Research map / Label-efficient and robust learning / Recognition and long-tail
Rotation-Invariant Spherical Watermarking via Third-Order SO(3) Representation Coupling
Pengzhen Chen, Yanwei Liu, Xiaoyan Gu, Antonios Argyriou, Wu Liu, Weiping Wang
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 |
|---|---|---|---|
| replace third-order bispectrum extraction with second-order power spectrum (ablation) | TRIAD · 3D rotation, 32-bit payload | bit accuracy (%): 100.0 -> 61.3 | Hurt |
| replace bispectrum with power spectrum at 16-bit payload (ablation) | TRIAD · 3D rotation, 16-bit payload | bit accuracy (%): 100.0 -> 92.4 | Hurt |
| add rotation-based data augmentation to baseline watermarkers | planar baselines (StegaStamp, TrustMark etc.) · random SO(3) rotations | bit accuracy | Mixed |
| use bispectrum-based invariant extraction instead of augmentation | TRIAD · random SO(3) rotations | bit accuracy | Improved |
| embed only in low degree V4 (ablation) | TRIAD · embedding subspace ablation | robustness vs fidelity: robustness >99%, fidelity compromised | Mixed |
| embed only in high degree V16 (ablation) | TRIAD · embedding subspace ablation | robustness vs fidelity | Mixed |
| use multi-degree Vembed = V6+V8+V14 (ablation) | TRIAD · embedding subspace ablation | PSNR and robustness: PSNR 39.22, near-perfect recovery | Improved |
| increase lmax from 16 to 28 and add higher-degree subspaces (ablation) | TRIAD · 3D rotation | PSNR: 39.22 -> 37.19 | Hurt |
| increase lmax from 16 to 28 (ablation) | TRIAD · 3D rotation | bit accuracy: 1.000 -> 1.000 | No clear effect |
| increase lmax from 4 to 24 (ablation) | TRIAD · inference latency | latency | No clear effect |
| aggressive rotation augmentation for baselines forces stronger embedding | baseline watermarkers · panorama watermarking | perceptual quality | Hurt |
Closest papers by meaning
- SphereDiffusion: Spherical Geometry-Aware Distortion Resilient Diffusion Model (AAAI 2024 · similarity 0.82)
- Selective Disk Bispectrum: A Complete and Rotation Invariant Image Descriptor (NeurIPS 2026 · similarity 0.82)
- WateRF: Robust Watermarks in Radiance Fields for Protection of Copyrights (CVPR 2024 · similarity 0.82)
- 3D-GSW: 3D Gaussian Splatting for Robust Watermarking (CVPR 2025 · similarity 0.82)
- GaussianMarker: Uncertainty-Aware Copyright Protection of 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.82)
- GuardSplat: Efficient and Robust Watermarking for 3D Gaussian Splatting (CVPR 2025 · similarity 0.81)
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.
