Hand Gesture Detection

Hand Gesture Detection using Neural Network

Comment

Inspiration

The growing demand for Machine Learning and Neural Network like technologies made me do such a project also Neural Network being an important tech in today's world of technology made me do this project.

What it does

The project contains three files one for data collection, the second file is Neural Network model and the last one contains test file. The project has a Convolutional Neural Network which can be trained with a dataset and then it can be used for detecting different hand gestures.

How I built it

The first task was to create a huge dataset containing almost 100-130 images per label and then I had to augment the images so that the model gives proper results during prediction the next task is to create a Convolutional Neural Network model which could predict results on the basis of dataset provided for training the model, the last step was to create a code to detect and gestures.

Challenges I ran into

The major challenge was to create such a huge dataset for each of the labels given the next step was to create a Neural Network which was a major challenge for me as I had to select proper amount of layers for Convolutional Neural Network and then selecting other layers which included layers such as MaxPool, Dropouts layers and Dense layers and even selecting number of filters for Conv2D layers the final challenge was create a code that could detect hand gestures in real time using the web camera.

Accomplishments that I am proud of

The Accomplishments that I am proud of is that I was able to create the project in such a short time duration with an accuracy of 99.3% and also learning a few new technologies which could help me in long run.

What I learned

The major thing I learned that nothing is impossible even if it has do be done in a short duration and also I got to learn about many new technologies which I think could be very much helpful for me.

Built With

Share this project:

Updates