- Rating
- 2
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
StrataMesh reconstructs compact triangle meshes of large scenes from multiview images. It assigns geometrically reliable regions to a Core and weakly constrained regions to a Shell for appearance modeling. These roles evolve during training. Candidate refinements are ranked by conditional fitting gain per added face. This gain measures the benefit of new degrees of freedom beyond the existing parameters. Collapses recover faces under a scene budget. Geometry and appearance are transferred after topology edits, followed by optimization with a triangle-soup proxy, diffusion, and geometric validity checks.
Critical strengths
-
The allocation criterion is well motivated. Conditional gain identifies errors beyond the capacity of existing geometry and appearance parameters. Removing it increases face count from 2.94M to 8.68M without improving F1, PSNR, or LPIPS, supporting its role in reducing redundant refinement.
-
The meshes are compact with good triangle quality. On T&T, StrataMesh uses fewer faces than MILo-base. Only 0.845% of its triangles have q ≤ 0.1, compared with roughly 8% for MILo, with no observed degenerate faces. On Mip-NeRF 360, it achieves the highest PSNR and SSIM among the evaluated methods and nearly the best LPIPS with fewer faces than MILo-dense.
-
Resource use is low and evaluation is reasonably broad. On T&T, peak GPU memory is 4.61 GiB, with lower memory use and runtime than MILo and GGGS. The paper provides per-scene results and ablations of representation roles, budget transfer, and optimization strategies.
Critical weaknesses
-
Conditional gain lacks direct empirical validation. The paper does not test whether predicted gain per added face corresponds to actual improvement during training. The ablation's replacement criterion, candidate counts, and scoring time are unspecified.
-
MeshSplatting is omitted. MeshSplatting (CVPR 2026) optimizes opaque meshes for standard graphics engines and compares with MILo on Mip-NeRF 360 and T&T. This closely related method is neither cited nor evaluated.
-
Shell initialization is unclear. The main paper initializes an empty Shell, but the appendix refines it before the first explicit role transition. The initial construction step is missing. The relationship between source-face deactivation in Appendix D.1 and migration from Core to Shell also needs clarification.
-
Key experimental settings are missing. The paper does not specify the calibration procedure used, target error, extra face reserve for conforming refinement and quality constraints, migration thresholds, persistence length, or gain damping.
-
Data inconsistencies and missing definitions hinder verification. Figure 1 and Table 7 report different face counts for GOF (24.53M vs. 23.10M) and RaDe-GS (22.67M vs. 63.12M). Any configuration or postprocessing differences should be explained. The appendix also contains an undefined loss symbol (p. 36), three unresolved references (pp. 36 and 42), and two duplicated sentences (p. 26, lines 1366 to 1371).
Questions for authors
-
What refinement criterion replaces conditional gain in the ablation? Can you test whether predicted gain per added face corresponds to actual improvement on a small set of candidates and report the scoring time?
-
Can you include a comparison with MeshSplatting?
-
How are the initial Shell geometry and appearance constructed? How is the total face budget determined in the experiments, and what values are used for the key parameters listed above?
7 comments
Response (2)
W2. Comparison with MeshSplatting
Thank you for suggesting this comparison. Section 2 of the main paper cites MeshSplatting (Held et al., CVPR 2026). Appendix B.6 evaluates both methods on Tanks and Temples.
- Higher reconstruction fidelity with fewer faces. StrataMesh achieves higher F1 on all five scenes with ground-truth geometry (Appendix Table 23), with nearly three times the mean F1 (0.560 vs. 0.188). It uses fewer faces on four of those scenes, with 34% fewer faces on average (3.12M vs. 4.70M). On the rendered images, StrataMesh achieves lower mean LPIPS (0.296 vs. 0.353).
