Haptics Final Project Presentation
Download as PPT, PDF0 likes654 views
The document summarizes a final project that used a sensor glove to classify hand postures and gestures for writing in the air. It discusses using a neural network classifier with the CyberGlove device to accurately classify two postures. Templates were collected from 3 users for 4 sketch gestures and tested consecutively. The system was able to accurately classify some of the gesture sequences but failed on others, leading to the conclusion that the $1 classifier was insufficient.
Haptics Final Project Presentation
- 1. Haptics Final Project: Using a Sensor Glove to Write in the Air Paul Taele Spring 2008
- 2. Goals Write stuff in the air without a pen.
- 4. Original Posture Classifier Setup Tools: P5 CyberGlove Posture Classifiers: k-Nearest Neighbor Naïve Bayes Neural Network
- 5. Postures - Results P5 NB: 10% kNN: 50% NN: 70% CyberGlove NN: 75% (all 23 sensors) NN: 100% (3 index finger sensors)
- 6. Postures - Analysis Desired 100% for posture classification. Used CyberGlove device and Neural Network classifier for postures. Used two easily separable gestures instead of four.
- 7. Hand Gesture Segmentation Simple for two very separable gestures. Classify each time state in an instance using the trained NN.
- 8. Final Project Setup Tools: CyberGlove Language: Java Posture Classifier: Neural Network Sketch Classifier: $1 # of Postures: 2 # of Gestures: 4
- 10. Final Gestures
- 11. Training Data ($1) Created templates from 3 users. Each user gave 5 examples for each sketch gesture.
- 16. Test Data ($1) Data was tested on consecutively-inputted sketch gesture. Postures first extracted from gesturing stream. Time points of those postures used to classify sketch gestures.
- 17. Target: Circle -> Triangle Actual: Rectangle -> Triangle
- 18. Target: Rectangle -> Circle Actual: Rectangle -> Rectangle
- 19. Target: Triangle -> X Actual: X -> X
- 20. Target: X -> Rectangle Actual: Rectangle -> X