Skip to content

Hand gesture recognition using CNNs. Using the dataset to train the best known network for gesture classification on images with just one hand. Testing hyper-parameters and preprocessing techniques to improve the performance.

Notifications You must be signed in to change notification settings

mahadmughal/Arabic-hand-gestures-classification

Repository files navigation

Arabic-hand-gestures-classification

Hand gesture recognition using CNNs. Using the dataset to train the best known network for gesture classification on images with just one hand. Testing hyper-parameters and preprocessing techniques to improve the performance.

Steps to run the code:

1)Place the dataset directory namely 'augmented_dataset' in the root directory of the project. 2)Run the command 'python3 main.py' from the terminal window OR run the 'main.py' in any IDE to kick-start the code execution.

NOTE: 'data_augmentation.py' is only executed when you don't have the augmented dataset. It is more important to note that the names of dataset directories or folders must match with the names defined in the code.

Thank.

About

Hand gesture recognition using CNNs. Using the dataset to train the best known network for gesture classification on images with just one hand. Testing hyper-parameters and preprocessing techniques to improve the performance.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published