🌀 ComfyUI Zoom Sequence Node

A ComfyUI custom node that performs smooth, aspect-preserving Zoom In and Zoom Out effects over image sequences or video frames.
It maintains the original canvas size and aspect ratio while applying customizable zoom speed and easing curves.

---

✨ Features

• 🎞️ Works with videos, image sequences, or image lists
• 🔍 True aspect-preserving Zoom In / Zoom Out
• ⚙️ Configurable Pixels per frame (supports fractional values, e.g. 1.5)
• 🧩 Smooth easing options: Linear, Ease_In, Ease_Out, Ease_In_Out
• 🖼️ Canvas size and proportions always stay the same
• ⚡ GPU-accelerated with PyTorch (no extra dependencies)

---

🧩 Inputs & Outputs

Inputs

| Name | Type | Description | |------|------|-------------| | images | IMAGE | Sequence of frames ([B, H, W, C]) from a video or image list | | mode | Zoom In / Zoom Out | Zoom direction. Zoom In progressively crops inward; Zoom Out reverses it | | pixels_per_frame | FLOAT | Number of pixels per frame (per-side) on the smaller image dimension. Supports fractional values (e.g. 1.5) for finer control | | ease | STRING | Easing curve controlling zoom speed over time: Linear, Ease_In, Ease_Out, or Ease_In_Out |

---

Outputs

| Name | Type | Description | |------|------|-------------| | images | IMAGE | Sequence of zoomed frames (same dimensions as input) | | info | STRING | Text summary of zoom parameters, margins, and aspect adjustments |

---

⚙️ Installation

1. Go to your ComfyUI/custom_nodes/ folder.
2. Create a new folder, e.g.:

   ComfyUI/custom_nodes/ComfyZoomSequence/
   

3. Copy the following files into it:

   comfy_zoom_sequence.py
   __init__.py
   requirements.txt
   

4. Restart ComfyUI.

Folder structure

ComfyUI/
└─ custom_nodes/
   └─ ComfyZoomSequence/
      ├─ comfy_zoom_sequence.py
      ├─ __init__.py
      └─ requirements.txt

---

🧰 Requirements

torch>=1.10

ComfyUI already includes PyTorch; this line simply ensures version compatibility for external installs.

---

🪄 Usage Example

Typical workflow:

[Load Video / Image Sequence]
          │
          ▼
 [Zoom Sequence (Zoom In/Out, Easing)]
          │
          ├─▶ (images) → [Save Video] or [Preview Image]
          └─▶ (info)   → [Print Text] or [Console Log]

Example parameters:

| Parameter | Example | Description | |------------|----------|-------------| | mode | Zoom In | Zooms progressively inward | | pixels_per_frame | 1.5 | Crops 1.5px per side per frame (on the smaller image dimension) | | ease | Ease_In_Out | Starts and ends smoothly |

---

🧪 Example Output

When you run the node, the info output will look something like:

Frames: 120, Canvas: 1920x1080, Mode: Zoom In, Ease: Ease_In_Out
Requested small-dim max margin: 178.50 px
Applied small-dim max margin:   178.50 px (safe limit: 539 px)
Note: margins are proportional per axis to preserve aspect; integer crop indices are used.

---

💡 Notes

• Keeps original aspect ratio and canvas size exactly.
• Works seamlessly with Load Video, Batch Images, and Save Video nodes.
• pixels_per_frame applies to the smaller image dimension for proportional scaling.
• Fractional pixel speeds (e.g. 1.25) are supported and automatically rounded during cropping.
• Margins are clamped so crops never exceed safe limits.

---

🧑‍💻 Credits

Created by [Your Name or Handle]
Tested with ComfyUI portable and PyTorch 2.7.1+cu128.

---

📜 License

MIT License