SHield: Watermarking 4D Gaussian Splatting via Spherical Harmonic Coefficient Perturbation
Organizations: IIIT Hyderabad · Miko · Tufts University
Abstract
4D Gaussian Splatting (4DGS) models are computationally expensive to train but distributed as easily copied parameter files. Existing watermarks protect these assets by fine-tuning the scene against a neural decoder to extract bit strings from rendered video, a heavy process vulnerable to the temporal dynamics of dynamic scenes. However, if the threat model targets the redistributed file itself, verification can be vastly simplified. Because the canonical Spherical-Harmonic (SH) coefficients of 4DGS are time- and view-invariant, parameter-space perturbations can be recovered via deterministic arithmetic. We propose SHield, the first non-blind, decoder-free watermarking approach that natively embeds full-resolution RGB images into 4DGS files. SHield tiles a zero-mean payload across the SH coefficients of a key-selected Gaussian subset, applying a brief anchored optimization to preserve visual fidelity. To verify ownership, it geometrically re-associates the key using canonical Gaussian centres to survive file tampering and recovers the payload in closed form. Across 30 scenes (D-NeRF, HyperNeRF, DyNeRF), SHield guarantees near-perfect payload recovery (Pearson ) on all 150 tested models with a mean rendering penalty of ~dB. Held-out evaluation yields perfect detection on 400 true trials and zero false positives across wrong-subset null trials. We also strictly bound the method's robustness: SHield survives INT8 quantization and opacity pruning for the image payload, and on aggressive SH-degree-1 truncation bit payloads are still decodable.
Figures & tables
| Dataset (models) | PSNR | PSNR [worst] | SSIM ( ) | LPIPS ( ) | WM-PSNR | min |
| D-NeRF (35) ( Pumarola et al., 2021 ) | 35.31 | 0.048 [ 0.113] | 0.9842 ( 0.0004) | 0.0164 ( 0.0002) | 44.77 | 0.9990 |
| HyperNeRF (85) ( Park et al., 2021 ) | 26.92 | 0.047 [ 0.150] | 0.7761 ( 0.0012) | 0.2949 ( 0.0009) | 36.27 | 0.9910 |
| DyNeRF (30) ( Li et al., 2022 ) | 31.19 | 0.127 [ 0.314] | 0.9353 ( 0.0011) | 0.1099 ( 0.0012) | 36.59 | 0.9952 |
| Opacity pruning | SH quantization (whole file) | ||||||
| WM-PSNR | WM-PSNR | ||||||
| 0 % | 38.32 | 0.998 | 150 | FP32/16 | 38.32 | 0.998 | 150 |
| 10 % | 23.38 | 0.965 | 150 | INT12 | 38.25 | 0.998 | 150 |
| 30 % | 18.21 | 0.892 | 150 | INT10 | 37.35 | 0.998 | 150 |
| 50 % | 15.38 | 0.790 | 150 | INT8 | 32.08 | 0.994 | 150 |
| 80 % | 12.70 | 0.544 | 52 | ||||
| re-associated | oracle indices | |||||||
| Fine-tuned | steps | PSNR | mean | min | mean | min | matched (%) | passes |
| SH only | 100 | 0.01 | 0.9615 | 0.9282 | 0.9615 | 0.9282 | 100.0 | 5/5 |
| SH only | 500 | 0.04 | 0.8994 | 0.8458 | 0.8994 | 0.8458 | 100.0 | 5/5 |
| SH only | 2,000 | 0.18 | 0.7740 | 0.6634 | 0.7740 | 0.6634 | 100.0 | 5/5 |
| SH only | 5,000 | 0.39 | 0.5257 | 0.4312 | 0.5257 | 0.4312 | 100.0 | 1/2 |
| SH only | 10,000 | 0.63 | 0.3946 | 0.3109 | 0.3946 | 0.3109 | 100.0 | 0/2 |
| 32 bits | 48 bits | 64 bits | ||||||||||
| Method | Acc. | PSNR | SSIM | LPIPS | Acc. | PSNR | SSIM | LPIPS | Acc. | PSNR | SSIM | LPIPS |
| Frame-level, render domain: decode rendered frames | ||||||||||||
| CIN | 67.66 | 35.62 | .9814 | .0261 | 64.22 | 34.96 | .9794 | .0290 | 59.64 | 34.49 | .9780 | .0278 |
| HiDDeN | 82.21 | 31.48 | .9550 | .0425 | 79.11 | 31.07 | .9566 | .0375 | 76.85 | 31.67 | .9556 | .0478 |
| StegaStamp | 85.12 | 33.03 | .9793 | .0472 | 80.59 | 32.64 | .9718 | .0554 | 76.09 | 32.32 | .9707 | .0540 |
| SSL | 85.24 | 30.73 | .9651 | .0436 | 81.52 | 29.98 | .9620 | .0486 | 78.37 | 29.56 | .9596 | .0504 |
Appendix figures & tables22 assets
Supplementary material from the paper’s appendix.
