Neural Compression for 3D Geometry Sets¶
Conference: ICCV 2025 arXiv: 2405.15034 Code: GitHub Area: 3D Vision Keywords: 3D geometry compression, neural compression, TSDF, auto-decoder, geometry sets
TL;DR¶
This paper proposes NeCGS, the first neural compression paradigm capable of compressing geometry sets containing thousands of diverse 3D mesh models at ratios up to 900×, achieving high-fidelity reconstruction via a TSDF-Def implicit representation and a quantization-aware auto-decoder.
Background & Motivation¶
3D mesh models are widely used in computer graphics, VR, robotics, and related fields. As geometric data grows increasingly complex, efficient compression techniques become essential.
Limitations of prior work:
- Voxel/point-cloud-based methods (GPCC/VPCC) require high resolutions (≥\(2^{10}\)) for accurate representation, introducing redundancy
- SDF/TSDF representations require tensors of varying sizes, and complex models demand extremely large tensors
- Neural implicit methods (DeepSDF) exhibit limited capacity when handling large collections of models from diverse categories
- Most methods target single models or temporally correlated sequences, and cannot handle diverse, unrelated geometry sets
Method¶
Two-Stage Pipeline¶
Stage 1: Regular Geometry Representation (RGR) — Converts irregular 3D mesh models into unified, fixed-size regular 4D tensors. Stage 2: Compact Neural Representation (CNR) — Exploits intra- and inter-model geometric similarity via a quantization-aware auto-decoder.
TSDF-Def Representation¶
Extends the conventional TSDF by introducing an additional deformation at each grid point:
where \(\mathbf{V} \in \mathbb{R}^{K \times K \times K \times 4}\), and the deformation offsets are consumed during surface extraction via Differentiable Marching Cubes (DMC).
Optimization objective: $\(\min_{\mathbf{V}} \mathcal{E}_{Rec}(\texttt{DMC}(\mathbf{V}), \mathbf{S}) + \lambda_{Reg}\|\mathbf{V}[...,1:3]\|_1\)$
The \(\ell_1\) regularization suppresses unnecessary deformations, since most regions can already be accurately represented by TSDF alone.
Quantization-Aware Auto-Decoder¶
Each model \(\mathbf{V}_i\) is associated with a latent feature \(\mathbf{F}_i \in \mathbb{R}^{K' \times K' \times K' \times C}\), where \(K' \ll K\):
Differentiable quantization \(\mathcal{Q}(\cdot)\) is integrated into training to reduce quantization error.
Loss & Training¶
where \(\mathbf{M}_i\) is a mask assigning higher weight to grid points near the surface.
Entropy Coding¶
Quantized latent features and network parameters are compressed into a bitstream via Huffman coding.
Key Experimental Results¶
Compression Efficiency Across Datasets¶
| Method | Compression Time (h) | Decompression Time (ms) |
|---|---|---|
| GPCC | 0.62 | 562.56 |
| VPCC | 39.34 | 762.87 |
| PCGCv2 | 1.76 | 100.32 |
| Draco | 0.06 | 365.18 |
| NeCGS | 10.01 | 98.95 |
Ablation Study: TSDF vs. TSDF-Def¶
| Representation | CD↓ | NC↑ | F1-0.005↑ |
|---|---|---|---|
| TSDF K=64 | High | Low | Low |
| TSDF K=128 | Medium | Medium | Medium |
| TSDF-Def K=64 | Low | High | High |
| TSDF-Def K=128 | Lowest | Highest | Highest |
Key Findings¶
- NeCGS achieves compression ratios approaching 900× on the DT4D dataset while preserving geometric details.
- TSDF-Def preserves fine structures such as thin surfaces at low resolution (K=64), whereas conventional TSDF fails even at K=128.
- NeCGS achieves the fastest decompression speed (98.95 ms), which is critical for downstream applications.
- The framework supports dynamic scenes: new models can be incrementally added to an already-compressed set.
Highlights & Insights¶
- Elegant design of TSDF-Def — Introducing per-grid-point deformation offsets enables low-resolution tensors to represent fine geometric structures, unifying the representation size across models of varying complexity.
- Set-level compression — Exploiting cross-model geometric similarity yields compression efficiency far beyond what single-model methods can achieve.
- Quantization-aware training — Integrating quantization into the training loop effectively reduces quantization error.
- Incremental capability — Supporting dynamic addition of new models substantially enhances practical utility.
Limitations & Future Work¶
- Compression is time-intensive (~10 hours), making it an offline process.
- Performance degrades on mixed datasets with high inter-category diversity.
- The decoder architecture is fixed; different compression ratios are achieved by adjusting decoder size.
Related Work & Insights¶
- Single-model compression: GPCC, VPCC, Draco, PCGCv2
- Sequence compression: SLRMA, SMPL/SMAL-driven methods
- Neural implicit representations: DeepSDF, various SDF/UDF methods
Rating¶
- Novelty: ⭐⭐⭐⭐ (TSDF-Def + set-level neural compression)
- Technical Depth: ⭐⭐⭐⭐ (complete two-stage design, quantization-aware training)
- Experimental Thoroughness: ⭐⭐⭐⭐⭐ (four datasets, comprehensive ablations)
- Value: ⭐⭐⭐⭐ (900× compression ratio, supports dynamic model addition)