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

He Zhang

cs.CVAug 26, 2026

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

Authors: He Zhang

Organizations: Independent Researcher

Abstract

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

Figures & tables

Figure 1: The wall-first network and its two readouts. One encoder feeds a grammar-masked sequence decoder (emit) and a dense branch of three heatmaps; the graph readout (detect) assembles walls from the junction and centerline heatmaps without a trained parameter. The room-centric system is a separate network (Raster2Seq) followed by a reconciliation step.

Input:

H_{J},H_{C}\in[0,1]^{S\times S}

, sub-pixel offsets

O

; thresholds

\theta_{J},\theta_{C}

, coverage

\kappa

, chain angle

\alpha

1. Nodes.

V\leftarrow

local maxima of

H_{J}

with

H_{J}\geq\theta_{J}

(non-maximum suppression), refined by

O

2. Candidate edges. For every pair

(u,v)\in V^{2}

: sample the chord

\overline{uv}

n

interior points (trimming the ends); accept if the fraction of samples with

H_{C}\geq\theta_{C}

is at least

\kappa

\|u-v\|\geq\ell_{\min}

, and no third node lies within

\epsilon

of the chord (composite chords are rejected; their sub-edges carry them).

3. Chain. At each node, continue an edge into its single best anti-parallel partner within angle

\alpha

; walk chains in both directions to obtain maximal near-straight walls. Disabled in every frozen configuration: the ground-truth walls of all three domains are split at junctions, and chaining costs

22

pp on CubiCasa5K validation ( Appendix E ).

Output: walls as chained segments with shared node instances, hence closed by construction; rooms as faces of the planar graph.

Table 1: Graph readout from dense heatmaps to a wall graph.

H_{J}

and

H_{C}

are the network’s junction and centerline heatmaps; no parameter is trained.

readout	[email protected]	[email protected]	P	R	$r_{\mathrm{val}}$	rooms	open
graph readout (detect)	0.818	0.787	0.885	0.775	0.956	0.591	—
sequence decode (emit)	0.790	0.736	0.802	0.787	0.196	0.512	0.391
r2s_mix (rooms+reconcil.)	0.781	0.750	0.906	0.709	0.916	0.784	0.848

Table 2: Same trained network (walls-only arm) with the wall readout swapped, CubiCasa5K official test (

n{=}296

). Detection is ahead of autoregressive decoding on every wall measure. Rooms are the faces of the wall graph; the readout emits no openings. The room-centric system is shown for reference (

n{=}287

). Per-system 95% bootstrap half-widths on wall F1 are

0.012

–

0.018

Figure 2: Qualitative results. Columns: input; sequence decode (walls-only arm); graph readout of the same network; room-centric system after reconciliation; fusion (base in purple, ink-gated donor walls in red); ground truth. Rows 1–4 are CubiCasa5K official-test plans (a typical plan, a plan where the readout is far ahead of decoding, a plan where fusion repairs a weak room-centric output, and a failure case on which every system is poor); rows 5–6 are ResPlan-FP zero-shot plans, one on which the room-centric arm emits no rooms and the fused system falls back to the readout walls. The ResPlan readout panels use the frozen CubiCasa5K thresholds (the zero-shot protocol of Table 10 ), not the in-domain-calibrated variant. Wall F1 at $t=0.05$ in each title.

domain	$n$	dark-pixel frac.	GT walls	seq. decode	graph readout	readout $-$ decode
CubiCasa5K (scans)	296	0.078	28.9	0.790	0.818	$+2.7$ $[1.4,4.0]$
open-plan (clean synth)	30	0.112	35.0	0.908	0.825	$-8.2$ $[-12.3,-4.7]$
ResPlan, zero-shot (frozen thr.)	1000	0.094	31.0	0.640	0.621	$-1.9$ $[-2.7,-1.2]$
ResPlan, zero-shot (calibrated thr.)	1000	0.094	31.0	0.640	0.677	$+3.7$ $[+3.0,+4.4]$
ResPlan, finetuned	1000	0.094	31.0	0.968	0.915	$-5.3$ $[-5.5,-5.0]$