- Better-shaped triangles and cleaner surfaces. StrataMesh leads on both triangle-shape metrics in each of the six scenes (Appendix Table 24). Mean triangle quality is 0.689 vs. 0.540, and the fraction of low-quality triangles () is 0.766% vs. 3.971%. Appendix Fig. 8 shows paired RGB and mesh crops on Truck and Train. In the RGB crops, the truck fender and the train steps show smoother surfaces with fewer visible creases. The mesh crops show more regular triangulation around the truck wheels and cleaner boundaries around the train steps and side vents. They also illustrate how StrataMesh uses finer triangles near geometric details and coarser faces on simpler surfaces. The comparison highlights local surface continuity and triangle layout in the shown regions.
- Lower peak GPU memory. StrataMesh uses 4.83 vs. 24.92 GiB of mean peak GPU memory across the five scenes with F1. It maintains indexed connectivity throughout fitting using local, budgeted splits and collapses with manifold checks. MeshSplatting builds connectivity through restricted Delaunay triangulation, then fine-tunes the connected opaque mesh. The peak memory numbers cover the full pipelines.
Reconstruction F1, LPIPS, and mesh resources on five scenes
| Scene | Method | F1 ↑ | LPIPS ↓ | Faces (M) ↓ | Peak GPU (GiB) ↓ |
|---|---|---|---|---|---|
| Truck | MeshSplatting | 0.29 | 0.236 | 3.31 | 9.11 |
| Truck | StrataMesh | 0.62 | 0.261 | 2.11 | 3.12 |
| Barn | MeshSplatting | 0.12 | 0.357 | 4.78 | 30.24 |
| Barn | StrataMesh | 0.61 | 0.297 | 2.69 | 4.12 |
| Caterpillar | MeshSplatting | 0.15 | 0.437 | 4.39 | 30.14 |
| Caterpillar | StrataMesh | 0.45 | 0.374 | 2.67 | 3.97 |
| Ignatius | MeshSplatting | 0.24 | 0.395 | 4.4 | 24.83 |
| Ignatius | StrataMesh | 0.76 | 0.258 | 5.4 | 8.79 |
| Meetingroom | MeshSplatting | 0.14 | 0.338 | 6.6 | 30.26 |
| Meetingroom | StrataMesh | 0.36 | 0.289 | 2.72 | 4.13 |
| Mean (5 scenes) | MeshSplatting | 0.188 | 0.3526 | 4.696 | 24.916 |
| Mean (5 scenes) | StrataMesh | 0.560 | 0.296 | 3.118 | 4.826 |
Triangle quality on all six scenes
Train is included in the shape comparison. The F1 table covers the five scenes with reference geometry.
| Scene | Method | Mean q ↑ | q ≤ 0.1 (%) ↓ |
|---|---|---|---|
| Truck | MeshSplatting | 0.552 | 2.831 |
| Truck | StrataMesh | 0.676 | 0.987 |
| Barn | MeshSplatting | 0.501 | 6.290 |
| Barn | StrataMesh | 0.679 | 0.700 |
| Caterpillar | MeshSplatting | 0.544 | 4.363 |
| Caterpillar | StrataMesh | 0.685 | 0.598 |
| Ignatius | MeshSplatting | 0.565 | 3.054 |
| Ignatius | StrataMesh | 0.690 | 0.784 |
| Meetingroom | MeshSplatting | 0.513 | 4.912 |
| Meetingroom | StrataMesh | 0.733 | 0.856 |
| Train | MeshSplatting | 0.565 | 2.376 |
| Train | StrataMesh | 0.671 | 0.672 |
| Average | MeshSplatting | 0.540 | 3.971 |
| Average | StrataMesh | 0.689 | 0.766 |
Both methods report zero degenerate faces under the triangle-shape criterion.
Response (3)
W3. Shell initialization and source-face deactivation
Thank you for pointing out the missing step in the initialization schedule and asking how deactivation relates to role migration.
- The initial Shell is created before the first Shell refinement. At initialization, we build a colored patch in each occupied octree cell from its SfM points and their RGB values, then stitch the patches into one indexed mesh and assign all faces to Core (Sec. 3 of the main paper). We render this mesh from the calibrated training views and estimate parallax for each visible region. In the first role assignment, stronger-parallax regions remain in Core and weaker-parallax regions form the initial Shell. Shell vertex colors are initialized from the Core-only RGB renders. Panel 1 of the Ignatius walkthrough (Appendix Fig. 4) illustrates this assignment. The initial budget reallocation then refines the Shell that is already in place, followed by Shell appearance warm-up and joint Core–Shell fitting. The revised reconstruction schedule in Appendix E.4 now shows the initial assignment before the first refinement.
- Deactivation is the Core-side step of a later Core-to-Shell migration. Once a region meets the persistent demotion criteria in Appendix E.1 and the migration is accepted, we deactivate its source faces in Core. The faces stop contributing to the Core render and receiving Core updates. The reservoir retains the source geometry for possible later promotion while it remains compatible with the current topology. If a Shell patch is nearby, we extend it to cover the region. Otherwise, we detach the selected patch from the Core connectivity and reassign it as a Shell patch. The region then receives Shell geometry and appearance updates.
W4. Experimental budget and key settings
- Budget calibration and target error. The reported experiments use the simpler gradient-based estimate described in the revised Appendix C.1.4. This replaces the earlier rule based on calibrated regional coefficients and a prescribed target error. We identify adjacent pairs of vertices whose gradient magnitudes exceed , estimate the vertex count assuming one insertion per pair, and refresh the budget estimate every few iterations.
- Extra face reserve. We use a shared Core–Shell face budget with zero extra reserve. Each accepted refinement is charged its full face-count increase, including conformity splits on neighboring faces (Sec. 3.1.2 of the main paper and Appendix E.2.1). QAMO maintains mesh quality through regularization and geometric validity checks during fitting (Sec. 3.2 of the main paper).
- Migration, persistence, and damping. For Core-to-Shell demotion, a face is eligible if it has fewer than two visible views or fewer than one parallax-supporting view and its mean squared residual is at or below the 35th percentile among observed Core faces. The condition must hold for two consecutive checks (Appendix E.1). Conditional-gain damping is (Appendix E.2.1 gives the scoring equation).
W5. Face-count discrepancies and appendix corrections
Thank you for catching these inconsistencies.
- Truck results in Figure 1. The teaser in Figure 1 of the main paper is intended to summarize the Truck evaluation reported in Table 7 in Appendix B.2. Two face-count labels were entered incorrectly when preparing the teaser. Appendix Table 7 gives the correct counts of 23.10M for GOF and 63.12M for RaDe-GS. We have corrected the teaser to use these face counts from Appendix Table 7.
- Appendix corrections. In the revised appendix, p. 40 defines the image-loss gradient and resolves the two proof references. Appendix E.2.3 (p. 45) gives the quota derivation in place of the unresolved reference. The state-transfer passage in Appendix C.2.4 (p. 30) now contains one copy of the previously duplicated sentences.
Response (1)
Thank you for pointing out the missing implementation details and for suggesting MeshSplatting as a baseline. We respond to each weakness and question below.
Critical Weaknesses
W1. Empirical validation and scoring details for conditional gain
Higher predicted gain per face leads to greater realized improvement in the Ignatius group comparison. We refine the regions ranked 1–5 and 6–10 in two separate runs from the same checkpoint at step 12,000. Each run refines several patches together because the change caused by a single split can be smaller than the loss fluctuations across training views. The groups add similar numbers of faces, and each branch receives 1,000 QAMO steps. Scoring uses fixed visibility. After the splits are applied, QAMO rerenders the edited mesh and recomputes visibility and occlusion.
The higher-ranked group achieves 2.1× the loss reduction of the lower-ranked group (Appendix Fig. 10 and Table 27). This tests whether the ranking directs a comparable face increase to regions that benefit more from refinement after nonlinear fitting. The loss is measured for each group of edits. Individual split correlations require separate trials with controlled view sampling.