DreaMark: Rooting Watermark in Score Distillation Sampling Generated Neural Radiance Fields

TL;DR

DreaMark introduces a method to embed watermarks directly during the generation of neural radiance fields, enhancing copyright protection without degrading quality and resisting various attacks.

Contribution

The paper presents a novel watermarking approach that embeds secret messages during NeRF generation, addressing vulnerabilities of prior post-generation methods.

Findings

Watermarking does not degrade NeRF quality.

Watermarks achieve over 90% accuracy under attacks.

Effective against image- and model-level attacks.

Abstract

Recent advancements in text-to-3D generation can generate neural radiance fields (NeRFs) with score distillation sampling, enabling 3D asset creation without real-world data capture. With the rapid advancement in NeRF generation quality, protecting the copyright of the generated NeRF has become increasingly important. While prior works can watermark NeRFs in a post-generation way, they suffer from two vulnerabilities. First, a delay lies between NeRF generation and watermarking because the secret message is embedded into the NeRF model post-generation through fine-tuning. Second, generating a non-watermarked NeRF as an intermediate creates a potential vulnerability for theft. To address both issues, we propose Dreamark to embed a secret message by backdooring the NeRF during NeRF generation. In detail, we first pre-train a watermark decoder. Then, the Dreamark generates backdoored NeRFs in a way that the target secret message can be verified by the pre-trained watermark decoder on an arbitrary trigger viewport. We evaluate the generation quality and watermark robustness against image- and model-level attacks. Extensive experiments show that the watermarking process will not degrade the generation quality, and the watermark achieves 90+% accuracy among both image-level attacks (e.g., Gaussian noise) and model-level attacks (e.g., pruning attack).