What question did this study set out to answer?

To improve the recognition accuracy of complex dynamic gestures captured by Kinect sensors for Internet of Things human-computer interaction.

December 18, 2018IEEE AccessOpen Access

Dynamic Gesture Recognition in the Internet of Things

View Full Paper

Ask AI

Bookmark

To improve the recognition accuracy of complex dynamic gestures captured by Kinect sensors for Internet of Things human-computer interaction.
Extracted tangent angles and morphological variations across palm trajectory timestamps from Kinect depth sensor streams as primary motion features.
Reduced trajectory tangent dimensionality via quantization codes before conducting Hidden Markov Model (HMM) parameter training.
Applied Dempster-Shafer (D-S) evidence theory to execute combinatorial logic judgments on model outputs for final dynamic gesture classification.
Achieved enhanced dynamic gesture recognition rates by combining palm trajectory tangent modeling with evidence-theory logic judgment.
Demonstrated improved classification capability for complex dynamic hand movements under variable illumination and cluttered background conditions.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Algorithm development study demonstrates improved dynamic gesture recognition for Kinect sensors, indicating enhanced human-computer interaction in IoT systems.

Cite This Study

Li et al. (2018) studied this question.

synapsesocial.com/papers/6a0c058995872b300be87b74 https://doi.org/10.1109/access.2018.2887223

View Full Paper

Ask AI

Bookmark

Implication

Algorithm development study demonstrates improved dynamic gesture recognition for Kinect sensors, indicating enhanced human-computer interaction in IoT systems.

Dynamic Gesture Recognition in the Internet of Things

Key Points

Discussion

Implication

Cite This Study

Dynamic Gesture Recognition in the Internet of Things

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider