- Rating
- 1
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
The paper targets object-level restoration of single-photon LiDAR point clouds, which are sparse, incomplete and contaminated by background detections. It introduces SPORE, a simulated benchmark built from ShapeNet meshes with ray casting, Poisson photon statistics and three corruption levels, and PRISM, a restoration method in LION's hierarchical latent space. PRISM first estimates the reliability of each local latent anchor and repairs unreliable ones using reliable neighbors, then runs coupled global and local flow matching conditioned on the recorded repair states. On SPORE it ranks first on all four metrics under High corruption and is competitive under Low and Mid, with one qualitative example on real SPAD data.
Critical strengths
- The benchmark fills a real gap. Paired clean geometry with physically motivated single-photon corruption across 55 categories and three severity levels doesn't exist yet, and the simulator is documented in enough detail to be reproduced.
- The method is reasonably motivated and the ablation is thoughtfully designed.
Critical weaknesses
- The [email protected] results are hard to trust. The same model gets a higher [email protected] as the input gets worse (0.200, 0.214 and 0.342 for Low, Mid and High), and under Low it scores the same as the noisy input. This suggests the metric mostly reflects sampling alignment with the FPS-sampled reference rather than restoration quality, which matters because the main High-regime gains are on [email protected] and HD95.
- The ablation results are difficult to interpret. Removing a single conditioning component drops [email protected] from 0.342 to 0.05–0.07 while CD changes much less, so it's unclear what each module actually contributes.
- The real-world evaluation is limited to one qualitative example. Since the benchmark is simulated on single ShapeNet objects, the paper doesn't show that the method transfers to real single-photon data.
- The technical novelty is limited. The method mainly combines existing components, such as LION's latent space, flow matching and FiLM-style conditioning, and under Low corruption it is behind PGD and PD-LTS on CD and P2M.
Questions for authors
- Why does PRISM's [email protected] increase with corruption for the same model?
- The autoencoder is trained on clean FPS-sampled clouds. Do PRISM's outputs inherit that sampling pattern? If so, how does PRISM compare with methods that only displace input points under a sampling-insensitive metric?
- Can you provide quantitative real-data results, for example against a reference scan, including the strongest baselines?
- How sensitive is the method to the simulator settings?
18 comments
Response to Question 2
We thank the reviewer for raising this important question. Farthest Point Sampling (FPS) is a well-established sampling strategy in point cloud processing, widely used to obtain spatially distributed representative points. For example, PointNet++ (Qi et al., 2017) employs FPS for hierarchical point sampling. Following this established practice, we used FPS to construct the original 2,048-point clean references for evaluation. Importantly, PRISM does not explicitly apply FPS to its restored outputs during inference. GALR selectively corrects local spatial anchors according to their geometric reliability, while GLFM performs repair-state-conditioned global–local latent transport. The decoder subsequently predicts coordinate residuals relative to the transported local anchors. Thus, the output coordinates arise from input-conditioned geometric restoration rather than explicit FPS resampling.
To assess the influence of reference sampling, we conducted an additional reference-resampling experiment. Keeping the denoised point clouds from all methods unchanged, we randomly sampled 2,048 reference points from the same underlying clean mesh surfaces to construct Random-2048 references and recomputed CD-L2, [email protected], and HD95. The results show that PRISM maintains the highest [email protected] and lowest HD95 across all three corruption regimes. Under High corruption, PRISM achieves a CD-L2 of 5.401, [email protected] of 0.3186, and HD95 of 0.0316, outperforming PGD and PD-LTS on all three metrics. Compared with the original FPS-2048 evaluation, its [email protected] decreases by 6.8%, comparable to the changes observed for PGD (6.2%) and PD-LTS (7.5%). These results indicate that PRISM's performance advantage persists when the reference sampling strategy changes.
