English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

BrickFlow: Connectivity-Guided Brick Reconstruction from LEGO Part Sets and Images

Forum topic · 小凯 · 2026-11-13

Summary

BrickFlow, a paper by Peter Kulits and Cordelia Schmid (arXiv:2611.17321), addresses the task of inferring the pose of each LEGO part to assemble an object given only a set of parts and a photo of the assembled result. Unlike traditional 3D reconstruction, the problem involves discrete physical constraints that make it combinatorial and highly sensitive to small pose errors. The authors train a set-conditioned SE(3) flow model on procedurally generated assemblies to learn a prior over how parts fit together, then finetune it to inject image conditioning using only 1% as much data. Because the learned flow captures global structure without guaranteeing connectivity, they introduce an analytic connectivity-field guidance that pulls predictions toward connected assemblies at test time. Experiments show that model scale, pretraining, and connectivity guidance jointly improve construction quality. The model and data are publicly available for research.

Overview

  • Field: Computer Vision
  • Authors: Peter Kulits, Cordelia Schmid
  • Published: 2026-11-12
  • arXiv: 2611.17321
  • Abstract

    Given a set of LEGO parts and an image of an assembled object, the task is to infer the pose of each part to assemble the object. In contrast with traditional 3D reconstruction, doing so requires satisfying discrete physical constraints, which provide structure but render the problem combinatorial and highly sensitive to small pose errors.

    To learn a prior over how parts fit together, the authors train a set-conditioned SE(3) flow model on procedurally generated assemblies. They then finetune this model to inject image conditioning, using 1% as much data.

    While the learned flow captures global structure, it does not enforce connectivity. To address this, the paper introduces an analytic connectivity-field guidance that pulls predictions toward connected assemblies at test time.

    Key points

  • Task: assemble a set of LEGO parts to match a single reference photo — a discrete, combinatorial alternative to continuous 3D reconstruction.
  • Method: a set-conditioned SE(3) flow model pretrained on procedurally generated assemblies.
  • Data-efficient finetuning: image conditioning is added with only 1% of the data.
  • Novelty: analytic connectivity-field guidance at test time ensures predictions form connected assemblies.
  • Findings: model scale, pretraining, and connectivity guidance jointly improve construction quality.

Resources

Model and data are openly available at: https://kulits.github.io/BrickFlow/

--- *Auto-collected on 2026-11-14.*

Tags

#computer-vision#3d-reconstruction#lego-assembly#flow-models#se3#diffusion-guidance#paper#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178637112