Dynamic LIBRAS Gesture Recognition via CNN over Spatiotemporal Matrix Representation

1. 1Extract 21 skeletal keypoints of the hand using MediaPipe Hand Landmarker
2. 2Create a spatiotemporal matrix representation of the keypoints
3. 3Train a CNN to classify gestures from the matrix representation
4. 4Evaluate the method on LIBRAS gesture recognition

This research benefits AI engineers and machine learning researchers working on computer vision and gesture recognition tasks, as it provides a novel approach to dynamic hand gesture recognition

💡 Combining MediaPipe Hand Landmarker with CNN and spatiotemporal matrix representation achieves effective dynamic hand gesture recognition

💡 Dynamic hand gesture recognition via CNN and spatiotemporal matrix representation for LIBRAS!