Dataset Information

MGRA: Motion Gesture Recognition via Accelerometer.

ABSTRACT: Accelerometers have been widely embedded in most current mobile devices, enabling easy and intuitive operations. This paper proposes a Motion Gesture Recognition system (MGRA) based on accelerometer data only, which is entirely implemented on mobile devices and can provide users with real-time interactions. A robust and unique feature set is enumerated through the time domain, the frequency domain and singular value decomposition analysis using our motion gesture set containing 11,110 traces. The best feature vector for classification is selected, taking both static and mobile scenarios into consideration. MGRA exploits support vector machine as the classifier with the best feature vector. Evaluations confirm that MGRA can accommodate a broad set of gesture variations within each class, including execution time, amplitude and non-gestural movement. Extensive evaluations confirm that MGRA achieves higher accuracy under both static and mobile scenarios and costs less computation time and energy on an LG Nexus 5 than previous methods.

SUBMITTER: Hong F

PROVIDER: S-EPMC4851044 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

MGRA: Motion Gesture Recognition via Accelerometer.

Hong Feng F You Shujuan S Wei Meiyu M Zhang Yongtuo Y Guo Zhongwen Z

Sensors (Basel, Switzerland) 20160413 4

Accelerometers have been widely embedded in most current mobile devices, enabling easy and intuitive operations. This paper proposes a Motion Gesture Recognition system (MGRA) based on accelerometer data only, which is entirely implemented on mobile devices and can provide users with real-time interactions. A robust and unique feature set is enumerated through the time domain, the frequency domain and singular value decomposition analysis using our motion gesture set containing 11,110 traces. Th ...[more]

PMID: 27089336

Similar Datasets

Project description:A mixed design was created using text and game-like multimedia to instruct in the content of physics. The study assessed which variables predicted learning gains after a 1-h lesson on the electric field. The three manipulated variables were: (1) level of embodiment; (2) level of active generativity; and (3) presence of story narrative. Two types of tests were administered: (1) a traditional text-based physics test answered with a keyboard; and (2) a more embodied, transfer test using the Wacom large tablet where learners could use gestures (long swipes) to create vectors and answers. The 166 participants were randomly assigned to four conditions: (1) symbols and text; (2) low embodied; (3) high embodied/active; or (4) high embodied/active with narrative. The last two conditions were active because the on-screen content could be manipulated with gross body gestures gathered via the Kinect sensor. Results demonstrated that the three groups that included embodiment learned significantly more than the symbols and text group on the traditional keyboard post-test. When knowledge was assessed with the Wacom tablet format that facilitated gestures, the two active gesture-based groups scored significantly higher. In addition, engagement scores were significantly higher for the two active embodied groups. The Wacom results suggest test sensitivity issues; the more embodied test revealed greater gains in learning for the more embodied conditions. We recommend that as more embodied learning comes to the fore, more sensitive tests that incorporate gesture be used to accurately assess learning. The predicted differences in engagement and learning for the condition with the graphically rich story narrative were not supported. We hypothesize that a narrative effect for motivation and learning may be difficult to uncover in a lab experiment where participants are primarily motivated by course credit. Several design principles for mediated and embodied science education are proposed.

Dataset Information

MGRA: Motion Gesture Recognition via Accelerometer.

Publications

MGRA: Motion Gesture Recognition via Accelerometer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets