- Rating
- 3
- Confidence
- 4
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper addresses the loss of sub-voxel geometric information in discrete semantic occupancy representations. The authors propose QuadricUNet, which predicts local implicit quadric surfaces from semantic voxel grids, enabling continuous surface reconstruction and analytical LiDAR ray intersection. Experiments on SemanticKITTI and Waymo demonstrate improved geometric reconstruction over the evaluated baselines, while simulated LiDAR sweeps yield better sim-to-real segmentation performance.
Critical strengths
-
The paper addresses the loss of geometric detail caused by coarse voxelization, which is particularly relevant to 3D scene reconstruction and physics-aware sensor simulation. Representing sub-voxel surfaces using learned quadrics provides a meaningful connection between data-driven geometric inference and analytical operations such as ray intersection.
-
The method achieves substantial improvements over the evaluated baselines on SemanticKITTI and Waymo. Importantly, the improvements extend beyond geometric metrics to sim-to-real LiDAR segmentation, where the macro-mIoU increases from 0.3672 with Marching Cubes to 0.4806 with QuadricUNet, providing evidence of practical downstream benefits.
Critical weaknesses
-
The main baselines are Marching Cubes and NKSR, neither of which is trained specifically to recover sub-voxel geometry from the given semantic occupancy representation. However, existing approaches such as Neural Dual Contouring can learn surface geometry directly from binary voxel grids without requiring input surface normals. Although their output representations differ, the absence of such learning-based comparisons makes it difficult to establish whether the reported reconstruction gains arise specifically from the proposed quadric representation or more generally from learned geometric priors.
-
Appendix A states that the input and supervised voxel sets are restricted to voxels for which reference quadrics can be fitted from aggregated LiDAR observations. This reference-dependent filtering introduces an additional condition into the evaluation inputs, while all reported experiments use ground-truth rather than predicted occupancy grids. Consequently, the results demonstrate effectiveness under an idealized and filtered occupancy setting, but do not yet establish robustness to independently generated or imperfect occupancy representations.
Questions for authors
-
Could the authors provide comparisons with learning-based voxel-to-surface methods, particularly Neural Dual Contouring or a voxel-conditioned implicit surface decoder, under comparable input and supervision settings? A learned planar representation would also help isolate the contribution of second-order geometry.
-
Could the authors clarify how many occupied voxels are excluded by the reference-quadric fitting requirement in Appendix A, and whether this filtering uses geometric information unavailable in a realistic inference setting? Results using occupancy inputs constructed independently of the reference LiDAR geometry would help establish the generality of the voxel-only claim.
-
Table 6 reports a substantial speedup in surface sampling from the same fitted quadric field, but excludes network inference and shows higher memory consumption. Could the authors provide end-to-end computational comparisons, including geometry prediction and LiDAR simulation, to better establish the practical efficiency advantages of the complete framework?
4 comments
Response to Reviewer AC4r [2/2]
Q3. Could the authors provide an end-to-end cost comparison?
We revised the cost tables to time every method end-to-end on the GPU.
Table 1. Dense sampling, per frame. Means over 97 frames of sequence 08 (SemanticKITTI) on one NVIDIA H200. NKSR's network time includes its normal estimation and solver.
| Method | Network (ms) | Field queries (ms) | Mesh extraction (ms) | Surface sampling (ms) | End to end (ms) | GPU memory (GB) |
|---|---|---|---|---|---|---|
| Reference quadrics (upper bound) | – | – | – | 77.3 | 77.3 | 1.3 |
| QuadricUNet | 25.9 | – | – | 79.0 | 104.9 | 1.5 |
| QuadricUNet (SDF head) | 25.8 | 182.7 | 43.8 | 18.9 | 271.1 | 8.1 |
| UniScene | 11.9 | 219.4 | 36.0 | 16.4 | 283.7 | 6.5 |
| NKSR | 1964.8 | – | 396.8 | 14.3 | 2375.9 | 3.4 |
| Marching Cubes | – | – | 1.6 | 10.6 | 12.1 | 1.0 |
Table 2. One LiDAR sweep, per frame.
| Method | Network (ms) | Mesh extraction and build (ms) | Ray casting or rendering (ms) | End to end (ms) | GPU memory (GB) |
|---|---|---|---|---|---|
| Reference quadrics (upper bound) | – | – | 24.0 | 24.0 | 0.5 |
| QuadricUNet | 25.5 | – | 23.1 | 48.6 | 1.4 |
| QuadricUNet (SDF head), volume rendering | 25.6 | – | 35.4 | 61.0 | 1.9 |
| QuadricUNet (SDF head), sphere tracing | 26.2 | – | 1799.6 | 1825.9 | 1.4 |
| UniScene, volume rendering | 10.6 | – | 39.6 | 50.2 | 2.2 |
| UniScene, sphere tracing | 10.7 | – | 951.1 | 961.8 | 0.9 |
| NKSR | 1964.8 | 401.7 | 39.4 | 2405.9 | 6.5 |
| Marching Cubes | – | 2.4 | 23.8 | 26.2 | 2.2 |
Table 3. Training and storage.
