseid":"cheng-sun-griss-davis-li-you-nuactivrecognizingunseennewactivitiesusingsemanticattributebasedlearning-2013","role":"author","urls":{"Paper":"https://doi.org/10.1145/2462456.2464438"},"keyword":["active learning","activity recognition","context-aware computing","machine learning","mobile sensing","semantic attributes","wearable computing","zero-shot learning"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["active learning","activity recognition","context-aware computing","machine learning","mobile sensing","semantic attributes","wearable computing","zero-shot learning"],"search_terms":["nuactiv","recognizing","unseen","new","activities","using","semantic","attribute","based","learning","cheng","sun","griss","davis","li","you"],"title":"NuActiv: recognizing unseen new activities using semantic attribute-based learning","year":2013}