Implicit and explicit representations of geometry are central in computer graphics applications and come with their pros and cons. With surface sampling being faster on explicit representations and spatial queries being faster on implicit representations, finding efficient conversions between them is a highly desirable property. We present TetraCoder, a single resolution hash-free adaptive tetrahedral encoding that links an explicit tetrahedral spatial encoder with an underlying implicit signed distance field (SDF). The resulting hybrid implicit-explicit space is continuous piecewise affine (CPWA) by design, allowing to obtain an exact correspondence between the implicit SDF and the explicit analytic mesh. TetraCoder's design comes with additional perks, in particular: (a) it scales better than multi-resolution encoding methods due to its single resolution optimizable grid, (b) unlike hash-based methods, it is not affected by collisions, (c) it enables fast tetrahedral ray-tracing, geometry and appearance editing, directly through its adaptive tetrahedral scaffold. We demonstrate the versatility of TetraCoder by showing an improvement over state-of-the-art methods on multiple tasks: 2D/3D fitting from images and point clouds, 3D novel view synthesis, 3D reconstruction, and geometry/appearance editing.
Figures & tables
Figure 1. We present TetraCoder , an adaptive tetrahedral encoding for neural fields. Multiresolution hash encoding ( left ) interpolates features on fixed spatial grids at multiple resolutions. Its trilinear interpolation does not produce a piecewise-affine encoding, making analytic mesh extraction impractical. To address that tetrahedral multiresolution hash encoding ( center ) subdivides these grids into fixed tetrahedra and uses barycentric interpolation to obtain a piecewise-affine encoding. TetraCoder ( right ) instead uses a single-level, hash-free encoding defined on an adaptive tetrahedral volumetric mesh. Combined with a piecewise-affine decoder, this representation enables fast analytic mesh retrieval. Unlike other encodings, TetraCoder supports novel view synthesis, scene reconstruction, 3D appearance and geometry editing within a unified framework. Three diagrams compare multiresolution Cartesian hash grids, multiresolution tetrahedral hash grids, and a single adaptive tetrahedral mesh.
Figure 2. Tetrahedral ray tracing. For a sample xj along the ray r(t)=o+td , an adjacency walk (purple) locates the containing cell. Shading highlights the visited cells, while green arrows indicate face normals. Dashed lines illustrate barycentric interpolation of vertex features within the target triangle (a,b,c) . Gray markers denote additional samples along the ray.
Image
Method
PSNR (dB) ↑
Parameters ↓
Albert
Hash
50.31
5.75M
TetraCoder
59.81
3.61M
Summer
Hash
37.94
6.57M
TetraCoder
38.18
4.82M
Pluto
Hash
44.03
23.96M
TetraCoder
48.63
17.27M
Table 1. R1 – Image fitting comparison with Hash Encoding (Instant-NGP). Our method consistently achieves higher PSNR with fewer parameters.
Metric
Method
Mean
CD ↓
TetraSDF L ( THash )
1.718
Ours ( TetraCoder )
1.611
Extract (s) ↓
TetraSDF L ( THash )
2.859
Ours ( TetraCoder )
0.446
Params. (M) ↓
TetraSDF L ( THash )
16.778
Ours ( TetraCoder )
5.221
Table 2. R2 – Surface reconstruction on the Stanford dataset. We compare reproduced TetraSDF (large) with our method using identical geometry supervision and 5,000 training updates. We report Chamfer Distance (CD, ×10−3 ), analytic extraction time in seconds, allocated parameters in millions, intrinsic Delaunay edge compliance (%), and skinny triangles (%).
NeRF Synthetic
DTU
Mean
Method
PSNR ↑
SSIM ↑
CD ↓
PSNR ↑
SSIM ↑
CD ↓
PSNR ↑
SSIM ↑
CD ↓
Tetra-NeRF ( Kulhanek and Sattler, 2023 ) ( Mesh )
29.71
0.948
14.24
27.09
0.905
2.65
28.04
0.927
8.35
NeuMesh++ ( Bao et al., 2025 ) ( Mesh )
32.32
0.955
–
28.39
0.940
–
30.36
0.9475
–
NeuS2 ( Wang et al., 2023 ) ( Hash )
31.58
0.953
4.10
32.14
0.938
0.711
31.86
0.9455
2.4055
Instant-NGP ( Müller et al., 2022 ) ( Hash )
32.75
0.960
4.71
32.27
0.942
1.77
32.51
0.9510
3.2400
Ours ( TetraCoder )
33.27
0.969
3.93
32.34
0.943
0.705
32.81
0.9560
2.3725
Table 3. R3 – Novel-view synthesis appearance and reconstruction accuracy on NeRF Synthetic and DTU. Mean reports the arithmetic average across the two datasets. CD is reported in millimeters.
Figure 3. R3 – Analytic extraction and grid-based extractors on the same frozen Lego field. SSDF measures mesh–field agreement.
Editing %
2D MSE ↓ ( ×10−2 )
3D MSE ↓
TetraCoder
Hash
TetraCoder
Hash
20
0.048
0.536
0.000
0.055
40
0.040
0.655
0.001
0.139
60
0.043
0.942
0.001
0.091
80
0.160
2.221
0.003
0.155
Table 4. R4 – Before–after editing MSE on the unpainted region for 2D and 3D scenes at different amounts of editing.
