Hey everyone! I’ve been building some n8n workflows to make it easier to integrate video background removal into AI generation pipelines. Thought I’d share what I’ve built so far in case it helps anyone working with Sora 2, Veo 3 or other AI video models.
The Problem
AI tools like Sora, Veo, or HeyGen often generate videos with backgrounds, so you need a way to automatically remove them, add custom scenes, and scale the process without manual work.
Here are the first two workflows I’ve built:
Workflow 1: Basic Video Composition
A simple background removal and composition workflow.
Use cases:
- AI-generated actors over custom scenes
- Product demos with removed backgrounds
- Any general video compositing need
Flow:
Upload foreground video → Remove background → Composite over background video → Export to Google Drive
Workflow 2: Screen Record of App UGC Ad Pipeline
Automates the entire UGC ad creation process for apps.
Use cases:
- App demo videos with AI presenters
- SaaS feature announcements
- Mobile app marketing videos
Flow:
Screen recording → Gemini analyzes it → Sora 2 generates AI actor → Remove actor background → Composite over screen recording → Export to Google Drive
The cool part here is that Gemini watches your screen recording and writes a contextual script, then Sora 2 generates an actor delivering that script. The actor is then composited on the bottom right over your app UI.
You can swap Sora 2 for Veo 3, Kling AI, or any other model just by updating one section. The background removal API supports composition, so you can easily adjust positioning and sizing.
The workflows are on GitHub with full setup instructions:
Each template includes:
- Complete n8n workflow JSON
- Detailed sticky notes
I’m planning to add more workflows for:
- Batch processing multiple videos
- Integrations with different AI video models
- …
I’d love to hear your thoughts are there any specific use cases you’d find helpful or questions about how these workflows work?