Research map / 3D vision / Gaussian splatting
WhisperSplat: Lossless Steganography in 3D Gaussian Splatting
Nicole Meng, Ronak Sahu, Miao Yin, Faysal Hossain Shezan, Yingjie Lao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use noise key on SH features instead of retraining 3DGS with GS-Hider | WhisperSplat · 9 scenes (avg) | secret recovery PSNR (dB): 25.178 -> 26.563 | Improved |
| use noise key instead of KeySS | WhisperSplat · 9 scenes (avg) | secret recovery PSNR (dB): 26.427 -> 26.563 | Improved |
| use noise key instead of KeySS | WhisperSplat · 9 scenes (avg) | clean-view PSNR (dB): 25.980 -> 26.518 | Improved |
| use noise key instead of GS-Hider | WhisperSplat · 9 scenes (avg) | clean-view PSNR (dB): 25.817 -> 26.518 | Improved |
| use noise key instead of 3DGS+Decoder | WhisperSplat · 9 scenes (avg) | secret recovery PSNR (dB): 21.249 -> 26.563 | Improved |
| use noise key instead of KeySS | WhisperSplat · Treehill | secret recovery PSNR (dB): 22.121 -> 26.074 | Improved |
| use noise key instead of KeySS | WhisperSplat · Bicycle and Stump | secret recovery PSNR (dB): 29.533 -> 29.394 | Hurt |
| learning rate 0.01 versus 0.1 (ablation) | WhisperSplat · 9 scenes (avg) | secret recovery PSNR (dB): 26.321 -> 26.748 | Improved |
| learning rate 0.01 versus 0.001 (ablation) | WhisperSplat · 9 scenes (avg) | secret recovery PSNR (dB): 26.648 -> 26.748 | No clear effect |
| increase random Gaussian pruning from 5% to 25% (ablation) | WhisperSplat · 13 scenes (mean, pre-refiner) | hidden-view PSNR change (dB): -0.23 -> -1.31 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GS-Hider — GS-Hider: Hiding Messages into 3D Gaussian Splatting (NeurIPS 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Splats in Splats: Robust and Effective 3D Steganography Towards Gaussian Splatting (AAAI 2026 · similarity 0.93)
- SecureGS: Boosting the Security and Fidelity of 3D Gaussian Splatting Steganography (ICLR 2025 · similarity 0.91)
- GS-Hider: Hiding Messages into 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.91)
- SplatFormer: Point Transformer for Robust 3D Gaussian Splatting (ICLR 2025 · similarity 0.88)
- MaskGaussian: Adaptive 3D Gaussian Representation from Probabilistic Masks (CVPR 2025 · similarity 0.87)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
