Human-Computer Interaction (HCI) has seen advancements and improvements since its development but still, major everyday interaction with machines is limited.Hand gesture Recognition system is a way for humans to interact with machines/computer by simply using our hand gestures This real-time hand gesture recognition system's main goal is to detect and recognise hand motions in real-time. Built using CNN leveraging TensorFlow and Keras for computation.
Future Work: The video sequences consist of both spatial and temporal features and using just CNN won’t be enough to capture both features, therefore we use both ConvNet and LSTM(Long short term memory) LSTMs, which are a variation of recurrent neural networks, RNNs. RNNs are built to process input sequences in which all of the inputs are connected to one another. The loop is a crucial component of RNN architecture since it can retain previous information in the internal state memory. This is done to identify both spatial and temporal features that have both short and long term information.
Built With
- cnn
- dl
- keras
- python
- pythonlibraries
- tenserflow
Log in or sign up for Devpost to join the conversation.