| Noise Level | Method | CD-L2 ↓ | P2M ↓ | [email protected] ↑ | HD95 ↓ |
|---|---|---|---|---|---|
| Low | PGD | 0.974 | 0.989 | 0.1638 | 0.0340 |
| Low | PD-LTS | 1.029 | 1.104 | 0.1706 | 0.0347 |
| Low | PRISM | 1.198 | 1.240 | 0.1904 | 0.0292 |
| Mid | PGD | 1.230 | 1.200 | 0.1300 | 0.0363 |
| Mid | PD-LTS | 1.254 | 1.201 | 0.1256 | 0.0384 |
| Mid | PRISM | 1.212 | 1.242 | 0.2045 | 0.0248 |
| High | PGD | 5.533 | 2.632 | 0.0844 | 0.0515 |
| High | PD-LTS | 5.528 | 3.545 | 0.0712 | 0.0574 |
| High | PRISM | 5.401 | 2.154 | 0.3186 | 0.0316 |
Our original bidirectional Point-to-Mesh (P2M) evaluation provides complementary evidence. P2M directly measures point-to-surface and surface-to-point geometric discrepancies against the ground-truth mesh, independently of the discrete reference sampling strategy. Therefore, the P2M values remain unchanged in the reference-resampling experiment. Under High corruption, PRISM achieves a P2M of 2.154, compared with 2.632 for PGD and 3.545 for PD-LTS, corresponding to error reductions of 18.2% and 39.2%, respectively. Together, the reference-resampling experiment and surface-based evaluation indicate that PRISM's improvements reflect geometric restoration beyond alignment with an FPS-specific reference distribution.
Response to Question 3 and Question 4
Q3: Thank you for this comment. We tested PRISM, PGD, and PD-LTS on eight real, experimentally acquired single-photon point-cloud samples. Since clean ground truth has not yet been obtained, we used two no-reference geometric descriptors, local surface variation (SV) and isolated-point ratio (IPR).
After removing exact duplicate points, SV was calculated using each point’s 20 nearest neighbors, excluding the query point itself. It is the ratio of the smallest neighborhood covariance eigenvalue to the sum of all eigenvalues, averaged over all points. Lower SV indicates lower local nonplanarity. IPR measures the percentage of points with fewer than three other neighbors within a specified radius. This radius was set to three times the median nearest-neighbor distance in the deduplicated noisy input and was shared by all methods within each sample. Lower IPR indicates fewer isolated points.
The following table reports the unweighted means across the eight samples. PRISM achieved the lowest IPR on 8/8 samples and the lowest SV on 6/8 samples. These results indicate that PRISM continues to outperform PGD and PD-LTS overall in terms of SV and IPR on the evaluated real single-photon point clouds, providing additional no-reference evidence of its denoising effectiveness in this setting. Detailed visualizations and analysis, together with the quantitative comparison, will be provided in Section 5.3, “REAL-WORLD RESULTS,” of the revised manuscript.
| Metric | Method | Sample 01 | Sample 02 | Sample 03 | Sample 04 | Sample 05 | Sample 06 | Sample 07 | Sample 08 | Mean |
|---|---|---|---|---|---|---|---|---|---|---|
| SV ↓ | Noisy | 0.1969 | 0.1136 | 0.1486 | 0.1944 | 0.0675 | 0.0719 | 0.1358 | 0.0610 | 0.1237 |
| SV ↓ | PGD | 0.1540 | 0.1027 | 0.1188 | 0.1728 | 0.0533 | 0.0697 | 0.1364 | 0.0551 | 0.1078 |
| SV ↓ | PD-LTS | 0.1890 | 0.1065 | 0.1470 | 0.1857 | 0.0584 | 0.0637 | 0.1267 | 0.0432 | 0.1150 |
| SV ↓ | PRISM | 0.1803 | 0.0254 | 0.0945 | 0.1754 | 0.0274 | 0.0285 | 0.1226 | 0.0331 | 0.0859 |
| IPR (%) ↓ | Noisy | 12.99 | 9.58 | 5.22 | 24.92 | 30.73 | 23.47 | 23.00 | 21.08 | 18.87 |
| IPR (%) ↓ | PGD | 1.32 | 3.54 | 0.39 | 0.37 | 29.77 | 19.25 | 18.78 | 20.24 | 11.71 |
| IPR (%) ↓ | PD-LTS | 6.78 | 6.68 | 1.61 | 13.64 | 26.84 | 22.17 | 14.96 | 20.15 | 14.10 |
| IPR (%) ↓ | PRISM | 1.26 | 0.05 | 0.12 | 0.09 | 1.91 | 0.19 | 0.67 | 1.16 | 0.68 |
Q4: We thank the reviewer for raising this important question. To provide a systematic evaluation, we have added a dedicated section, “Robustness to Simulator and Acquisition Shifts,” to the revised Appendix. We conduct 20 Mid-anchored experiments, comprising one nominal reference, 12 single-factor perturbations, three combined corruption settings, and four acquisition-geometry variations. These experiments cover changes in photon emission, background noise, dropout, timing and angular jitter, spurious detections, viewpoint configurations, and field of view. All evaluations use the same trained PRISM model without additional fine-tuning.