Appendix
| Condition | exact | mis-assigned | removed entries falsely matched | WM-PSNR / | bits |
| A Unmodified file | 192/192 | 0 | – | 0 / 0 | 100% |
| B Rows permuted | 192/192 | 0 | – | 0 / 0 | 100% |
| C 30% lowest opacity deleted, permuted | 117/192 | 0 | 123 (0.008%) | 0.006 dB / | 100% |
| D 20% random deleted, permuted | 28/192 | 0 | 1,189 (0.116%) | 0.017 dB / | 100% |
| Condition | min | median | max | decisions at | ||
| , retrospective | correct key, distributed model | 80 | 0.9910 | 0.9991 | 1.0000 | 80/80 detected |
| after INT8 | 80 | 0.9766 | 0.9963 | 0.9995 | 80/80 detected | |
| after pruning | 80 | 0.8147 | 0.9103 | 0.9737 | 80/80 detected | |
| , retrospective | other payload, same scene | 320 | 0.0951 | 0.0060 | 0.3170 | 0 false positives |
| wrong subset | 320 | 0.0382 | 0.1417 | 0.4858 | 0 false positives | |
| unmarked model (64 defined) | 80 | 0.0285 | 0.1400 | 0.3681 | 0 false positives |
| Image payload wm1 (30 scenes) | All image payloads | Bit payloads (42 runs) | ||||||
| SH degree | PSNR [worst] | aware | passes aware | passes naive | passes aware (naive) | bit accuracy | margin | PSNR vs. clean ( ) |
| 3 | 0.00 | 0.9997 | 30/30 | 30/30 | 150 (150) | 100% | 0.998 | 44.40 |
| 2 | 0.76 [ 3.41] | 0.749 | 30/30 | 19/30 | 150 (51) | 100% | 0.561 | 38.56 |
| 1 | 1.36 [ 4.54] | 0.500 | 0/30 | 0/30 | 36 † (0) | 100% | 0.249 | 36.69 |
| 0 | 2.27 [ 9.66] | 0.251 | 0/30 | 0/30 | 0 (0) | 100% | 0.062 | 33.75 |
| Dataset | degree | PSNR (dB) | aware / naive | passes aware / naive | bit models: PSNR (dB) |
| D-NeRF | 2 | 0.523 | 0.749 / 0.628 | 7/7 / 5/7 | 43.24 |
| 1 | 1.142 | 0.500 / 0.302 | 0/7 / 0/7 | 40.89 | |
| 0 | 2.410 | 0.251 / 0.076 | 0/7 / 0/7 | 37.40 | |
| HyperNeRF | 2 | 0.243 | 0.749 / 0.619 | 17/17 / 10/17 | – |
| 1 | 0.785 | 0.500 / 0.300 | 0/17 / 0/17 | – | |
| 0 | 1.420 | 0.251 / 0.079 | 0/17 / 0/17 | – |
| Scene / Split | PSNR | PSNR | PSNR | SSIM | LPIPS | WM-PSNR | Pearson |
| D-NeRF (7 synthetic monocular scenes) | |||||||
| Bouncingballs | 40.378 | 40.265 | 0.113 | 0.9940 | 0.0112 | 41.39 | 0.9994 |
| Hellwarrior | 28.731 | 28.709 | 0.022 | 0.9720 | 0.0296 | 42.54 | 0.9995 |
| Hook | 32.774 | 32.742 | 0.032 | 0.9760 | 0.0201 | 45.29 | 0.9998 |
| Jumpingjacks | 35.497 | 35.462 | 0.035 | 0.9855 | 0.0159 | 45.20 | 0.9998 |
| Mutant | 37.523 | 37.458 | 0.065 | 0.9875 | 0.0106 | 46.28 | 0.9998 |
| D-NeRF (7) | HyperNeRF (17) | DyNeRF (6) | ||||||
| Payload ( ) | Content | WM-PSNR | WM-PSNR | WM-PSNR | ||||