Figure 4. Qualitative results on T1 : Toy 2D image fitting, T2 : 3D point cloud to mesh, T3 : 3D scene reconstruction from images, T4 : 3D scene editing. TetraCoder can adapt to different decoders, tackling multiple tasks with a single unified representation. Four rows show 2D image fitting, mesh reconstruction from point clouds, novel-view synthesis, and 3D scene editing, including a physics simulation.
D
P
M
CD ↓
PSNR ↑
SSIM ↑
\ding 55
\ding 55
\ding 51
7.185
31.5678
0.959210
\ding 55
\ding 51
\ding 55
7.929
31.2010
0.955343
\ding 55
\ding 51
\ding 51
9.036
31.5586
0.959147
\ding 51
\ding 55
\ding 51
6.534
33.1690
0.968326
\ding 51
\ding 51
\ding 55
7.094
33.2307
0.968409
\ding 51
\ding 51
\ding 51
3.926
33.2647
0.968904
Table 5. Site-update ablation on NeRF Synthetic. D, P, and M denote densification, pruning, and site movement, respectively; ticks and crosses indicate enabled and disabled operations. We report mean CD ( ×10−3 , ↓ ), PSNR (dB, ↑ ), and SSIM ( ↑ ).
Extracting an explicit surface that exactly matches the zero-level set of a neural signed distance function (SDF) remains challenging. Sampling-based isosurfacing methods such as Marching Cubes introduce discretization error. In contrast, continuous piecewise affine (CPWA) analytic approaches typically require plain ReLU MLPs, which limits the ability to learn high-frequency SDFs in practice. We present TetraSDF, an analytic isosurface extraction framework for SDFs that retains the expressiveness of grid-based encoders while enabling exact zero-level set extraction, by representing the SDF with a ReLU MLP composed with a multi-resolution tetrahedral positional encoder. Our positional encoder's barycentric interpolation preserves a global CPWA structure, allowing us to track ReLU linear regions within an encoder-induced polyhedral complex. We further introduce a fixed analytic input preconditioner derived from the encoder's metric to reduce directional bias, thereby stabilizing training. Across multiple benchmarks, TetraSDF matches or surpasses existing grid-based encoders in SDF reconstruction accuracy, while faithfully recovering the network's zero-level set as a triangle mesh.
Seonghun Oh, Youngjung Uh, Jin-Hwa Kim
Yonsei University, Republic of Korea · NAVER AI Lab, Republic of Korea · Seoul National University, Republic of Korea
Meshes remain the most practical representation for geometry reasoning and integration into graphics pipelines, yet existing reconstruction methods struggle to produce high-quality meshes. Most state-of-the-art approaches initially learn an intermediate representation (NeRF/3DGS) and treat mesh extraction as a post-processing step, which often leads to oversmoothed surfaces or poor quality meshes with excessive triangle counts.Existing mesh-native optimization methods alleviate some of these issues but suffer from fixed-resolution discretizations and unstable optimization behavior. In this paper, we introduce an adaptive mesh-based optimization framework and a practical mesh rendering technique to address these challenges. Our representation combines an optimizable Delaunay-triangulated tetrahedral grid with a multi-resolution hash grid. The former is refined through point pruning and insertion, while the latter provides latent features for SDF/appearance value predictions. We use volumetric rendering to bootstrap a coarse geometry while leveraging mesh-based rendering for recovering fine-grained details. Additionally, we propose a differentiable, rasterization-based depth-offset rendering formulation, reducing geometric artifacts and improving reconstruction quality. Our method significantly outperforms existing mesh optimization approaches across a variety of object-centric benchmarks while being competitive with state-of-the-art NeRF/3DGS methods.
Johannes Weidenfeller, Shaofei Wang, Philipp Fürnstahl +1
ETH Zurich, Zurich, Switzerland · Beijing Institute for General Artificial Intelligence, Beijing, China · University of Zurich, Zurich, Switzerland +1
We propose Squeeze3D, a novel framework that leverages implicit prior knowledge learnt by existing pre-trained encoders and decoders to compress 3D data at extremely high compression ratios. Our approach bridges the latent spaces between a pre-trained encoder and a pretrained decoder model through trainable mapping networks. Any 3D asset represented as a mesh, point cloud, or radiance field is first encoded by the pre-trained encoder and then transformed (i.e. compressed) into a highly compact latent code by a mapping network. This latent code can effectively be used as an extremely compressed representation of the mesh, point cloud, or radiance field. A mapping network transforms the compressed latent code into the latent space of a powerful generative model; the decoder of this generative model then recreates the original 3D asset (i.e. decompression). Squeeze3D is trained entirely on generated synthetic data and does not require any 3D datasets. The Squeeze3D architecture can be flexibly used with existing pre-trained 3D encoders and existing generative models. It can flexibly support different formats, including meshes, point clouds, and radiance fields. Our experiments demonstrate that Squeeze3D achieves compression ratios of up to 2187× for textured meshes, 58.5× for point clouds, and more than 650× for radiance fields while maintaining visual quality comparable to many existing methods. Squeeze3D only incurs a small compression and decompression latency since it does not involve training object-specific networks to compress an object.