# Iskra: A System for Inverse Geometry Processing

arXiv: https://arxiv.org/abs/2602.12105 · HTML: https://mlsanity.com/paper/2602.12105/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 84 references that are in the map.
- First posted on arXiv 12 Feb 2026.

## Builds on (most cited first, 18 of 18)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Neural Ordinary Differential Equations](https://mlsanity.com/paper/1806.07366/) (2018)
- [OptNet: Differentiable Optimization as a Layer in Neural Networks](https://mlsanity.com/paper/1703.00443/) (2017)
- [Neural Fields in Visual Computing and Beyond](https://mlsanity.com/paper/2111.11426/) (2021)
- [Differentiable Convex Optimization Layers](https://mlsanity.com/paper/1910.12430/) (2019)
- [ChainQueen: A Real-Time Differentiable Physical Simulator for Soft Robotics](https://mlsanity.com/paper/1810.01054/) (2018)
- [DiffPD: Differentiable Projective Dynamics](https://mlsanity.com/paper/2101.05917/) (2021)
- [Isospectralization, or how to hear shape, style, and correspondence](https://mlsanity.com/paper/1811.11465/) (2018)
- [Cubic Stylization](https://mlsanity.com/paper/1910.02926/) (2019)
- [Differentiable Surface Triangulation](https://mlsanity.com/paper/2109.10695/) (2021)
- [HodgeNet: Learning Spectral Geometry on Triangle Meshes](https://mlsanity.com/paper/2104.12826/) (2021)
- [Differentiable solver for time-dependent deformation problems with contact](https://mlsanity.com/paper/2205.13643/) (2022)
- [BPQP: A Differentiable Convex Optimization Framework for Efficient End-to-End Learning](https://mlsanity.com/paper/2411.19285/) (2024)
- [Anderson Acceleration for Geometry Optimization and Physics Simulation](https://mlsanity.com/paper/1805.05715/) (2018)
- [Sparsity-Specific Code Optimization using Expression Trees](https://mlsanity.com/paper/2110.12865/) (2021)
- [SymX: Energy-based Simulation from Symbolic Expressions](https://mlsanity.com/paper/2303.02156/) (2023)
- [A Convex Optimization Framework for Regularized Geodesic Distances](https://mlsanity.com/paper/2305.13101/) (2023)
- [Differentiable Geodesic Distance for Intrinsic Minimization on Triangle Meshes](https://mlsanity.com/paper/2404.18610/) (2024)

## Abstract (arXiv)

We propose a system for differentiating through solutions to geometry processing problems. Our system differentiates a broad class of geometric algorithms, exploiting existing fast problem-specific schemes common to geometry processing, including local-global and ADMM solvers. It is compatible with machine learning frameworks, opening doors to new classes of inverse geometry processing applications. We marry the scatter-gather approach to mesh processing with tensor-based workflows and rely on the adjoint method applied to user-specified imperative code to generate an efficient backward pass behind the scenes. We demonstrate our approach by differentiating through mean curvature flow, spectral conformal parameterization, geodesic distance computation, and as-rigid-as-possible deformation, examining usability and performance on these applications. Our system allows practitioners to differentiate through existing geometry processing algorithms without needing to reformulate them, resulting in low implementation effort, fast runtimes, and lower memory requirements than differentiable optimization tools not tailored to geometry processing.