| wm1 ( ) | ICLR logo, flat colour | 47.81 | 0.9999 | 42.02 | 0.9996 | 42.48 | 0.9997 | 0.029 |
| wm2 ( ) | Road/bridge at dusk | 48.61 | 0.9999 | 42.28 | 0.9995 | 42.56 | 0.9996 | 0.368 |
| wm3 ( ) | Bold “SMILE” on white | 44.02 | 0.9998 | 32.41 | 0.9963 | 32.97 | 0.9972 | – |
| wm4 ( ) | Detailed colour poster | 41.70 | 0.9995 | 32.59 | 0.9956 | 32.85 | 0.9961 | 0.214 |
| wm5 ( ) | Anime group scene | 41.71 | 0.9995 | 32.06 | 0.9955 | 32.10 | 0.9959 | 0.067 |
| Scene | wm1 | wm2 | wm3 | wm4 | wm5 | ratio |
| coffee_martini | 0.067 / 0.061 | 0.066 / 0.076 | 0.268 / 0.184 | 0.152 / 0.146 | 0.252 / 0.146 | 1.31 |
| cook_spinach | 0.111 / 0.106 | 0.098 / 0.076 | 1.084 / 0.395 | 0.539 / 0.316 | 0.557 / 0.337 | 1.94 |
| cut_roasted_beef | 0.220 / 0.009 | 0.041 / 0.037 | 0.593 / 0.092 | 0.728 / 0.037 | 0.549 / 0.045 | 9.64 |
| flame_salmon_1 | 0.087 / 0.017 | 0.062 / 0.020 | 0.286 / 0.037 | 0.176 / 0.006 | 0.491 / 0.055 | 8.14 |
| flame_steak | 0.130 / 0.026 | 0.083 / 0.011 | 0.862 / 0.031 | 0.457 / 0.010 | 0.705 / 0.062 | 15.96 |
| sear_steak | 0.262 / 0.181 | 0.138 / 0.133 | 1.257 / 0.462 | 0.756 / 0.388 | 0.667 / 0.403 | 1.96 |
| PSNR | WM-PSNR | Pearson | |
| 0.01 | 0.011 | 26.82 | 0.9770 |
| 0.05 | 0.022 | 36.16 | 0.9969 |
| 0.10 | 0.063 | 38.32 | 0.9980 |
| 0.20 | 0.250 | 40.17 | 0.9986 |
| 0.50 | 1.377 | 43.07 | 0.9994 |
| Metric | hook (D-NeRF) | tamping (HyperNeRF) | flame_salmon_1 (DyNeRF) | |
| Image | PSNR (dB) | [ , ] | ||
| WM-PSNR (dB) | ||||
| Pearson | ||||
| after pruning | ||||
| after pruning | [ , ] | |||
| at SH degree 2 (aware) |
| bit acc. | PSNR (dB) | SSIM | LPIPS | per-slot RMS | file RMS | energy | |||
| 0.01 | 100% | 62.48 [56.99, 67.40] | 0.99994 | 0.00006 | 0.00998 | 0.00316 | 1.2% | 72.4% | 57.8% |
| 0.05 | 100% | 50.32 [47.81, 53.63] | 0.99875 | 0.00095 | 0.04988 | 0.01577 | 6.0% | 99.2% | 79.9% |
| 0.10 | 100% | 44.29 [41.57, 47.62] | 0.99564 | 0.00373 | 0.09981 | 0.03156 | 12.0% | 100% | 95.1% |
| Hide-in-Motion (ICRA 2025) | SHield | |
| Representation | E-D3DGS with an extra composite per-Gaussian attribute that also drives the deformation | unchanged 4DGS model |
| When embedded | jointly during reconstruction training | post hoc on a trained model |
| Carrier | learned composite feature | existing canonical SH of key-selected Gaussians |
| Extraction | feature map rendered at a check viewpoint, private U-Net decoder | closed-form arithmetic against stored reference SH; no rendering or learned decoder |
| Payloads | 2D images; text, QR codes, audio, video via modality-specific decoders | RGB images, bit messages |
