Original Summary

Hello HN!<p>The inspiration for Etchv came directly from friends who raised concerns about the growing ease of unauthorized media copying without attribution. To address this, Etchv embeds an invisible watermark inside images, PDFs, or short videos that survives modifications like screenshots, format changes, and file compression.<p>You can try it on our homepage without signing up. There&#x27;s also a live demo where a watermarked PDF page gets screenshotted and the watermark ID is still successfully recovered: <a href="https:&#x2F;&#x2F;etchv.com&#x2F;demo&#x2F;trace" rel="nofollow">https:&#x2F;&#x2F;etchv.com&#x2F;demo&#x2F;trace</a><p>Instead of training a model from scratch, our model is custom-trained on a blend of the COCO Dataset and a collection of proprietary datasets optimized to mimic real-world media variations.<p>On 208 test photos, the watermark survived the shrinking and compression that a social feed or chat app applies, every time. Tiny thumbnails mostly killed it—only 16 of 208 survived. Full benchmark results can be found here: <a href="https:&#x2F;&#x2F;etchv.com&#x2F;solutions&#x2F;creative-delivery#survival" rel="nofollow">https:&#x2F;&#x2F;etchv.com&#x2F;solutions&#x2F;creative-delivery#survival</a><p>Limits: video is restricted to short clips for now (a few seconds at 360p), and PDFs max out at 8 pages. Additionally, someone who knows what they&#x27;re doing can always remove the mark.<p>We offer an API with official SDKs in six languages, as well as an MCP server you can integrate directly into your workflow.

  • 情报分类：技术学习与提效
  • 分类依据：内容涉及技术、AI、软件工具或工程实践
  • 信息来源：Hacker News 新项目
  • 发布时间：2026/10/8 20:28:57