Pith. sign in

REVIEW 3 major objections 8 minor 30 references

Mitigating Domain Shift in Conditioned Floor Plan Generation: Synthetic Pre-training for Data-Efficient Adaptation

T0 review · 3 major / 8 minor · reviewed 2026-07-08 · glm-5.2

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Pre-training on ugly floor plans beats training on real ones

desk verdict First cross-domain study for conditioned floor plan generation; synthetic pre-training strategy works but the mechanistic claim is only half-validated. read the letter →

arxiv 2607.06483 v1 pith:DSN6DYRX submitted 2026-07-07 cs.CV

Matthieu Ospici , Arnaud Gueze , Luc Bourrat , Adrien Bernhardt This is my paper · ORCID
classification cs.CV
keywords floorplangenerationdomainshiftsyntheticdataproceduralrandomizationcross-domaintransfergenerativemodelsefficiency
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper addresses domain shift in conditioned floor plan generation, a previously unstudied problem. The authors first demonstrate that state-of-the-art generative models from two fundamentally different paradigms (arrangement-based flow matching and vertex-level diffusion) suffer severe performance degradation, up to an order of magnitude, when transferred across three real-world datasets (RPLAN, MagicPlan, Swiss Dwellings) reflecting distinct architectural cultures. To mitigate this, they introduce a procedural synthetic dataset of 135k floor plans that enforces strict physical constraints (non-overlapping rooms, valid doors, graph consistency) while deliberately abandoning architectural realism through aggressive geometric perturbation. The central claim is that pre-training on these implausible layouts forces models to internalize the combinatorial logic of spatial assembly rather than overfitting to domain-specific visual regularities. The authors show that this synthetic pre-training yields zero-shot cross-domain performance that surpasses in-domain training on MagicPlan for one model, and provides a superior initialization for fine-tuning that outperforms real-world cross-domain initialization by up to 40% in low-data regimes, across both paradigms.

What carries the argument

Procedural synthetic data pipeline with geometric perturbation and constraint enforcement

What would settle it

If the zero-shot and fine-tuning gains disappear when the synthetic pipeline is re-seeded with a fundamentally different shape pool and no augmentations are added, the transferability would be attributable to shape statistics rather than assembly rules.

Watch

Extended reading notes

Core claim

The core discovery is that for conditioned geometric generation tasks, a procedurally generated synthetic dataset that enforces underlying physical constraints while maximizing geometric diversity and sacrificing all architectural realism produces more transferable representations than any single real-world dataset. The mechanism is decoupling: by training on layouts that share no visual resemblance to any target domain but obey the same assembly rules (non-overlapping rooms, valid door placement, graph consistency), models learn the combinatorial logic of spatial assembly rather than dataset-specific shortcuts. This is validated by an ablation showing that removing shape augmentations (redu

Load-bearing premise

The claim that transferability arises from assembly rules rather than RPLAN-specific shape priors rests on a re-seeding ablation where one of the two models requires an added augmentation to recover transfer, which the authors acknowledge as a limitation.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 8 minor

Summary. This paper presents the first systematic study of domain shift in conditioned floor plan generation, evaluating two generative paradigms (arrangement-based DPFM and vertex-level Gueze et al.) across three real-world datasets (RPLAN, MagicPlan, Swiss Dwellings). The authors propose a procedural synthetic pre-training dataset that enforces physical constraints (non-overlapping rooms, valid doors, graph consistency) while deliberately sacrificing architectural realism through aggressive geometric perturbation. The central empirical claims are: (1) both paradigms suffer severe bidirectional domain shift, (2) synthetic pre-training enables zero-shot transfer that surpasses in-domain training on MagicPlan for DPFM, and (3) synthetic initialization accelerates fine-tuning, outperforming real-world cross-domain initialization by up to 40% in low-data regimes. The experimental design is competent, with ablations on dataset size, shape complexity, and a re-seeding ablation using ProcTHOR shapes to test pool independence.

