Wearable Activity Recognition with Crowdsourced Annotation

  title={Wearable Activity Recognition with Crowdsourced Annotation},
  author={Long-Van Nguyen-Dinh},
xiii Zusammenfassung xvii 


Wearable Activity Tracking in Car Manufacturing
A context-aware wearable computing system could support a production or maintenance worker by recognizing the worker's actions and delivering just-in-time information about activities to be performed.
Weakly Supervised Recognition of Daily Life Activities with Wearable Sensors
Two learning schemes for activity recognition are explored that effectively leverage such sparsely labeled data together with more easily obtainable unlabeled data and are robust to the presence of erroneous labels occurring in real-world annotation data.
Toward wearable social networking with iBand
The iBand technology and feedback from an initial study suggest that control over personal information is an ongoing issue, but they highlight the possibility for wearable devices to enable the creation of a set of invented techno-gestures with different affordances and constraints that might be more appropriate for certain social interaction applications.
Robust online gesture recognition with crowdsourced annotations
SegmentedLCSS and WarpingLCSS are presented, two template-matching methods offering robustness when trained with noisy crowdsourced annotations to spot gestures from wearable motion sensors, and to use their methods to filter out the noise in the crowdsourced annotation before training a traditional classifier.
Towards a unified system for multimodal activity spotting: challenges and a proposal
A unified system which works with any available wearable sensors placed on user's body to spot activities and is compatible with respect to modalities and body-worn positions is proposed.
Activity Recognition from Accelerometer Data
This paper reports on the efforts to recognize user activity from accelerometer data and performance of base-level and meta-level classifiers, and Plurality Voting is found to perform consistently well across different settings.
Tagging human activities in video by crowdsourcing
This work presents for the first time an approach to achieve good quality for activity annotation in videos through crowdsourcing on the AmazonMechanical Turk platform (AMT), and shows that the proposed filtering strategies can increase the accuracy by up to 40%.
Online Context Recognition in Multisensor Systems using Dynamic Time Warping
The system uses Dynamic Time Warping (DTW) to recognize multimodal sequences of different lengths, embedded in continuous data streams, using accelerometer data acquired from performing two hand gestures and NOKIA's benchmark dataset for context recognition.
Video Annotation and Tracking with Active Learning
A novel active learning framework for video annotation is introduced, and it is shown that it can obtain excellent performance by querying frames that, if annotated, would produce a large expected change in the estimated object track.