Gesture_Recognition_Jester_Dmo

This is demo of dynamic gesture recognition.

• Deep learning model: ResNet3D
• Dataset: 20BN-jester
• Pretrained Model: Uploaded In Google Drive
• Environment:Python 3.7 Keras 2.2.4 TensorFlow 1.13.1 opencv-python 4.1.2.30
• Accuracy: 85.49% on validation

Annaconda Environment File is listed above as environment.yaml. For installing conda environment, please use

conda env create -f environment.yml

Video Presentation

Please have a look at videos in Folder Jester_video_demo to see how far did I go. The names of actions in each video are listed on these .mp4 names.

Attention

Please make sure a camera is avaible on your laptop/PC.

For any question, please feel free to pull your problem. My gmail is [email protected]