Significance. The paper addresses a genuine gap: no prior work has formally studied domain shift in conditioned floor plan generation. The cross-domain evaluation across three datasets and two paradigms is well-constructed and the findings are informative for the community. The synthetic data pipeline is a reasonable application of domain randomization principles to the layout generation setting. The zero-shot result (synthetic beating in-domain MagicPlan training for DPFM) is striking and the fine-tuning data-efficiency results are practically useful. The re-seeding ablation (Table 6) is a commendable attempt to falsify the alternative explanation that gains stem from RPLAN shape priors. The paper is transparent about limitations, including the NGED metric failure for Gueze et al. and the DPFM re-seeding residual.

major comments (3)
  1. Sec. 5, Table 6: The mechanistic claim that transferability arises from 'spatial assembly rules rather than RPLAN-specific priors' is only partially supported. For Gueze et al., ProcTHOR re-seeding yields nearly identical results (e.g., 53.8→53.7 on Swiss Dwellings), which is convincing. For DPFM, however, the ProcTHOR-seeded result requires an added small-rotation augmentation (marked †) and still shows meaningful degradation: MagicPlan 23.3→27.6 MPE (the domain where the headline claim of 'synthetic beats in-domain' is made) and Swiss Dwellings 49.4→63.0 MPE (a 28% increase). The paper acknowledges this as 'a limitation of the current ablation.' Since the headline zero-shot results in Table 2 use RPLAN-seeded synthetic data, and the paradigm-general framing ('across two fundamentally different generative paradigms') requires the mechanism to hold for both paradigms, the DPFM gap is a负载
  2. Table 6: The 'small-rotation augmentation (†)' applied to the DPFM ProcTHOR condition is not described in the Method section (Sec. 3.2) or anywhere else in the paper. The augmentation pipeline in Sec. 3.2 specifies rotations in {0°, 90°, 180°, 270°} only. The introduction of an undocumented augmentation specifically to recover DPFM transfer under ProcTHOR seeding raises the question of whether this augmentation was selected post-hoc. The authors should either describe this augmentation in the Method section with justification, or run the ProcTHOR DPFM condition without it and report the un-augmented result for transparency.
  3. Sec. 4.2, Table 3: NGED is omitted for Gueze et al. across all conditions, not just under domain shift. The explanation given is that 'severely displaced rooms can accidentally intersect with decoupled door segments, yielding false connectivities that completely skew the topological evaluation.' This is a reasonable concern, but it means the paper's claim to evaluate with 'geometric and topological metrics (MPE, NGED)' (Sec. 1, bullet 1) is only fully satisfied for DPFM. For Gueze et al., only MPE is reported. The paper should clarify in the contributions list that NGED is reported for one of two paradigms, and consider whether a modified topological metric (e.g., one robust to door-segment displacement) could be applied.
minor comments (8)
  1. Sec. 3.2: The distribution for 'Number of rooms per scene' is described as 'weighted toward medium-density layouts' but the weighting function is not specified. Stating the exact distribution or sampling probabilities would aid reproducibility.
  2. Table 1: The 'Connected components' and 'Isolated rooms' metrics are informative but their definitions are not provided. A brief footnote defining these terms (e.g., whether isolated rooms count as separate connected components) would help interpretation.
  3. Sec. 4.1: The door noise injection ('randomized door lengths') is described briefly. The range or distribution of randomized lengths should be specified, as this protocol modification affects all reported results.
  4. Figure 5a: The y-axis range for the RPLAN target panel (10.0–27.5) differs substantially from the MagicPlan panel (15–30) and Swiss Dwellings panel (30–60). While this is necessary to show detail, a note indicating the different scales would prevent misreading. Consider adding a zoomed inset.
  5. Sec. 4.3: The claim of 'up to 40% in a low-data regime' appears in the abstract. The precise result supporting this is the 41.7% reduction on RPLAN with 1k samples (Table 4: 7.05 vs 12.1). The abstract should cite the exact figure and the correct figure and location.
  6. Reference [7] (DPFM) is cited as 'AAAI 2026 Main Conference, Jan 2026,' which at the time of this review may not yet be publicly available. If the paper is not yet accessible, providing an arXiv preprint link would make the work accessible to readers.
  7. Sec. 3.2: The phrase 'a distribution weighted toward medium-density layouts' is vague. Given that the synthetic dataset has a mean of 4.42 rooms per scene (Table 1), which is lower than all real-world datasets, the term 'medium-density' may be misleading. Consider clarifying that the weighting targets 4–6 rooms per scene.
  8. Table 2: The NGED value for RPLAN under RPLAN training is reported as '0.46±0.01' but appears to be missing the ± notation consistency present in other entries (e.g., '0.90±0.01'). Ensure consistent notation across all entries.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for a careful and constructive reading of our manuscript. The referee correctly identifies the core contributions and fairly assesses the experimental design. We address each major comment below. In brief: (1) we agree the DPFM re-seeding gap weakens the paradigm-general framing of the mechanism claim and will revise the manuscript to scope it more carefully; (2) the undocumented small-rotation augmentation will be fully described in the Method section and the un-augmented ProcTHOR DPFM result will be reported transparently; (3) we will clarify in the contributions list that NGED is reported for DPFM only and discuss the possibility of a door-displacement-robust topological metric as future work.