Table 3: Emit versus detect across domains (wall [email protected]) with the mean dark-pixel fraction of the

256

px input and the paired bootstrap 95% interval of readout minus decode. The walls-only arm is used everywhere except the finetuned ResPlan row. Readout thresholds: CubiCasa5K frozen; open-plan calibrated on an in-domain set (

0.819

with the frozen CubiCasa5K thresholds); finetuned ResPlan calibrated on ResPlan validation (

0.914

with the frozen CubiCasa5K thresholds); the two zero-shot rows show the frozen thresholds carried over and the effect of calibrating only the thresholds on ResPlan validation (no training). The winner is not monotone in ink density or in plan size.

system	wall F1	wall P	wall R	rooms	open	edit
r2s_cc5k (rooms+reconcil.)	0.769	0.894	0.703	0.777	0.840	80.7
r2s_mix (rooms+reconcil.)	0.781	0.906	0.709	0.784	0.848	79.3
wf_cc5k (full seq)	0.747	0.762	0.742	0.374	0.341	117.1
wf_mix (full seq)	0.751	0.764	0.746	0.395	0.367	113.9

Table 4: Matched-data comparison, CubiCasa5K official test, fpeval @0.05. Room-centric rows are scored on the

290

(r2s_cc5k) and

287

(r2s_mix) plans on which the model produces a decodable output, wall-first rows on all

296

; on the common

287

-plan roster wf_mix has wall F1

0.751

and edit cost

110.2

( Table 11 ). “rooms” for wall-first is derived from faces; room-centric rooms are the Raster2Seq stage-1 output scored under fpeval against skv4, not under the official protocol (which gives

0.887

, Section 5 ). Per-system 95% bootstrap half-widths on wall F1 are

0.013

–

0.017

. Wall-first has higher wall recall; the room-centric system is better on wall F1, rooms, openings, watertightness, and edit cost.

arm	room-boundary walls ( $n{=}3337$ )	non-boundary walls ( $n{=}835$ )
r2s (rooms+reconcil.)	0.865	0.302
wall-first native	0.827	0.663

Table 5: Wall recall by room-boundary membership (150-plan subset of the CubiCasa5K official test, tol 0.05). The representational effect is a coverage limit rather than a topology-quality effect.

Figure 3: Wall classes. Room-bounding walls can be recovered from room polygons; non-room-bounding walls (dangling, stub, exterior) cannot.

system	wall F1	wall P	rooms	doubles/plan	$r_{\mathrm{val}}$
r2s_cc5k + reconcil.	0.769	0.894	0.777	0.00	0.938
r2s_cc5k NAIVE	0.708	0.758	0.778	7.90	0.983
wf_cc5k snap .06	0.747	0.762	0.374	$\sim$ 0	0.02
wf_cc5k SNAP0	0.778	0.794	0.087	0.70	0.01

Table 6: Post-processing ablation, CubiCasa5K test, fpeval @0.05. Without the reconciliation step or snapping, each system is better on its own native element; the aggregate ordering is set by the asymmetry of derivation direction.

system	donor	[email protected]	[email protected]	edit
r2s_mix alone	—	0.781	0.750	79.3
wf_mix alone (full sequence)	—	0.751	0.694	110.2
walls-only arm alone (seq. decode)	—	0.790	0.736	—
fusion, ungated	wf_mix	0.832	0.787	—
fusion + ink gate ( $\geq 0.5$ )	wf_mix	0.846	0.806	80.1
fusion final (ink gate, opening-strip)	walls-only	0.853	0.811	73.1

Table 7: Fusion steps, CubiCasa5K official test (

n{=}287

of 296), wall F1, with the donor stated per row. Each step requires no training and is deterministic. Opening-strip changes openings and edit cost only, not wall F1. The final system is the best overall (paired 95% interval of its gain over r2s_mix:

