Traceless AI

What is traceless AI?

Traceless AI is an online intelligent removal tool designed to eliminate watermarks, subtitles, and unwanted elements from videos.

The platform identifies each frame and redraws the obscured areas by completing the image, while also offering functions for removing unwanted elements from images and improving their clarity.

Core functions

Remove watermarks from videos

Video watermark removal can handle fixed or moving text, graphics, watermarks, and LOGOs, among other visual elements.

• The selection mode is suitable for watermarks that are always located in a fixed area.
• Full-screen mode is suitable for dynamic watermarks whose position keeps changing.
• AI will attempt to redraw the background covered by the watermark, rather than simply blurring it.
• Complex textures, rapid movement, and large areas of occlusion can leave traces of restoration.
• Only proprietary or authorized materials and logos can be processed.

Remove subtitles from video

The subtitle removal feature can target a specific area containing subtitles, or it can scan the entire screen to remove text there as well.

• For fixed subtitles, the selection mode is preferred, as it helps to minimize changes in unrelated areas.
• When the subtitle position changes or the coverage area is large, you can choose full-screen mode.
• After generation, the edges of the characters, their mouth shapes, and complex backgrounds need to be checked section by section.
• If translation is still needed, it is recommended to keep the original subtitle file and timeline first.
• It is not allowed to remove the copyright notice and republish someone else’s work without authorization.

Universal video removal

The video eraser is used to remove unwanted people, objects, obstructions, or specific elements in the image.

• The fixed position mode is suitable for static markers and objects with minimal changes.
• Smart tracking is suitable for tracking people or objects that are in constant motion within the frame.
• The larger the object and the longer it remains in cover, the more background information the AI needs to guess.
• After switching the camera view, it is necessary to reconfirm the selection area and tracking range.
• Mosaic removal cannot guarantee the restoration of the true original information.

Convert video to high definition

Video upscaling to high definition is used to improve the viewing quality of low-resolution content, offering high-definition or ultra-high-definition output options.

• It is suitable for videos that are heavy, have low resolution, or lack detail.
• Enhanced models may generate fake details that are not present in the original video.
• Check the face, text, product texture, and edges before outputting.
• High resolution does not necessarily mean an simultaneous increase in true clarity and amount of information.
• For formal project proposals, it is recommended to retain the original documents and compare the different versions.

Image removal

The universal image removal tool can delete watermarks, annotations, handwritten notes, or unwanted people from images.

• Try to select only the areas that need to be removed, in order to minimize the impact on the main part.
• Solid colors and regular textures are usually easier to repair than complex backgrounds.
• Text that is dense, overlapping edges, and transparent objects require close inspection.
• When handling documents, receipts, and certificates, they must not be used to forge or alter facts.
• The background change and enhancement of images are based on the real-time status of the workstation.

Guide to Using Traceless AI

1. Log in to Traceless AI, and select video to remove watermarks, subtitles, or pen marks.
2. Confirm that the material belongs to you or that you have obtained legitimate permission to use it.
3. Upload MP4 or MOV videos and check the duration and resolution limits.
4. For fixed elements, the selection mode is box selection; for moving elements, full-screen or smart tracking is used.
5. Preview the selection area to ensure that no important figures or main products are covered.
6. Check the estimated points consumption and current queue status before submitting the task.
7. Wait for the cloud processing to be completed; do not submit the same task again.
8. Check section by section the traces of repairs, image quality, audio-video synchronization, and the encoding results.
9. Download the final product within the material’s validity period and keep a local backup.
10. Confirm the copyright, platform rules, and authenticity of the content once more before publishing.

How to choose between fixed and dynamic watermarks

Price of the subscription package

As of August 30, 2026, the official website lists four different packages of points, with the price per point decreasing as the quantity purchased increases.

Prices, concurrency levels, and duration may vary; prior to purchase, the information on the top-up page, the order page, and the current agreement should be referred to.

Rules for consuming points

Video tasks are billed in 30-second intervals, with different tools and processing methods resulting in varying costs.

For durations of less than thirty seconds, the rules regarding rounding and handling failed tasks should be checked through the real-time prompts before submission.

Example of cost estimation

• Five minutes of fixed subtitles are estimated at ten billing units, which requires 50 points.
• Five minutes of full-screen subtitles are estimated at ten billing units, which requires 100 points.
• A five-minute selection of a watermark is estimated based on ten billing segments, and 80 points are required.
• A five-minute full-screen watermark is estimated across ten billing periods, requiring 150 points.
• Five minutes of intelligent tracking is estimated based on ten billing segments, and it requires 200 points.
• The actual fee charged will be based on the system’s rounding of the video length and the submission instructions.