read point-by-point responses

  1. Referee: Sec. 5, Table 6: The mechanistic claim that transferability arises from 'spatial assembly rules rather than RPLAN-specific priors' is only partially supported. For Gueze et al., ProcTHOR re-seeding yields nearly identical results, which is convincing. For DPFM, the ProcTHOR-seeded result requires an added small-rotation augmentation and still shows meaningful degradation on MagicPlan and Swiss Dwellings. Since the headline zero-shot results use RPLAN-seeded synthetic data, and the paradigm-general framing requires the mechanism to hold for both paradigms, the DPFM gap is a concern.

    Authors: The referee is correct that the DPFM re-seeding result does not fully support the mechanistic claim as strongly as the Gueze et al. result does. We acknowledge this asymmetry. Our intended revision is to scope the mechanistic claim more precisely: the claim that transferability arises from spatial assembly rules rather than RPLAN-specific priors is strongly supported for Gueze et al. (where ProcTHOR re-seeding yields near-identical results across all three target domains), and partially supported for DPFM (where RPLAN and MagicPlan transfer is recovered with a small-rotation augmentation, but a residual gap remains on Swiss Dwellings). We will revise the Discussion and the relevant sentence in the Conclusion to reflect this scoped claim rather than presenting it as uniformly paradigm-general. We will also add an explicit discussion of the DPFM residual: the most likely explanation is that DPFM's rigid-transformation formulation lacks the chain-code angle regularization present in Gueze et al., making it more sensitive to the slight non-orthogonality of RPLAN shapes. The Swiss Dwellings residual likely reflects additional pool properties (e.g., corner-count distributions) beyond global obliquity that differ between RPLAN and ProcTHOR. We agree this is a genuine limitation of the current ablation and will state so plainly. revision: partial

  2. Referee: Table 6: The 'small-rotation augmentation (†)' applied to the DPFM ProcTHOR condition is not described in the Method section (Sec. 3.2) or anywhere else in the paper. The augmentation pipeline in Sec. 3.2 specifies rotations in {0°, 90°, 180°, 270°} only. The introduction of an undocumented augmentation specifically to recover DPFM transfer under ProcTHOR seeding raises the question of whether this augmentation was selected post-hoc. The authors should either describe this augmentation in the Method section with justification, or run the ProcTHOR DPFM condition without it and report the un-augmented result for transparency.

