ComfyUI_SwiftCut

A simple ComfyUI plugin that Its purpose and function is to replicate some of the editing effects of capcut,jianying and pr.

📦 Installation

1. Clone this repository into your ComfyUI custom nodes directory:

cd ComfyUI/custom_nodes
git clone https://github.com/your-username/ComfyUI_SwiftCut.git

2. Install required dependencies:

pip install torch numpy pillow

3. Restart ComfyUI

https://github.com/user-attachments/assets/d353c911-2364-4d81-b6d8-7a28b06c8ce4

🛠️ Available Nodes

Image Processing Nodes

🎬 ImagePushPullLens

Category: SwiftCut/Image

Creates cinematic push-pull lens effects (dolly zoom) from single or multiple images.

Features:

• Single to Multi Mode: Generate multiple frames from a single image
• Multi to Multi Mode: Apply different crop ratios to existing frame sequences
• Three-stage Animation: Customizable start, middle, and end crop ratios
• Center-based Cropping: Maintains subject focus during zoom effects

Inputs:

• image: Input image(s)
• frames: Number of output frames (2-120)
• start_crop_ratio: Initial crop ratio (0.1-1.0)
• middle_crop_ratio: Middle point crop ratio (0.1-1.0)
• end_crop_ratio: Final crop ratio (0.1-1.0)
• middle_frame: Frame where middle ratio is reached
• input_mode: Processing mode selection

Outputs:

• frames: Processed image sequence
• masks: Corresponding crop masks

🎨 ImageOverlayBlend

Category: SwiftCut/Image

Advanced multi-frame blending with background overlay and transparency control.

Features:

• Dynamic Alpha Control: Three-stage transparency animation
• Multiple Blend Modes: Normal, Multiply, Screen, Overlay, Soft Light
• Custom Background: Hex color background support
• Professional Color Mixing: Industry-standard blend algorithms

Inputs:

• frames: Input frame sequence
• background_color: Hex color (e.g., "#000000")
• start_alpha_ratio: Initial transparency (0.0-1.0)
• middle_alpha_ratio: Middle transparency (0.0-1.0)
• end_alpha_ratio: Final transparency (0.0-1.0)
• middle_frame: Transition point frame
• blend_mode: Blending algorithm selection

Outputs:

• blended_frames: Processed frame sequence
• alpha_masks: Transparency masks

🔀 ImageBatchBlend

Category: SwiftCut/Image

High-performance batch image blending for complex compositions.

Features:

• Automatic Channel Matching: RGB/RGBA/Grayscale compatibility
• Frame Synchronization: Automatic length matching
• Professional Blend Modes: 7 industry-standard modes
• Batch Processing: Optimized for large sequences

Inputs:

• image1: First image batch
• image2: Second image batch
• blend_factor: Blend strength (0.0-1.0)
• blend_mode: Blending algorithm

Outputs:

• blended_images: Final composited sequence

Utility Nodes

📋 SelectImages

Category: SwiftCut/Utils

Precise image selection from batches with advanced indexing.

Features:

• Flexible Indexing: Single, range, and step selections
• Negative Indexing: Reverse selection support
• Error Handling: Configurable bounds checking

📋 SelectImagesAdvanced

Category: SwiftCut/Utils

Extended selection capabilities with complex pattern support.

Features:

• Complex Patterns: Mixed index formats
• Batch Processing: Multiple selection operations
• Safety Controls: Comprehensive error handling

If you benefit from this project, please buy the author a cup of coffee.

<img width="1536" height="841" alt="未标题-2" src="https://github.com/user-attachments/assets/a562fcfd-cb65-4695-a2a8-ca3eb6a6db67" />