The results demonstrate PRISM's consistent restoration capability across the tested parameter shifts. Across all 19 shifted conditions, PRISM improves all four evaluation metrics relative to the corresponding noisy inputs, reducing CD-L2 by 93.4–97.9% and P2M by 92.9–97.9%. For example, increasing the background rate from 0.15 to 0.60 raises the noisy CD-L2 from 32.813 to 66.762, while PRISM maintains a low CD-L2 of 1.421. Even under the combined low-photon and high-dropout condition, PRISM reduces CD-L2 from 180.318 to 8.318 and HD95 from 0.8230 to 0.0387. With only three acquisition views, it further reduces CD-L2 from 38.902 to 2.571 while improving [email protected] from 0.0833 to 0.2421.
These results provide quantitative evidence that PRISM maintains effective geometric restoration across diverse simulator configurations, including substantial changes in photon statistics, sensor noise, and acquisition geometry. The complete parameter settings, quantitative comparisons, and detailed analyses are now included in the updated Appendix.
Response to Question 1
We appreciate the reviewer's careful observation regarding the [email protected] trend across corruption regimes. This behavior can be understood from three perspectives: PRISM's reliability-conditioned restoration mechanism, the precision–recall relationship, and the threshold sensitivity of the F-score.
Reliability-conditioned Restoration. PRISM is designed to address spatially heterogeneous and structured corruption in single-photon point clouds. GALR selectively repairs observations according to their estimated reliability, while the resulting repair states guide GLFM's global–local latent transport. Under Low corruption, the observed geometry is already relatively reliable, leaving limited room for further refinement. Moreover, some points may remain outside the [email protected] distance threshold even after their geometric errors are reduced. Consequently, improvements in geometric accuracy do not necessarily translate into higher F-scores at a fixed threshold. In contrast, stronger corruption provides greater scope for reliability-guided restoration of unreliable observations and missing structures.
Precision–Recall Analysis. Under Low corruption, precision increases from 0.2300 to 0.2500, while recall decreases from 0.1759 to 0.1663, leaving [email protected] nearly unchanged (0.1993 to 0.1997). In contrast, under High corruption, both precision (0.0700 to 0.4200) and recall (0.0465 to 0.2884) improve substantially, increasing [email protected] from 0.0559 to 0.3419. These results reveal a precision–recall trade-off under Low corruption, whereas severe corruption benefits from simultaneous improvements in correspondence accuracy and geometric coverage.
Threshold Sensitivity. The F-score evaluates point-cloud correspondence within a fixed distance threshold and is therefore sensitive to nearest-neighbor distance distributions. At a stricter threshold of 0.0075, F-scores follow Low > Mid > High, with values of 0.1650, 0.1450, and 0.1100, respectively. However, at 0.01, the ordering reverses, with scores of 0.1997, 0.2143, and 0.3419. Increasing the threshold from 0.0075 to 0.01 produces a much larger F-score gain under High corruption (0.2319) than under Low (0.0347) or Mid (0.0693). This indicates differences in nearest-neighbor distance distributions across corruption regimes, suggesting that some geometric improvements may not be fully reflected by a fixed-threshold F-score.
Importantly, this non-monotonic behavior is not unique to PRISM. As reported in Table 1, PD-Flow achieves [email protected] scores of 0.1306 and 0.2781 under Low and High corruption, respectively, while PCN achieves 0.1856 and 0.2293. Both methods obtain higher [email protected] under High corruption than under Low corruption, and neither exceeds the noisy input's [email protected] of 0.1993 under Low corruption. This further suggests that the observed trend is not specific to PRISM but is associated with differences in geometric error distributions and the threshold-dependent nature of the F-score.
Overall, the limited [email protected] improvement under Low corruption may result from the combined effects of limited corrective refinement, the precision–recall trade-off, and fixed-threshold evaluation. The higher [email protected] under High corruption does not imply uniformly lower geometric errors than under Low corruption, but instead reflects differences in the distributions of point-cloud correspondence distances.