[5.9,8.3]

pp at

0.05

[5.1,7.2]

0.015

domain	donor	base $F_{1}$	$F_{1}/2$	$k$	$p$	fused $F_{1}$	$\Delta$
CubiCasa5K	wf_mix seq, no gate	0.781	0.391	7.4	0.51	0.832	$+5.1$
CubiCasa5K	wf_mix seq, ink gate	0.781	0.391	4.8	0.67	0.846	$+6.5$
CubiCasa5K	walls-only seq, no gate	0.781	0.391	7.1	0.56	0.844	$+6.3$
CubiCasa5K	walls-only seq, ink gate	0.781	0.391	4.9	0.69	0.853	$+7.1$
CubiCasa5K	graph readout, no gate	0.781	0.391	4.8	0.75	0.856	$+7.4$
CubiCasa5K	graph readout, ink gate	0.781	0.391	4.5	0.74	0.850	$+6.9$

Table 8: Measured donor purity

p

against the threshold

F_{1}/2

of Eq. 4 , wall [email protected].

k

is the mean number of ink-gated donor walls added per plan;

p

is the pooled fraction of those that match a ground-truth wall the base had missed. CubiCasa5K rows are on the

287

-plan roster; ResPlan is scored on the

733

zero-shot plans where the room-centric base decodes.

arm	deploy	zero-content	GT-content
no-conditioning (walls-only)	0.790 / 0.736
room-outline channel	0.762 / 0.708	0.762 / 0.710	0.764 / 0.709
fused-draft channel	0.768 / 0.720	0.792 / 0.742	0.795 / 0.745
symbolic coordinate prefix	0.756 / 0.696	0.783 / 0.717	0.781 / 0.715

Table 9: Input-side conditioning, CubiCasa5K official test, wall F1 (@.05 / @.015). The three columns for each arm are deployment / zero-content / ground-truth-content. No deployment arm beats the unconditioned wall-only model (0.790 / 0.736); the zero- and GT-content control rows reproduce that baseline within noise, and GT content adds at most

0.3

pt over zero-content. The paired noise scale on this test set is about

\pm 1.3

pt ( Section 6.1 ), so the

0.3

pt control differences are indistinguishable from zero while the

2

–

3

pt deploy-arm deficits are not.

Figure 4: The fused-draft-channel conditioning arm. The rasterized draft enters the wall-first network as an extra input channel; the room-outline arm replaces the draft raster with rasterized room outlines, and the symbolic-prefix arm instead feeds the draft’s coordinates to the decoder as a prefix (not drawn). What any of them can carry is a second reading of the same pixels, and none has an effect.

track	score	note
zero-shot, fusion system	0.688	readout fallback on failures
zero-shot, wall-first seq-decode alone	0.640	no failures
zero-shot, wall-first readout alone	0.621	frozen thr.; 0.677 calibrated
zero-shot, room-centric alone	0.383	26.7% hard failures
finetuned, room-centric	0.928	rooms/open/wt $0.986/0.989/0.986$
finetuned, wall-first seq-decode	0.968	—

Table 10: ResPlan-FP baseline tracks (wall F1 unless noted). The zero-shot fusion row keeps the strictly zero-shot protocol (frozen-threshold readout donor); fusion with the threshold-calibrated donor is an expected further gain and is not listed as a track.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

operation	r2s_mix	wf_mix	fusion
wall move	21.7	23.3	25.3
wall type change	8.5	2.5	10.3
wall create	7.5	6.0	4.0
wall delete	1.8	5.1	3.1
open-to-phantom convert	0.0	0.0	0.0
opening move	13.6	10.3	13.6

Appendix

Table 11: Mean edit operations per plan, CubiCasa5K official test (

n{=}287

), default weights. Rows are operation counts; the total is the weighted cost of Eq. 1 , in which a move is weighted by its displacement, so the total is not a weighted sum of the rows. The wall-first total here (