    Authors: The referee is correct that this augmentation is undocumented in the current manuscript, and we will fix this. To be transparent: the small-rotation augmentation was introduced during the re-seeding ablation to compensate for a specific architectural difference between the two models. DPFM predicts rigid transformations of input polygons and has no mechanism enforcing angular regularity; when seeded with ProcTHOR's strictly orthogonal shapes, the model loses exposure to the slight non-orthogonality present in RPLAN (and in all three target domains). Gueze et al. is unaffected because its chain-code representation includes implicit angle regularization. The augmentation consists of applying a small random rotation (uniformly sampled in [-5°, +5°]) to each room polygon independently before assembly, introducing mild non-orthogonality. We will describe this augmentation in the Method section with this justification. Furthermore, we agree that reporting the un-augmented ProcTHOR DPFM result is important for transparency, and we will add it to Table 6. We will also clarify in the text that this augmentation was introduced specifically for the re-seeding ablation and is not part of the main pipeline used for the headline results in Table 2. revision: yes

  3. Referee: Sec. 4.2, Table 3: NGED is omitted for Gueze et al. across all conditions. The paper's claim to evaluate with 'geometric and topological metrics (MPE, NGED)' is only fully satisfied for DPFM. For Gueze et al., only MPE is reported. The paper should clarify in the contributions list that NGED is reported for one of two paradigms, and consider whether a modified topological metric robust to door-segment displacement could be applied.

    Authors: We agree with the referee that the contributions list overstates the metric coverage. The bullet currently reads as if NGED is reported for both paradigms, which is not the case. We will revise the contributions bullet to clarify that NGED is reported for DPFM and that only MPE is reported for Gueze et al., with the explanation given in Section 4.2. Regarding the suggestion of a modified topological metric robust to door-segment displacement: this is a reasonable suggestion and we have considered it. The core difficulty is that Gueze et al. generates door segments decoupled from room boundaries under domain shift, and any connectivity metric that relies on door-segment/room-boundary intersection will be affected. A possible alternative would be to derive connectivity from room-pair adjacency (shared boundary length above a threshold) rather than from door segments specifically. However, this changes the semantics of what is being measured (room adjacency vs. door-mediated connectivity) and would require re-evaluation across all conditions for consistency, including DPFM. Given the scope of this revision, we will discuss this as a promising direction in the Limitations section rather than introducing a new metric that has not been validated across all settings. We believe this is the honest approach: the current NGED metric is well-defined for DPFM but not reliably applicable to Gueze et al. under domain shift, and we should state this limitation clearly rather than substituting an unvalidated alternative. revision: yes

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the synthetic dataset is generated from a procedural pipeline independent of evaluation datasets, and the two evaluated models are treated as fixed baselines trained via their standard procedures.

full rationale

The paper's central claim—that pre-training on a synthetic dataset of physically valid but architecturally implausible floor plans improves cross-domain transfer—is not circularly derived. The synthetic dataset (Sec. 3.2) is generated from a procedural pipeline that extracts isolated room polygons from RPLAN, applies geometric augmentations (aspect ratio distortion, bumps, flips, scale), assembles them via edge-packing, and enforces constraints (non-overlapping rooms, valid doors, graph consistency). The evaluation datasets (RPLAN, MagicPlan, Swiss Dwellings) are distinct real-world datasets not used in synthetic data generation beyond shape extraction. The two evaluated models (DPFM [7] and Gueze et al. [6]) are pre-trained using their standard procedures without modification (Sec. 4.1). Two of the four authors (Ospici, Gueze) are co-authors on both evaluated model papers, but the models are treated as fixed baselines—their architectures and training procedures are not re-derived or re-defined in this paper. The ProcTHOR re-seeding ablation (Table 6) tests whether results depend on RPLAN-specific shape priors; while the DPFM results show some degradation and require an added augmentation (†), this is an empirical limitation of the ablation, not a circularity. The paper acknowledges this limitation transparently. No step in the derivation chain reduces to its inputs by construction, no prediction is a renamed fit, and no self-citation is load-bearing for the central claim. The minor self-citation (authors citing their own model papers as evaluated baselines) is standard practice and does not constitute circularity.