Format, duration, and storage

• The official website states that it primarily supports MP4 and MOV formats.
• The official documentation states that the maximum resolution is 1080P.
• The homepage indicates that a single video can be up to fifteen minutes long; for some level categories, the maximum length is thirty minutes.
• The actual maximum duration depends on the account privileges and the task page.
• The user agreement states that the materials and finished products are stored in temporary cloud storage.
• The lower-tier subscription plans display the data for seven days, while other public tiers show it for thirty days.
• Important materials must be downloaded to your local device before the deadline.

Mobile devices and supported platforms

• Web version: It is possible to upload and process materials using Chrome or Edge.
• WeChat Mini Programs: These can be used on mobile devices by searching for the names of the programs that have been officially published.
• Points synchronization: The web version and the mini-program can share points when using the same phone number.
• Native app: The agreement states that if it is launched, it will be announced on the official website.
• Local offline processing: Currently, core AI tasks need to be uploaded to cloud computing.

Video watermark removal API

The authorities provide a watermark removal API for short videos used in bulk operations, which can be integrated into private workflow systems.

• The official website features a simple promotional interface and offers one-on-one technical support.
• Charging is based on actual usage, with no minimum consumption requirement.
• When connected to the service, businesses can access dedicated processing channels, which reduces the impact of regular queues.
• No public documentation for self-service interfaces or a unified API pricing list was found.
• Before connecting, it is necessary to confirm the authentication, callback, concurrency, storage, and retry rules in case of failures.

API, GitHub, and open-source status

• API: An interface for removing watermarks from videos is available; you need to contact the platform to request access to it.
• SDK: No official public SDK download page could be found.
• GitHub: No official code repository that corresponds explicitly to the current product was found.
• Open source: The web platform, the processing models, and the server-side code are not made available as open source.
• Private deployment: The official website does not currently display standard private deployment solutions.

Material privacy and temporary storage

The user agreement states that the materials uploaded and the final products generated constitute the user’s business data, with the platform providing temporary cloud storage for them.

• The platform states that it will not use or disclose users’ business data without authorization.
• The privacy policy briefing collects information necessary for the operation of the service, as well as details related to accounts and payments.
• The paid and bulk publishing features may require providing necessary information to the relevant service providers.
• Sensitive videos, commercial samples, and personal footage should have their upload risks assessed first.
• Temporary storage is not the same as long-term backup; expired files may be automatically deleted.

Copyright and compliance boundaries

• Only process videos and images that you have created yourself or for which you have obtained authorization.
• It is not allowed to infringe upon the rights of authors, platforms, or brands by removing rights indicators.
• Elimination tools must not be used to forge documents, receipts, contracts, or facts.
• Retain material licenses, project assignments, and source records for subsequent verification.
• Check the licensing for music, portraits, trademarks, and visual content before republishing.
• The handling of AI-generated identifiers and platform watermarks must comply with the rules of the publishing platform.

Which users are it suitable for

• Video creators: Remove subtitles, markers, and other elements from their own footage.
• E-commerce team: Handles video content for products for which authorization has been obtained from suppliers.
• Short Drama Team: Handles the subtitle sections for its own dramas in bulk.
• Advertising producers: Remove temporary markers and errors from the footage.
• Enterprise developers: Integrate processing capabilities into internal processes via APIs.
• Regular users: Fix distracting elements in personal videos and photos.

Usage restrictions and precautions

• AI reconstruction is based on predictions, and it cannot guarantee the restoration of the actual content that was obscured.
• Large areas of obstruction, complex movements, and rapid camera cuts can easily lead to repair flaws.
• Reencoding may affect the file size, compatibility, or detail quality.
• Points, duration, and storage benefits shall be based on the real-time account page.
• The rules regarding refunds in cases of processing failure or unsatisfactory subjective results should be reviewed in advance.
• Cloud processing is not suitable for highly sensitive materials that have not been evaluated.
• Watermark removal technology cannot replace legitimate authorization and copyright verification.

Frequently Asked Questions

What can traceless AI mainly handle?

It can handle video watermarks, subtitles, as well as characters, objects, and occlusions; it also provides tools for video enhancement and image removal.

How to choose between fixed watermarks and moving watermarks?

Fixed watermarks are selected using a prioritized approach; for movable or full-screen watermarks, the full-screen mode is employed, and intelligent tracking can be used for moving objects.

Is Traceless AI free?

New users can receive trial credits; to continue using the service, they need to purchase credits, with the specific amount offered indicated on the registration page.

Which video formats is Traceless AI compatible with?

The official website states that MP4 and MOV formats are supported; the limitations regarding resolution, duration, and file size are specified on the task page.

How long are the uploaded materials stored?

The accrued benefits are stored temporarily for either seven days or 30 days; the files may be deleted once the deadline expires, so it is necessary to download a local backup in a timely manner.

Does Traceless AI provide APIs?

An API for removing watermarks from videos is available; pricing is based on actual usage with no minimum charge. For details regarding the interface and costs, please contact the platform.

Can Traceless AI process any other person’s videos?

No, it is only possible to work with material that is owned or for which authorization has been obtained; distributing such material without removing the copyright markings may result in legal consequences.