The "Gesture-Recognition-Conv3D-vs-ConvRNN" project is focused on gesture recognition using deep learning techniques. In this project, two different models, Conv3D and ConvRNN, were implemented and compared for their accuracy in recognizing gestures from video data. The objective was to determine which model performs better in this specific task.
- Conv3D Model The Conv3D model is a 3D convolutional neural network designed to process spatiotemporal data, making it suitable for video-based tasks like gesture recognition.
- ConvRNN Model The ConvRNN model combines convolutional layers with recurrent layers, creating a hybrid architecture. This allows the model to capture both spatial and temporal features within video sequences.
The project uses a dataset containing video clips of various hand gestures. These video clips are labeled with corresponding gesture categories.
After training and testing both models on the dataset, the project found that the Conv3D model outperformed the ConvRNN model in terms of accuracy when it came to gesture recognition.
This project is licensed under the MIT License. See the LICENSE file for details.
For any questions or issues, please feel free to contact.
Enjoy exploring gesture recognition with Conv3D and ConvRNN in your project!