Visible Watermarking with Gradio

Summary

Generative artificial intelligence tools produce images, video, audio, and text that have become increasingly difficult to distinguish from real-world media captures. To improve synthetic media transparency, Hugging Face introduced visible watermarking capabilities directly into the Gradio web application framework. Developers can now overlay visual watermarks on image and video outputs by specifying a single watermark parameter using file paths, open image objects, or NumPy arrays. The implementation also supports QR code watermarks, which can match the visual style of generated media while conveying detailed background information. Furthermore, the Gradio Chatbot component accepts a text watermark parameter that automatically attaches attribution metadata whenever users copy generated responses to their clipboards.

Context

AI-generated images and videos have become realistic enough to be nearly indistinguishable from real camera footage, creating an urgent need for mechanisms that allow users to identify synthetic content.

Approach / What changed

Hugging Face added a watermark parameter to Gradio components, enabling developers to overlay watermarks onto gr.Image and gr.Video outputs using filenames, open images, or NumPy arrays, as well as attach text attribution to gr.Chatbot copy actions.

Takeaways

• Gradio components gr.Image and gr.Video support visible watermarking by accepting filenames, open image objects, or NumPy arrays through the watermark parameter.
• QR code watermarks can be embedded into visual outputs to provide detailed metadata while matching the visual style of the generated image or video.
• The gr.Chatbot component uses a text watermark parameter to automatically append attribution whenever users copy text from generated chatbot responses.

Related reading

• huggingface.co · Dec 8, 2021

  Training CodeParrot 🦜 from Scratch

  Training code generation models from scratch requires careful data filtering, tokenization, and compute management. To build CodeParrot, a 1.5-billion-parameter Python generation model, engineers extracted 20 million Python files from BigQuery and cleaned the resulting 180 GB corpus down to 50 GB after discovering extreme duplication. A custom GPT-2 tokenizer was trained over streamed samples, and the GPT-2 large architecture was initialized with layer-scaled, full-precision attention mechanisms. The training pipeline leveraged Hugging Face Accelerate alongside gradient checkpointing and a streaming iterable dataset designed to yield fixed-length concatenated token sequences. CodeParrot was trained on roughly 25 to 30 billion tokens and evaluated on coding tasks using the pass@k metric.

  Leandro von Werra

• huggingface.co · Dec 21, 2021

  Gradio is joining Hugging Face!

  Hugging Face has acquired Gradio, the open-source machine learning library designed for building and sharing interactive model interfaces. Gradio originated in 2019 when its founder struggled to share a medical computer vision model with a physician collaborator who did not write Python. Co-founded alongside Ali Abdalla, Ali Abid, and Dawood Khan, the project expanded from computer vision into text, speech, and video modalities. Over 300,000 demos have been built using Gradio, allowing interdisciplinary industry teams and researchers to debug models internally and showcase them externally to non-technical users. The acquisition unites Gradio with Hugging Face to broaden browser-based machine learning accessibility and expand hiring efforts across the joint team.