Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

StegaNeRF: Embedding Invisible Information within Neural Radiance Fields

About

Recent advances in neural rendering imply a future of widespread visual data distributions through sharing NeRF model weights. However, while common visual data (images and videos) have standard approaches to embed ownership or copyright information explicitly or subtly, the problem remains unexplored for the emerging NeRF format. We present StegaNeRF, a method for steganographic information embedding in NeRF renderings. We design an optimization framework allowing accurate hidden information extractions from images rendered by NeRF, while preserving its original visual quality. We perform experimental evaluations of our method under several potential deployment scenarios, and we further discuss the insights discovered through our analysis. StegaNeRF signifies an initial exploration into the novel problem of instilling customizable, imperceptible, and recoverable information to NeRF renderings, with minimal impact to rendered images. Project page: https://xggnet.github.io/StegaNeRF/.

Chenxin Li, Brandon Y. Feng, Zhiwen Fan, Panwang Pan, Zhangyang Wang• 2022

Related benchmarks

TaskDatasetResultRank
Digital WatermarkingBlender and LLFF (test)
Bit Accuracy (No Attack)93.15
39
3D Gaussian Splatting WatermarkingBlender and LLFF (test)
Bit Accuracy93.15
24
SteganographyDTU Bricks scene
PSNR21.67
8
SteganographyDTU Tools scene
PSNR22.38
8
SteganographyDTU Birds scene
PSNR20.2
8
SteganographyDTU Snowman scene
PSNR20.63
8
3D SteganographyTanks & Temples Message
PSNR28.114
4
3D Scene SteganographyTHuman_MV 2 scenes
Cover PSNR28.72
4
3D Scene SteganographyTHuman_MV 3 scenes
Cover PSNR28.01
4
3D SteganographyTanks & Temples Scene
PSNR27.251
4
Showing 10 of 20 rows

Other info

Follow for update