# SCORES: Shape Composition with Recursive Substructure Priors

arXiv: https://arxiv.org/abs/1809.05398 · HTML: https://mlsanity.com/paper/1809.05398/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Sep 2018.

## Built on by (most cited first, 24 of 24)

- [Flexible Isosurface Extraction for Gradient-Based Mesh Optimization](https://mlsanity.com/paper/2308.05371/) (2023)
- [VAEL: Bridging Variational Autoencoders and Probabilistic Logic Programming](https://mlsanity.com/paper/2202.04178/) (2022)
- [Navigating the Maze of Explainable AI: A Systematic Approach to Evaluating Methods and Metrics](https://mlsanity.com/paper/2409.16756/) (2024)
- [Patch-based 3D Natural Scene Generation from a Single Example](https://mlsanity.com/paper/2304.12670/) (2023)
- [Towards Target-Driven Visual Navigation in Indoor Scenes via Generative Imitation Learning](https://mlsanity.com/paper/2009.14509/) (2020)
- [Learning Fine-Grained Segmentation of 3D Shapes without Part Labels](https://mlsanity.com/paper/2103.13030/) (2021)
- [AutoMate: A Dataset and Learning Approach for Automatic Mating of CAD Assemblies](https://mlsanity.com/paper/2105.12238/) (2021)
- [Points2NeRF: Generating Neural Radiance Fields from 3D point cloud](https://mlsanity.com/paper/2206.01290/) (2022)
- [Semi-supervised 3D shape segmentation with multilevel consistency and part substitution](https://mlsanity.com/paper/2204.08824/) (2022)
- [Multi-resolution Monocular Depth Map Fusion by Self-supervised Gradient-based Composition](https://mlsanity.com/paper/2212.01538/) (2022)
- [SoccerChat: Integrating Multimodal Data for Enhanced Soccer Game Understanding](https://mlsanity.com/paper/2505.16630/) (2025)
- [Open-Jev Judgments on CallScreenBench: Calibrated One-Pass Scam Screening with a Small Language Model](https://mlsanity.com/paper/2609.23959/) (2026)
- [Learning Mesh Representations via Binary Space Partitioning Tree Networks](https://mlsanity.com/paper/2106.14274/) (2021)
- [HyperCube: Implicit Field Representations of Voxelized 3D Models](https://mlsanity.com/paper/2110.05770/) (2021)
- [Alloc-MoE: Budget-Aware Expert Activation Allocation for Efficient Mixture-of-Experts Inference](https://mlsanity.com/paper/2604.08133/) (2026)
- [Attention-based Part Assembly for 3D Volumetric Shape Modeling](https://mlsanity.com/paper/2304.10986/) (2023)
- [STEdge: Self-training Edge Detection with Multi-layer Teaching and Regularization](https://mlsanity.com/paper/2201.05121/) (2022)
- [Seg&Struct: The Interplay Between Part Segmentation and Structure Inference for 3D Shape Parsing](https://mlsanity.com/paper/2211.00382/) (2022)
- [An Agentic AI Scientific Community for Automated Neural Operator Discovery](https://mlsanity.com/paper/2607.12122/) (2026)
- [Learning Semantic Abstraction of Shape via 3D Region of Interest](https://mlsanity.com/paper/2201.04945/) (2022)
- [Expect the Unexpected? Testing the Surprisal of Salient Entities](https://mlsanity.com/paper/2604.10724/) (2026)
- [Flow Matching for Convective-Scale Precipitation Downscaling](https://mlsanity.com/paper/2606.00281/) (2026)
- [Robust Strategic Classification under Decision-Dependent Cost Uncertainty](https://mlsanity.com/paper/2606.30136/) (2026)
- [REEL: Learning to Retrieve, Evaluate, and Evolve Visual Memory for Any Long-Horizon Multi-Shot Video Generation](https://mlsanity.com/paper/2610.16632/) (2026)

## Abstract (arXiv)

We introduce SCORES, a recursive neural network for shape composition. Our network takes as input sets of parts from two or more source 3D shapes and a rough initial placement of the parts. It outputs an optimized part structure for the composed shape, leading to high-quality geometry construction. A unique feature of our composition network is that it is not merely learning how to connect parts. Our goal is to produce a coherent and plausible 3D shape, despite large incompatibilities among the input parts. The network may significantly alter the geometry and structure of the input parts and synthesize a novel shape structure based on the inputs, while adding or removing parts to minimize a structure plausibility loss. We design SCORES as a recursive autoencoder network. During encoding, the input parts are recursively grouped to generate a root code. During synthesis, the root code is decoded, recursively, to produce a new, coherent part assembly. Assembled shape structures may be novel, with little global resemblance to training exemplars, yet have plausible substructures. SCORES therefore learns a hierarchical substructure shape prior based on per-node losses. It is trained on structured shapes from ShapeNet, and is applied iteratively to reduce the plausibility loss.We showresults of shape composition from multiple sources over different categories of man-made shapes and compare with state-of-the-art alternatives, demonstrating that our network can significantly expand the range of composable shapes for assembly-based modeling.