| Attacks evaluated | none reported | quantization, reordering, removal, SH truncation, SH noise (bits), centre perturbation |
| Dynamic rendering | Recovery | ||||||
| Method | Base | PSNR | SSIM | LPIPS | PSNR | PSNR | SSIM |
| E-D3DGS ( Bae et al., 2024 ) | – | 25.641 | 0.714 | 0.283 | – | – | – |
| Hide-in-Motion | E-D3DGS | 25.573 | 0.709 | 0.297 | 0.068 | 50.341 | 0.998 |
| 4DGS ( Wu et al., 2024 ) | – | 25.000 | 0.676 | 0.394 | – | – | – |
| SHield (ours) | 4DGS | 24.950 | 0.674 | 0.395 | 0.049 | 36.345 | 0.921 |
| Dynamic rendering | Recovery | ||||||
| Method | Base | PSNR | SSIM | LPIPS | PSNR | PSNR | SSIM |
| E-D3DGS ( Bae et al., 2024 ) | – | 31.500 | 0.938 | 0.149 | – | – | – |
| Hide-in-Motion | E-D3DGS | 31.471 | 0.935 | 0.153 | 0.029 | 50.075 | 0.998 |
| 4DGS ( Wu et al., 2024 ) | – | 31.192 | 0.936 | 0.109 | – | – | – |
| SHield (ours) | 4DGS | 31.065 | 0.935 | 0.110 | 0.127 | 36.591 | 0.927 |
| D-NeRF (7) | HyperNeRF (17) | DyNeRF (6) | |||||
| Method | View type | WM-PSNR | WM-PSNR | WM-PSNR | |||
| 2D per-frame | Same-view (training) | 14.34 | 0.917 | 15.28 | 0.943 | 15.30 | 0.943 |
| 2D per-frame | Novel-view (test) | 6.49 | 0.010 | 7.78 | 0.000 | 8.30 | 0.011 |
| SHield (ours) | model file | 44.77 | 0.9997 | 36.27 | 0.9973 | 36.59 | 0.9977 |
| Attribute | D-NeRF (7) | HyperNeRF (17) | DyNeRF (6) |
| SH DC ( ) | 0.0084 | 0.0083 | 0.0085 |
| SH rest ( ) | 0.0191 | 0.0123 | 0.0088 |
| SH combined | 0.0086 | 0.0083 | 0.0093 |
| Scene | params | steps | PSNR | re-assoc. | oracle | matched (%) | verifies |
| bouncingballs | sh | 100 | 0.04 | 0.9808 | 0.9808 | 100.0 | yes |
| bouncingballs | sh | 500 | 0.03 | 0.9257 | 0.9257 | 100.0 | yes |
| bouncingballs | sh | 2,000 | 0.26 | 0.8035 | 0.8035 | 100.0 | yes |
| bouncingballs | sh | 5,000 | 0.59 | 0.6202 | 0.6202 | 100.0 | yes |
| bouncingballs | sh | 10,000 | 0.98 | 0.4782 | 0.4782 | 100.0 | no |
| bouncingballs | sh | 20,000 | 1.51 | 0.3701 | 0.3701 | 100.0 | no |
| Attack | Matcher | Matched (%) | Passes | |
| Duplicate | shipped | 89.96 | 0.9795 | 30/30 |
| lowest-numbered | 100.00 | 0.9997 | 30/30 | |
| Duplicate | shipped | 49.91 | 0.8544 | 30/30 |
| lowest-numbered | 100.00 | 0.9997 | 30/30 | |
| Duplicate | shipped | 0.00 | 0.0285 | 0/30 |
| lowest-numbered | 100.00 | 0.9997 | 30/30 |
Robust Watermarking
Robust watermarking embeds or verifies ownership and provenance signals that survive editing, compression, model changes, and deliberate removal.
See all 228 papers