UnmarkAI

Last updated: 2026-08-10

UnmarkAI is a browser-based AI video cleanup and localization platform that uses AI inpainting to erase hardcoded captions, watermarks, and scene objects from video while rebuilding the background. It then translates and dubs the cleaned footage into more than 20 languages through a self-serve REST API with async jobs and webhooks.

About UnmarkAI

UnmarkAI is a browser-based video cleanup and localization platform run by a small independent studio based in Hangzhou, China. It targets one recurring problem for creators and localization teams: source footage carrying hardcoded captions, platform watermarks, timestamps, or on-screen objects that need to come out cleanly before the clip can be re-subtitled, dubbed, or republished. The core mechanism is AI inpainting: the video text remover and object remover analyze each frame, erase the flagged region, a caption, a lower third, a passerby, a prop, and rebuild the background behind it instead of blurring or cropping over it, so the source resolution and frame rate stay intact. That cleaned frame then feeds a localization pipeline (clean, transcreate, voice, QA) that produces subtitle translations and AI dubbing in multiple languages, including Spanish, Portuguese, Arabic, Japanese, Korean, and Hindi. A separate AI upscaler and an image-focused magic eraser round out the toolset. Its clearest use cases are short-form video repurposing for creators clearing burned-in captions before a re-edit, overseas short-drama localization for studios releasing vertical dramas in new markets, and e-commerce teams turning one product demo into several dubbed market versions. A self-serve REST API, bearer-token authentication, async jobs, webhooks, and idempotency keys, lets teams fold cleanup and dubbing into an existing content pipeline instead of uploading through the browser one file at a time. UnmarkAI is web-only, with no desktop or mobile app. Access is priced on a points system with a free starter plan, a monthly subscription, one-time point packs, and custom Enterprise terms; points are charged by processing time and cleanup type rather than a flat per-video fee, and the exact breakdown is on the pricing page. Its terms of service and acceptable use policy require that uploaded content be owned, licensed, or explicitly authorized for editing, and separately prohibit removing watermarks or attribution in a way that breaches copyright, trademark, or platform rules, with a zero-tolerance, mandatory-reporting policy for child-exploitation material. The vendor has not published an independent third-party security certification, worth flagging for teams with a formal vendor-risk review process.

Pricing

Free: $0, 10 points on signup (roughly one to two short clips), output capped to the first 30 seconds and watermarked. Subscribe: $17.92/mo for 1,200 monthly points, files up to 2GB, videos up to 60 minutes. One-time Buy Points pack: $49 for 1,500 points, files up to 500MB, videos up to 5 minutes. Enterprise: custom monthly points, team workflow support, higher upload and export limits, sales contact required. Video translation and dubbing use 10 points per 30 seconds of output; API usage bills a separate USD wallet by billable video seconds, rate varies by cleanup mode and account tier.

Key Features

• Video Text & Watermark Remover: AI inpainting erases hardcoded captions, timestamps, lower thirds, and platform watermarks frame by frame and rebuilds the background instead of blurring or cropping over it, keeping the source resolution and frame rate.
• Video Object Remover: Clears scene distractions, a passerby, a prop, a cable, a review overlay, from footage without cropping the frame.
• AI Video Upscaler: Increases resolution on cleaned or source footage as a separate step in the same dashboard.
• Video Translation Studio: Runs subtitle translation and AI dubbing across dozens of languages, including Thai, Indonesian, French, and Vietnamese, through a four-stage clean, transcreate, voice, and QA pipeline built for short-drama and product-video localization.
• Points-Based Batch Dashboard: Processes videos in bulk with side-by-side before and after task previews before download, so teams can check a result before spending on a full export.
• Self-Serve REST API: Offers bearer-token authentication (uma_live_* keys), asynchronous jobs, webhooks, and idempotency keys for safe retries, aimed at teams automating cleanup or dubbing inside an existing content pipeline.

Pros

• Ships a working self-serve API, bearer-token auth, async jobs, webhooks, idempotency keys, with a default concurrency of 4 parallel jobs, so the automate-via-API claim is checkable against real endpoint documentation rather than being sales-only.
• Runs a named four-stage clean-transcreate-voice-QA localization workflow aimed specifically at short-drama and e-commerce dubbing, not just generic auto-subtitling bolted onto a text tool.
• States a specific, named zero-tolerance policy with mandatory NCMEC reporting for child-exploitation material in its Acceptable Use Policy, a more concrete governance disclosure than most small watermark-removal competitors publish.

Cons

• Free plan caps output to a short preview window and stamps a watermark on the export, so judging full-length results requires a paid plan.
• No published SOC 2, ISO 27001, GDPR, or HIPAA attestation, and the privacy policy does not state whether uploaded video is used to train models, worth checking before enterprise use.
• Web-only with no desktop or mobile app, and the vendor's own limitations notice says results vary with heavy motion, low source resolution, or complex backgrounds.
• Operated as a single-person Individual Business registered in Hangzhou, China, per its own Terms of Service, rather than an incorporated, independently audited company, relevant for procurement teams doing vendor-risk review.

Visit UnmarkAI Official Website