# When Should a Network Emit Geometry, and When Should It Detect It? Readout, Reconciliation, and Representation in Floorplan Vectorization

arXiv: https://arxiv.org/abs/2608.25608 · HTML: https://mlsanity.com/paper/2608.25608/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 38 references that are in the map.
- First posted on arXiv 26 Aug 2026.

## Builds on (most cited first, 28 of 28)

- [Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World](https://mlsanity.com/paper/1703.06907/) (2017)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [Show and Tell: A Neural Image Caption Generator](https://mlsanity.com/paper/1411.4555/) (2014)
- [Pix2seq: A Language Modeling Framework for Object Detection](https://mlsanity.com/paper/2109.10852/) (2021)
- [Structured3D: A Large Photo-realistic Dataset for Structured 3D Modeling](https://mlsanity.com/paper/1908.00222/) (2019)
- [PolyGen: An Autoregressive Generative Model of 3D Meshes](https://mlsanity.com/paper/2002.10880/) (2020)
- [ATISS: Autoregressive Transformers for Indoor Scene Synthesis](https://mlsanity.com/paper/2110.03675/) (2021)
- [A Unified Sequence Interface for Vision Tasks](https://mlsanity.com/paper/2206.07669/) (2022)
- [PolyFormer: Referring Image Segmentation as Sequential Polygon Generation](https://mlsanity.com/paper/2302.07387/) (2023)
- [Efficient Interactive Annotation of Segmentation Datasets with Polygon-RNN++](https://mlsanity.com/paper/1803.09693/) (2018)
- [SceneScript: Reconstructing Scenes With An Autoregressive Structured Language Model](https://mlsanity.com/paper/2403.13064/) (2024)
- [HouseDiffusion: Vector Floorplan Generation via a Diffusion Model with Discrete and Continuous Denoising](https://mlsanity.com/paper/2211.13287/) (2022)
- [CubiCasa5K: A Dataset and an Improved Multi-Task Model for Floorplan Image Analysis](https://mlsanity.com/paper/1904.01920/) (2019)
- [Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries](https://mlsanity.com/paper/2211.15658/) (2022)
- [PolyDiffuse: Polygonal Shape Reconstruction via Guided Set Diffusion Models](https://mlsanity.com/paper/2306.01461/) (2023)
- [Floor-SP: Inverse CAD for Floorplans by Sequential Room-wise Shortest Path](https://mlsanity.com/paper/1908.06702/) (2019)
- [FloorNet: A Unified Framework for Floorplan Reconstruction from 3D Scans](https://mlsanity.com/paper/1804.00090/) (2018)
- [HEAT: Holistic Edge Attention Transformer for Structured Reconstruction](https://mlsanity.com/paper/2111.15143/) (2021)
- [Deep Floor Plan Recognition Using a Multi-Task Network with Room-Boundary-Guided Attention](https://mlsanity.com/paper/1908.11025/) (2019)
- [MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans](https://mlsanity.com/paper/2103.11161/) (2021)
- [PolyRoom: Room-aware Transformer for Floorplan Reconstruction](https://mlsanity.com/paper/2407.10439/) (2024)
- [ResPlan: A Large-Scale Vector-Graph Dataset of 17,000 Residential Floor Plans](https://mlsanity.com/paper/2508.14006/) (2025)
- [FRI-Net: Floorplan Reconstruction via Room-wise Implicit Representation](https://mlsanity.com/paper/2407.10687/) (2024)
- [WAFFLE: Multimodal Floorplan Understanding in the Wild](https://mlsanity.com/paper/2412.00955/) (2024)
- [FloorplanVLM: A Vision-Language Model for Floorplan Vectorization](https://mlsanity.com/paper/2602.06507/) (2026)
- [CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction](https://mlsanity.com/paper/2509.15459/) (2025)
- [LLM-Guided Agentic Floor Plan Parsing for Accessible Indoor Navigation of Blind and Low-Vision People](https://mlsanity.com/paper/2604.23970/) (2026)
- [Raster2Seq: Polygon Sequence Generation for Floorplan Reconstruction](https://mlsanity.com/paper/2602.09016/) (2026)

## Abstract (arXiv)

A network trained to recover the walls, openings, and rooms of a rasterized floorplan can produce its output in two ways: by emitting the geometry as an autoregressive coordinate sequence, or by detecting it on dense junction and centerline heatmaps and assembling a graph. We compare the two readouts on the same trained network. On real scans (CubiCasa5K) detection is better on every wall measure (+2.7 wall F1 at tolerance 0.05, +5.1 at 0.015; paired bootstrap intervals exclude zero), and reading an opening heatmap the decoder never used raises opening F1 by 2.6x without retraining. Within real scans the readout's advantage grows with plan size and reverses on small plans; on clean vector renders sequence decoding is better by 5 to 8 points where its training covered the render style, while under full domain shift the readout, given calibrated thresholds, stays ahead; neither ink density nor plan size explains the reversal. With matched data and recipe, a room-centric system with a reconciliation step and a wall-first sequence model reach comparable wall quality, so the output representation matters less than is usually assumed. A prior from the other family helps at the output but not at the input: deterministic fusion of the two outputs raises wall F1 by 7 points, whereas conditioning one model on the other's output gives no gain in three forms, including two ground-truth-content controls. We also provide an edit-cost metric that scores a draft by the human work needed to correct it, corrected CubiCasa5K annotations, and ResPlan-FP, a CC BY 4.0 benchmark of 16,998 plans with frozen splits and three baseline tracks. Code, the benchmark, and the corrected annotations are available at https://github.com/Cyprinus12138/fpvec-lab