muapi Logomuapi Logo
muapi Logomuapi Logo

Seedance 2 Watermark Remover: AI Video Editor

🎉 FREE for a limited time — Remove SD 2.0 watermarks from videos using LaMa AI inpainting. Automatically detects the watermark region, builds a precise mask via Canny edge detection, and inpaints each frame for artifact-free results. No credits deducted — requires a positive balance to access.

Interactive model controls

🎉 FREE for a limited time — Remove SD 2.0 watermarks from videos using LaMa AI inpainting. Automatically detects the watermark region, builds a precise mask via Canny edge detection, and inpaints each frame for artifact-free results. No credits deducted — requires a positive balance to access.

📝

Overview

About this model

🎉 Free for a limited time. SD 2.0 Watermark Remover is a free AI tool purpose-built to remove the SD 2.0 watermark from generated videos. It uses LaMa (Large Mask inpainting) to detect the SD watermark region automatically using Canny edge detection on a mean-frame average, then inpaints each frame with zero visible artifacts. Unlike generic delogo tools, the patch-only inpainting strategy processes only the watermark area — making it 50–100× faster than full-frame approaches while preserving the surrounding video perfectly. Available at $0.00 per use for any account with a positive credit balance, for a limited time.

1Cleanup: Remove the SD 2.0 watermark from AI-generated videos before publishing or sharing.
2Post-production: Clean SD-generated footage for use in branded content or commercial projects.
3Archival: Store watermark-free copies of SD 2.0 video generations for personal or professional archives.
4Prototyping: Quickly produce clean preview videos from SD 2.0 outputs during creative development.
5Social media: Prepare watermark-free SD 2.0 clips ready for upload to YouTube, Instagram, or TikTok.
💰

Pricing & Value

Cost analysis

muapiapp🎉 Free for a limited time ($0.00)

Currently $0.00 per use for all muapiapp users with a positive account balance. This is a limited-time promotional offer — pricing may be introduced in the future. No credits are deducted per use during this period.

Fal.aiNot available

Fal.ai does not offer a dedicated SD 2.0 watermark removal tool.

ReplicateNot available

Replicate does not offer a dedicated SD 2.0 watermark removal tool.

** Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Video URLstring

URL of the Seedance v2.0 video to remove the watermark from.

Default Valuehttps://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-i2v.mp4
📖

Implementation Guide

Developer documentation

How to Use the SD 2.0 Watermark Remover

  1. Generate your video with SD 2.0

    • Use the seedance-pro or seedance-lite endpoints to generate a video.
    • Copy the output video URL.

  2. Submit the watermark removal request

    • Send a POST request to /api/v1/sd-2.0-watermark-remover with your video URL: 
      {
  "video_url": "https://your-seedance-output.mp4"
}
    • You will receive a request_id immediately.

  3. Poll for the result

    • Call GET /api/v1/predictions/{request_id}/result periodically.
    • Status will move from processing → completed.
    • The response will contain the cleaned video URL in output.video.

  4. Download your clean video

    • Fetch the URL from output.video — the SD 2.0 watermark has been removed by LaMa inpainting.

Note: Processing time scales with video length. The LaMa inpainting model processes the watermark patch only, keeping times manageable even for longer clips.

❓

Common Questions

Frequently asked

What watermark does this remove?

It is specifically designed to remove the **SD 2.0** watermark placed by the SD video generation model. The auto-detection algorithm scans video corners for the characteristic SD text overlay and builds a precise inpainting mask.

How does the detection work?

The tool samples up to 60 frames and computes a mean frame — the static SD 2.0 watermark becomes clearly visible while moving content averages out. Canny edge detection then traces the letter strokes of the watermark text, creating a precise mask that avoids over-masking uniform backgrounds.

What inpainting model is used?

LaMa (Large Mask inpainting), a state-of-the-art model that fills masked regions using fast Fourier convolutions. It is significantly better than ffmpeg's delogo/removelogo filter, producing no blocky artifacts.

How much does it cost?

🎉 The SD 2.0 Watermark Remover is **free for a limited time** — $0.00 per use. This is a promotional offer; pricing may be introduced in the future. You only need a positive credit balance (> $0) on your account to access it while it's free.

Why do I need credits if it's free?

A positive credit balance is required to prevent abuse and ensure fair use during the free period. The processing is computationally intensive (runs LaMa inpainting on every frame), so we reserve it for active paying users. Once the free period ends, a per-use cost will apply.

How long is it free for?

The free period is for a limited time only. We will announce any pricing changes in advance. Use it now to take advantage of $0.00 per generation while the offer lasts.

Will it work on non-SD videos?

The auto-detection is tuned for the SD 2.0 watermark position and style. It may work on other corner-placed text watermarks, but results are not guaranteed for other video generators.