Remove Background Video API
AI-powered video background removal for developers. Remove backgrounds, create transparent video, and automate production workflows.
curl -X POST https://api.videobgremover.com/v1/jobs \
-H "X-API-Key: $VIDEO_BG_API_KEY" \
-H "Content-Type: application/json" \
-d '{"video_url":"https://example.com/video.mp4"}'
# Then start background removal
curl -X POST https://api.videobgremover.com/v1/jobs/JOB_ID/start \
-H "X-API-Key: $VIDEO_BG_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"videobgremover-light",
"background":{"type":"transparent"}}'REST API, SDKs, webhooks, and delivery history included.
Already have a green screen?
Try Online Chroma Key & Export in WebMDeveloper-First Video Processing
Official SDKs
TypeScript-first Node.js SDK and Pythonic SDK with type hints. Handle authentication, job management, and video composition automatically.
Complete Workflow
Remove backgrounds with AI, then compose with custom backgrounds locally using FFmpeg. One SDK handles both API calls and video processing.
Professional Output
Export in H.264, VP9 WebM, ProRes MOV, or PNG sequences. From web apps to professional video editing workflows.
How to Build with Our SDKs
1. Install SDK
- npm install @videobgremover/sdk
- pip install videobgremover
- Get your API key from dashboard
2. Remove Background
- Load video with Video.open()
- Call video.removeBackground({ client })
- Get transparent video automatically
3. Compose & Export
- Create custom backgrounds locally
- Position and layer videos with precision
- Export in professional formats
Choose the Right AI Model
4 specialized models for general, speed-focused, prompted, and human-focused workflows
Original
videobgremover-original
Highest quality segmentation with detailed edge detection. Best for complex scenes.
Best for:
Complex scenes, detailed segmentation
Light
videobgremover-light
Fast and efficient processing optimized for speed. Great for simple backgrounds.
Best for:
Fast processing, simple backgrounds
Pro
videobgremover-pro
Advanced background removal for complex objects. Enables text prompting for precise segmentation.
Best for:
Complex objects, text-based segmentation ("red car", "golden retriever")
Human
videobgremover-human
Fast matting algorithm optimized for portraits and facing people.
Best for:
Human subjects, portraits, face cam videos
| Model | Speed | Quality | Text Prompts | Pricing |
|---|---|---|---|---|
Original videobgremover-original | Standard | Highest | — | 60 credits/min |
Light videobgremover-light | Speed-optimized | High | — | 45 credits/min (25% cheaper) |
Pro videobgremover-pro | Slowest | Highest | Yes | 180 credits/min |
Human videobgremover-human | Human-optimized | High | — | 45 credits/min (25% cheaper) |
Video Processing API Code Examples
Remove Background Video API Pricing
Usage-based credits with the exact charge returned when each processing job starts
API Features
- RESTful API with JSON responses
- API-key authentication
- MP4, MOV, and WebM input support
- Asynchronous processing for long videos
Processing Features
- AI-powered background removal
- Custom backgrounds & transparent videos
- Processed video + mask output
- No green screen required
Pay As You Go API Processing
No subscriptions. Pay only to process your videos via API.
Remove Background Video API FAQ
What is the Remove Background Video API?
Our Remove Background Video API allows developers to integrate AI-powered video background removal directly into their applications. The API automatically detects subjects in videos and removes backgrounds without requiring green screens, making it perfect for apps, websites, and automated workflows.
How does the Video Background Removal API work?
The API uses advanced AI models to process videos in four simple steps: 1) Create a job and get an upload URL, 2) Upload your video file, 3) Start processing with optional background customization (solid colors, transparent formats), and 4) Retrieve the processed video. You can specify color backgrounds (#FF0000), transparent outputs (WebM, ProRes MOV, PNG sequence), or use the default green screen. The entire process is automated and requires no manual intervention.
What video formats does the Remove Background API support?
The API accepts MP4, MOV, and WebM video files. Jobs are processed asynchronously, so your application does not need to keep an HTTP request open while processing runs. Check the current API reference for upload and account limits.
How much does the Video Background Removal API cost?
The API uses model-based credits. Original costs 60 credits per video minute, Light and Human cost 45 credits per minute, and Pro costs 180 credits per minute. High-frame-rate processing can use an additional FPS multiplier. The start-job response returns the exact credits charged.
How long does video processing take?
Processing time varies with video length, frame rate, resolution, selected model, output format, and current capacity. Jobs run asynchronously; use the status endpoint or webhook notifications to know when processing has completed. Customer webhook deliveries are not currently HMAC-signed, so verify the job through the authenticated API before taking sensitive actions.
What are the API rate limits and quotas?
Limits can vary by endpoint and account. Refer to the current API reference and API responses for the limits that apply to your integration, or contact us before planning a high-volume workload.
Do you provide SDKs for the Remove Background Video API?
Yes! We provide official SDKs for Python and Node.js that make integration even easier. The SDKs handle authentication, job management, and video composition automatically. You can also use the REST API directly with any programming language using our comprehensive cURL examples and documentation.
Can the API remove backgrounds without green screens?
Yes. The models are designed to remove backgrounds without a green screen. Results vary with the subject, motion, lighting, resolution, and selected model, so test representative footage before production use.
What output formats does the Remove Background Video API provide?
The API offers flexible output options: 1) Default green screen MP4 videos, 2) Custom color backgrounds in MP4 format, 3) Transparent videos in WebM VP9, ProRes MOV, or PNG sequence formats, and 4) High-quality mask videos. You can specify your preferred output format when starting processing, making it perfect for any workflow from web applications to professional video editing.
Can I get the video mask to use in my own compositing workflow?
Yes! Along with the processed video, you receive a high-quality mask video that shows exactly which pixels contain the subject (white) versus background (black). This mask is perfect for advanced compositing workflows, allowing you to apply custom backgrounds, effects, or integrate with video editing software like After Effects, Premiere Pro, or custom compositing pipelines.
Is there webhook support for processing completion?
Yes. You can provide a webhook URL and receive lifecycle notifications such as job.started, job.completed, and job.failed. Delivery attempts are recorded so you can inspect failed deliveries; polling remains available as a fallback. Customer webhooks are not currently HMAC-signed, so use an unpredictable HTTPS URL and validate job IDs through the API.
How do I get started with the Video Background Removal API?
To get started: 1) Sign up, 2) Create an API key in the management dashboard, 3) Purchase processing credits, 4) Make your first API call to create a job, and 5) Upload and process a test video. Creating an API key does not add free credits.
Which AI model should I use?
We offer 4 AI models to fit different needs: videobgremover-original for detailed segmentation and general use, videobgremover-light for speed-focused processing at 45 credits per minute, videobgremover-pro for prompt-guided segmentation at 180 credits per minute, and videobgremover-human for human subjects and portraits at 45 credits per minute. Choose based on your footage, credit rate, and whether you need prompt support.
What are text prompts and how do they work?
Text prompts are supported by videobgremover-original and videobgremover-pro. Instead of relying only on automatic subject detection, you can describe what to keep, such as 'person wearing red jacket' or 'golden retriever dog'. Set prompt.mode to 'text' and provide the description when starting a job.
Can I switch between models for different videos?
Yes! You can specify which model to use for each video processing job by setting the 'model' parameter when calling startJob(). This allows you to use videobgremover-original for complex videos requiring highest quality, videobgremover-light for simple backgrounds where speed matters, videobgremover-pro when you need text prompts, and videobgremover-human for portrait videos. All models work with the same API and your credits can be used across any model.