# On the Effectiveness of Weight-Encoded Neural Implicit 3D Shapes

arXiv: https://arxiv.org/abs/2009.09808 · HTML: https://mlsanity.com/paper/2009.09808/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Sep 2020.

## Built on by (most cited first, 29 of 29)

- [Efficient Geometry-aware 3D Generative Adversarial Networks](https://mlsanity.com/paper/2112.07945/) (2021)
- [Deep Marching Tetrahedra: a Hybrid Representation for High-Resolution 3D Shape Synthesis](https://mlsanity.com/paper/2111.04276/) (2021)
- [Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes](https://mlsanity.com/paper/2101.10994/) (2021)
- [Advances in Neural Rendering](https://mlsanity.com/paper/2111.05849/) (2021)
- [StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generation](https://mlsanity.com/paper/2112.11427/) (2021)
- [3D Neural Field Generation using Triplane Diffusion](https://mlsanity.com/paper/2211.16677/) (2022)
- [Modulated Periodic Activations for Generalizable Local Functional Representations](https://mlsanity.com/paper/2104.03960/) (2021)
- [Neural Lumigraph Rendering](https://mlsanity.com/paper/2103.11571/) (2021)
- [Generative Neural Articulated Radiance Fields](https://mlsanity.com/paper/2206.14314/) (2022)
- [Compressive Neural Representations of Volumetric Scalar Fields](https://mlsanity.com/paper/2104.04523/) (2021)
- [GenDexGrasp: Generalizable Dexterous Grasping](https://mlsanity.com/paper/2210.00722/) (2022)
- [Learning Deep Implicit Functions for 3D Shapes with Dynamic Code Clouds](https://mlsanity.com/paper/2203.14048/) (2022)
- [D2CSG: Unsupervised Learning of Compact CSG Trees with Dual Complements and Dropouts](https://mlsanity.com/paper/2301.11497/) (2023)
- [Neural Surface Maps](https://mlsanity.com/paper/2103.16942/) (2021)
- [Fast Training of Neural Lumigraph Representations using Meta Learning](https://mlsanity.com/paper/2106.14942/) (2021)
- [NeuForm: Adaptive Overfitting for Neural Shape Editing](https://mlsanity.com/paper/2207.08890/) (2022)
- [Template-free Articulated Neural Point Clouds for Reposable View Synthesis](https://mlsanity.com/paper/2305.19065/) (2023)
- [Neural Geometry Processing via Spherical Neural Surfaces](https://mlsanity.com/paper/2407.07755/) (2024)
- [NURBS-Diff: A Differentiable Programming Module for NURBS](https://mlsanity.com/paper/2104.14547/) (2021)
- [Neural Convolutional Surfaces](https://mlsanity.com/paper/2204.02289/) (2022)
- [ROAD: Learning an Implicit Recursive Octree Auto-Decoder to Efficiently Encode 3D Shapes](https://mlsanity.com/paper/2212.06193/) (2022)
- [Implicit Neural Deformation for Sparse-View Face Reconstruction](https://mlsanity.com/paper/2112.02494/) (2021)
- [Discretization-independent surrogate modeling over complex geometries using hypernetworks and implicit representations](https://mlsanity.com/paper/2109.07018/) (2021)
- [ReFiNe: Recursive Field Networks for Cross-modal Multi-scene Representation](https://mlsanity.com/paper/2406.04309/) (2024)
- [Low-Rank Adaptation of Neural Fields](https://mlsanity.com/paper/2504.15933/) (2025)
- [Theoretical and Empirical Analysis of a Fast Algorithm for Extracting Polygons from Signed Distance Bounds](https://mlsanity.com/paper/2111.05778/) (2021)
- [Analysis of Sampling Strategies for Implicit 3D Reconstruction](https://mlsanity.com/paper/2304.03999/) (2023)
- [CN-DHF: Compact Neural Double Height-Field Representations of 3D Shapes](https://mlsanity.com/paper/2304.13141/) (2023)
- [INRet: A General Framework for Accurate Retrieval of INRs for Shapes](https://mlsanity.com/paper/2501.15722/) (2025)

## Abstract (arXiv)

A neural implicit outputs a number indicating whether the given query point in space is inside, outside, or on a surface. Many prior works have focused on _latent-encoded_ neural implicits, where a latent vector encoding of a specific shape is also fed as input. While affording latent-space interpolation, this comes at the cost of reconstruction accuracy for any _single_ shape. Training a specific network for each 3D shape, a _weight-encoded_ neural implicit may forgo the latent vector and focus reconstruction accuracy on the details of a single shape. While previously considered as an intermediary representation for 3D scanning tasks or as a toy-problem leading up to latent-encoding tasks, weight-encoded neural implicits have not yet been taken seriously as a 3D shape representation. In this paper, we establish that weight-encoded neural implicits meet the criteria of a first-class 3D shape representation. We introduce a suite of technical contributions to improve reconstruction accuracy, convergence, and robustness when learning the signed distance field induced by a polygonal mesh -- the _de facto_ standard representation. Viewed as a lossy compression, our conversion outperforms standard techniques from geometry processing. Compared to previous latent- and weight-encoded neural implicits we demonstrate superior robustness, scalability, and performance.