110.2

) differs from Table 4 (

113.9

n{=}296

) because of the roster; the room-centric total is unchanged because Table 4 already scores it on these

287

plans.

bed/bath	share	parameter	value
0/1 (studio)	0.5%	area band, 1 bed	30–80
1/1	15.2%	area band, 2 bed	50–130
2/2	28.4%	area band, 3 bed	75–200
3/2	36.0%	area band, 4 bed	110–350
4/3	16.8%	balcony	0.70
5/4	2.8%	service yard	0.60 ( $\geq$ 2 bed), 0.25 otherwise

Appendix

Table 12: Unit statistics used by the generator. Left: bedroom/bathroom joint distribution fitted to 18,203 parsed listings. Right: area bands (

\mathrm{m}^{2}

) per bedroom count and prevalence of optional spaces.

Indoor Scene Reconstruction

Indoor scene reconstruction builds 3D models of rooms and homes from images, video, depth, or sensor scans for mapping, simulation, and robotics.

See all 40 papers

PlanCraft: Sketch, Refine, and Furnish for Architect-Inspired Progressive 3D Residential Scene Generation
Jul 26, 2026Pengyu Zeng, Yuqin Dai, Jun Yin +73D Scene Generation 3D Scene Layout Generation
Tsinghua University
HypergraphFormer: Learning Hypergraphs from LLMs for Editable Floor Plan Generation
May 18, 2026Nikita Klimenko, Hesam Salehipour, Parham Eftekhar +2Floor Plan Generation Indoor Scene Generation
Autodesk Research · York University · UC Berkeley
PolyLayout: Multi-room Manhattan Layout Estimation
Aug 4, 2026Gustav Hanning, Shaohui Liu, Rémi Pautrat +33D Scene Understanding
Lund University · ETH Zurich · Microsoft Spatial AI Lab

Jul 26, 2026J/K move · Enter open · S save

PlanCraft: Sketch, Refine, and Furnish for Architect-Inspired Progressive 3D Residential Scene Generation

Pengyu Zeng, Yuqin Dai, Jun Yin, Ziyang Han+6

Tsinghua University

3D Scene Generation 3D Scene Layout Generation Floor Plan Generation Indoor Scene Generation

Open paper AskPDF arXiv

Two structural insights have been overlooked in automated residential floor plan generation. First, design is inherently progressive. Architects begin with rough strokes and refine them over time, whereas existing methods typically require their conditioning representation to be fully specified before generation, a fundamental mismatch with how design actually works. Second, the 2D floor plan is not an optional intermediate but an irreplaceable spatial contract. Once room boundaries, doors, and windows are fixed, furnishing reduces from open-ended spatial reasoning to bounded constraint satisfaction. Bypassing this contract, as existing 3D systems do by delegating layout to language models, yields overlapping rooms and implausible proportions; directly calling general-purpose language models likewise produces geometrically invalid layouts. Guided by these insights, we present PlanCraft. SketchPlan supplies the missing training signal by replaying the architect's drawing process on 80K real floor plans, producing partial sketches at every completeness level. PlanCraft-Diff progressively sharpens an incomplete sketch into a geometrically precise, vectorizable floor plan through a coarse-to-fine strategy. With the spatial contract established, PlanCraft-Agent then furnishes the scene within well-defined room boundaries. Experiments show that PlanCraft achieves a 61.1% lower FID than the best existing 2D method and surpasses existing 3D systems by 15 points in expert-rated spatial rationality, with a sketch at only 25% completion already outperforming all fully specified baselines.

Abstract

Indoor Scene Reconstruction

Explore similar work

PlanCraft: Sketch, Refine, and Furnish for Architect-Inspired Progressive 3D Residential Scene Generation

HypergraphFormer: Learning Hypergraphs from LLMs for Editable Floor Plan Generation

PolyLayout: Multi-room Manhattan Layout Estimation