| Method | Training (min) | Stored per frame (MB) |
|---|---|---|
| QuadricUNet | 92 | 4.0, 10 numbers per voxel |
| QuadricUNet (SDF head) | 129 | 256, dense volume |
| UniScene | 125 | 12.7 sparse, 32 numbers per voxel; 256 dense |
| NKSR | n/a | 50.9, mesh |
- Dense sampling: QuadricUNet is the fastest learned method and uses the least GPU memory of all learned baselines. It is more than twice as fast as both learned baselines and more than an order of magnitude faster than NKSR. Its sampling step costs about as much as sampling the reference quadrics, so the predicted coefficients add no overhead. Marching Cubes is faster because it involves no learned model, but its surfaces are far less accurate.
- One LiDAR sweep: QuadricUNet matches the speed of UniScene's volume rendering at a clearly lower error, and is far faster than sphere tracing either implicit model. Its ray casting is the cheapest step of all learned methods. Marching Cubes is faster because its sweep is a mesh ray cast, but its error is several times higher.
- Same U-Net with a query head: QuadricUNet (SDF head) has the same network cost, so the whole gap is the head. It is slower on both outputs and needs several times more GPU memory for dense sampling and far more storage. The quadric head, not the U-Net, is what makes QuadricUNet cheap.
- Storage: QuadricUNet stores fewer numbers per voxel than any learned baseline. Its output is several times smaller than UniScene's sparse features and an order of magnitude smaller than NKSR's mesh.
The savings come from needing no network at query time.
Thank you for the detailed rebuttal and additional experiments. The new learning-based baselines and end-to-end efficiency comparisons adequately address my main concerns and strengthen the paper. I appreciate the authors' efforts and maintain my positive recommendation (3: Weak Acceptance).
Response to Reviewer AC4r (follow-up)
We thank the reviewer for their response and for maintaining the positive score. We are glad we could address your concerns. Your constructive feedback helped strengthen our submission.
Response to Reviewer AC4r [1/2]
Thank you for the detailed and valuable feedback.
W1 and Q1. Could the authors provide comparisons with learned baselines under comparable input and supervision settings?
We added two learned models with our exact input. QuadricUNet (SDF head) is the voxel-conditioned implicit decoder as suggested. It keeps our U-Net and replaces the quadric head with a query MLP. UniScene is a published occupancy-to-LiDAR model [1], trained on our data.
This comparison separates the two factors. In Table 1, the SDF head is on par with QuadricUNet in Chamfer distance but has a clearly lower F1@10. It predicts surfaces in most of the voxel volume which lowers the Chamfer distance (good completeness) but also ends up predicting incorrect surfaces (low accuracy), which F1 penalises. Learning-based setups overall have an edge in reconstruction performance. Our representation has the edge on accuracy and LiDAR results back this: on the same backbone, the query decoder's LiDAR error is much higher. Rendering QuadricUNet through UniScene's slower volume renderer gives the same per-ray error as our analytical intersection (14.6 cm with both), so the gap is not due to the renderer.
We thank the reviewer for pointing to Neural Dual Contouring [2]. QuadricUNet (SDF head) addresses the same question and is trained on the same reference points. We do not train a learned planar representation, as fitted planes reach 4.53 cm Chamfer and 61.8% F1@5 (Table 1), against 3.92 cm and 66.9% for fitted quadrics. This shows that the representational capacity of planes is already lower than quadrics for our setting.
W2 and Q2. Could the authors clarify the effect of filtering and the results in a realistic inference setting? Results using occupancy inputs constructed independently of the reference LiDAR geometry would help establish the generality of the voxel-only claim.
The filter was used to clean up noisy voxels during preprocessing (10.5% of the static voxels on sequence 08, which hold 0.011% of the reference points). It therefore uses LiDAR points, but only at dataset creation. We tested and found that a model trained without it performs similarly (4.98 cm Chamfer without the filter against 4.97 ± 0.04 cm with it), as the removed voxels hold almost no reference points. Inference requires only the semantic voxel grid. We agree that all results are uniformly evaluated on ground-truth grids. We avoid evaluation on predicted occupancy grids for two reasons. First, the ground-truth grids are built from the LiDAR-mapped surfaces that serve as our reference, whereas a predicted grid need not align with the reference, so the reference cannot score it meaningfully. Second, ground-truth grids keep the errors of upstream occupancy prediction models out of our evaluation. Predicted grids therefore need their own evaluation protocol, which we leave to future work.
The end-to-end cost comparison (Q3) follows in the next comment.
References
[1] B. Li et al., "UniScene: Unified Occupancy-centric Driving Scene Generation," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
[2] Z. Chen et al., "Neural Dual Contouring," ACM Transactions on Graphics (TOG), vol. 41, no. 4, 2022.