K-Groove

# Getting Started To get started with K-Groove, follow these steps: 1. Install Visual Studio Code (VSCode) on your computer. 2. Install the Jupyter Notebook extension for VSCode. 3. Locate the 'expert-dance.mp4' file, which contains a portion of NewJeans' "How Sweet" dance. 4. Practice the choreography shown in the video, focusing on the approximately 19-second segment. 5. Record your own dance performance, mimicking the choreography from the expert video. 6. Save your dance video as 'user_dance.mp4' and place it in the same directory as the project files. 7. Set up the GEMINI_API_KEY environment variable (see instructions below). 8. Open and run the 'Untitled-1.ipynb' file in VSCode. ## Setting up the GEMINI_API_KEY environment variable Before running the application, you need to set up the GEMINI_API_KEY environment variable. Follow the instructions for your operating system: ### Windows: 1. Open the Start menu and search for "Environment Variables". 2. Click on "Edit the system environment variables". 3. In the System Properties window, click on the "Environment Variables" button. 4. Under "User variables", click "New". 5. Set the Variable name as GEMINI_API_KEY and the Variable value as your actual Gemini API key. 6. Click "OK" to save. Alternatively, you can use the command prompt: ### macOS and Linux: 1. Open your terminal. 2. Open your shell configuration file (e.g., ~/.bash_profile, ~/.zshrc, or ~/.bashrc) in a text editor. 3. Add the following line at the end of the file: export GEMINI_API_KEY=<your-api-key-here> 4. Save the file and exit the editor. 5. Run the following command to apply the changes: ## Creating Pose Images and Videos If you want to generate pose images and videos: 1. Run the 'Untitled-2.ipynb' file in VSCode. 2. This will create image files in the expert and user folders. 3. It will also generate pose videos: 'expert_pose.mp4' and 'user_pose.mp4'. Note: Errors may occur during image file generation. If an error occurs after creating expert image files, follow these steps: - Restart the kernel - Run from the beginning, but skip this part: This is python code: with create_pose_landmarker() as landmarker: print("Processing expert video...") user_poses = process_video(expert_video, landmarker, show_first_n_frames=720, path="expert")