ComfyUI_BASICSAdvancedDancePoser

🕺 ComfyUI_BASICSAdvancedDancePoser

Professional COCO-WholeBody 133-keypoint dance animation system for ComfyUI

Transform any audio into synchronized dance animations with intelligent gesture recognition, musical facial expressions, and beat-responsive visual effects.

✨ Features

🎵 Musical Intelligence

• Beat Detection - Automatic rhythm analysis drives movement intensity
• Tonality Analysis - Major/minor key detection creates appropriate facial expressions
• Frequency Band Mapping - Bass, mids, and highs control different body parts
• BPM Analysis - Automatic tempo detection (60-200 BPM supported)

💃 Professional Gesture Library

• Victory Poses - Hands up triumphant, pump fists
• Classic Gestures - Peace signs, thumbs up, pointing, clapping
• Dance Moves - Jazz hands, disco pointing, running man, wave hands
• Power Poses - Arms crossed, hands on hips, reaching gestures
• Hand Articulation - Full 21-keypoint finger movement per hand

🎭 Facial Expression System

• Happy - Major keys + bright tones trigger smiles
• Sad - Minor keys + dark tones create contemplative expressions
• Surprised - High brightness creates wide-eyed looks
• Concentrated - Complex music triggers focused expressions

🦶 Coordinated Movement

• Foot Patterns - Step in place, marching, shuffling, bouncing
• Leg Coordination - Knees and ankles move naturally with feet
• Beat Tapping - Toe taps and heel strikes on strong beats
• Weight Distribution - Realistic balance and support leg adjustments

🎨 Professional Visualization

• Visual Hierarchy - Different colors and line weights for body parts
• Beat-Responsive Effects - Colors pulse and glow with music
• Dance Style Modifiers - Each style has unique visual personality
• Multiple Color Schemes - Black, white, and gray background support

🎯 Dance Styles

| Style | Characteristics | Gestures | Movement | |-------|----------------|----------|----------| | Energetic | High-energy, dynamic | Pump fists, victory poses, jazz hands | Quick transitions, beat emphasis | | Smooth | Flowing, graceful | Peace signs, gentle reaches, thumbs up | Fluid movements, sway | | Dramatic | Powerful, expressive | Victory poses, arms crossed, power gestures | Exaggerated movements | | Robot | Mechanical, precise | Pointing, structured poses | Quantized movements | | Bounce | Rhythmic, bouncy | Running man, disco points | Synchronized bouncing |

🚀 Quick Start

1. Install the node in your ComfyUI custom_nodes folder
2. Restart ComfyUI
3. Find the node: BASICS → Audio → 🕺 BASICS Advanced Dance Poser
4. Connect your audio input
5. Choose your dance style
6. Generate amazing dance animations!

📋 Requirements

• ComfyUI (latest version recommended)
• Python Libraries:
  • torch (for tensor operations)
  • numpy (for numerical computations)
  • librosa (for audio analysis)
  • opencv-python (for video rendering)

🎛️ Parameters

Core Settings

• audio - Input audio file (any ComfyUI audio format)
• start_time - Start position in seconds (0.0-3600.0)
• max_duration - Maximum clip length (3.0-180.0 seconds)
• width/height - Output video dimensions
• fps - Frame rate (12-60 fps)

Dance Control

• dance_style - Overall animation style (energetic, smooth, dramatic, robot, bounce)
• movement_intensity - Base movement strength (0.0-1.0)
• beat_sensitivity - How responsive to musical beats (0.0-1.0)
• tempo_factor - Speed modifier (0.25x, 0.5x, 1x, 2x)

Body Parts

• arm_movement_style - Arm animation type (flowing, sharp, waves, reaches, gestures)
• hand_gesture_style - Hand poses (auto, open_palm, peace_sign, rock_horns, etc.)
• leg_movement_style - Leg movement pattern (step, shuffle, march, bounce)
• foot_pattern_style - Foot coordination (auto, step_in_place, march_step, etc.)

Visual Effects

• enable_facial_expressions - Musical facial expressions on/off
• show_hands - Display hand details
• show_face_details - Display facial features
• show_visual_effects - Beat-responsive glow and pulse effects
• background_color - Black, white, or gray background

🏗️ Architecture

The system uses a modular architecture with specialized components:

ComfyUI_BASICSAdvancedDancePoser/
├── advanced_dance_poser.py        # Main node orchestrator
├── core/
│   ├── audio_analyzer.py          # Musical analysis
│   ├── skeleton_controller.py     # COCO-WholeBody structure
│   ├── facial_expression_driver.py # Tonality → expressions
│   ├── body_animation_driver.py   # Gesture library
│   └── constants.py               # COCO keypoint definitions
├── patterns/
│   └── foot_movement_patterns.py  # Coordinated foot/leg movement
└── renderers/
    └── pose_renderer.py           # Professional visualization

🎨 Technical Details

COCO-WholeBody Keypoints (133 total)

• Body: 17 keypoints (COCO standard)
• Feet: 6 keypoints (big toe, small toe, heel per foot)
• Face: 68 keypoints (facial landmarks)
• Hands: 42 keypoints (21 per hand)

Audio Analysis

• Beat Tracking - Librosa beat detection with frame-accurate timing
• Spectral Analysis - FFT-based frequency band separation
• Chroma Features - Musical key and tonality analysis
• Temporal Smoothing - Anti-aliasing for smooth animation

Animation System

• Gesture Templates - Pre-defined pose libraries with anatomical accuracy
• Blend Trees - Smooth transitions between poses
• Beat Synchronization - Musical timing drives gesture changes
• Coordinated Movement - Body parts work together naturally

🤝 Contributing

Contributions welcome! Areas for improvement:

• Additional gesture libraries
• New dance styles
• Enhanced facial expressions
• Performance optimizations
• Visual effects

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• COCO-WholeBody dataset for keypoint standards
• Librosa library for audio analysis capabilities
• ComfyUI community for the amazing framework
• OpenPose research for pose estimation foundations

🔧 Troubleshooting

Common Issues

"Audio too short" error

• Ensure audio is at least 3 seconds long
• Check start_time doesn't exceed audio duration

Missing dependencies

pip install librosa opencv-python numpy torch

Poor beat detection

• Try different beat_sensitivity values (0.1-0.3 range)
• Ensure audio has clear rhythmic content
• Check audio quality and format

Robotic movements

• Increase smoothing parameter (0.7-0.9)
• Lower movement_intensity
• Try "smooth" dance style

Performance Tips

• Use lower FPS (12-18) for faster generation
• Reduce max_duration for testing
• Disable visual_effects for better performance
• Use smaller output dimensions for drafts

📞 Support

For issues, feature requests, or questions:

• Open an issue on GitHub
• Check existing discussions
• Review troubleshooting section

---

Made with ❤️ for the ComfyUI community

Transform your audio into expressive dance animations with professional-grade skeletal animation and musical intelligence.