SPATIAL POSTURE ANALYSIS :
Automate the recognition of sign language gestures using human gestures and convert them into speech. It can be trained to recognize any sign language, even without a pre-existing dataset. It uses: Camera input to capture gestures. Pre-processing techniques (median and Gaussian filters) to reduce noise. HSV color space to distinguish hand from background. Histogram equalization and hand segmentation using a 5×10 box approach. TensorFlow object detection API for object identification. Convolutional Neural Networks (CNNs) to automate feature learning and recognition. The system achieves good accuracy and is capable of recognizing 46 gestures.
https://youtube.com/shorts/mTcm9RsAaZc?si=gKIsaKt8zzZ9544j
No comments:
Post a Comment