Assumptions & free parameters 9 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new theoretical entities, particles, forces, or dimensions. The synthetic dataset is a procedural construction, not a postulated object. The free parameters are all in the data generation pipeline and training protocol; they are hand-set but not fitted to the evaluation results. The axioms are domain assumptions about transferability and metric adequacy, plus standard math for the metrics.

free parameters (9)
  • Aspect ratio distortion range = [0.8, 1.25]
    Chosen by hand for the shape augmentation pipeline (Sec. 3.2). No sensitivity analysis provided.
  • Number of bumps per polygon = 1-3
    Hand-set range for geometric perturbation (Sec. 3.2). No justification for this range.
  • Global scale factor range = [0.7, 1.6]
    Hand-chosen uniform sampling range for room size variance (Sec. 3.2).
  • Door length fractions = small [0.18,0.35], medium [0.35,0.55], large [0.55,0.85]
    Mixture strategy for door placement (Sec. 3.2). Categories and ranges are manually defined.
  • Door endpoint snap tolerance = 0.03
    Tolerance in normalized [-1,1] coordinates (Sec. 3.2). No sensitivity analysis.
  • Number of rooms per scene = 2-10
    Range for scene assembly, weighted toward medium-density (Sec. 3.2). Weighting distribution not specified.
  • Dataset size = 135k scenes
    Chosen scale; ablation (Table 5) shows 60k vs 135k yields diminishing returns, so this is not critical.
  • Fine-tuning learning rate reduction = factor of 10
    Applied uniformly across all fine-tuning experiments (Sec. 4.1). No LR sweep reported.
  • Fine-tuning steps = 50k
    Fixed across all setups (Sec. 4.1). Authors state no overfitting observed.
assumptions (5)
  • domain assumption Domain randomization transfers to constrained geometric generation tasks
    The paper's central strategy is based on the assumption that the domain randomization principle (established in robotics/vision) applies to floor plan generation. This is empirically validated but not derived from first principles.
  • domain assumption The two evaluated models (DPFM [7] and Gueze et al. [6]) are representative of the two dominant vectorial generative paradigms
    Sec. 1 and Sec. 4.1 claim these two models represent arrangement-based and coordinate-based paradigms. The generalizability of findings depends on this representativeness.
  • domain assumption MPE and NGED are adequate metrics for cross-domain evaluation
    Sec. 4.1 uses MPE (geometric) and NGED (topological) as the sole quality metrics. No human evaluation or alternative metrics are used to validate that these capture layout quality across domains.
  • domain assumption RPLAN room shapes provide a reasonable weak prior for basic polygon properties
    Sec. 3.2 justifies seeding from RPLAN shapes rather than generating from scratch to avoid degenerate polygons. The ProcTHOR re-seeding (Table 6) partially tests this but only for one model without degradation.
  • standard math Standard math: Euclidean distance, graph edit distance, optimal rigid alignment
    Metric definitions in Sec. 4.1 rely on standard mathematical operations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mitigating Domain Shift in Conditioned Floor Plan Generation: Synthetic Pre-training for Data-Efficient Adaptation." pith.science (2026). https://pith.science/paper/DSN6DYRX

@misc{pith2026260706483,
  author       = {Pith},
  title        = {Pith review of: Mitigating Domain Shift in Conditioned Floor Plan Generation: Synthetic Pre-training for Data-Efficient Adaptation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DSN6DYRX}},
  note         = {Machine review of arXiv:2607.06483}
}
read the original abstract

Robustness to domain shift is a key requirement for floor plan generative models to be applicable beyond the single dataset they were trained on, as floor plans vary widely across regions due to distinct architectural cultures, spatial constraints, and construction practices, while acquiring new annotated datasets remains costly and domain-specific. Yet, no prior work has studied this robustness in the context of conditioned floor plan generation. In this paper, we evaluate state-of-the-art models from two fundamentally different generative paradigms across three public datasets (RPLAN, MagicPlan and Swiss Dwellings) and show that they are highly sensitive to domain shift, with up to an order of magnitude performance degradation when transferred across domains. To mitigate this with minimal target-domain supervision, we introduce a procedural method to generate a large-scale synthetic training dataset that enforces strict physical constraints (non-overlapping rooms, valid door placement, graph consistency) while intentionally sacrificing architectural realism through highly irregular spatial arrangements and aggressive geometric perturbation of room shapes. We show that pre-training on this synthetic data considerably improves zero-shot cross-domain performance, outperforming in-domain training on MagicPlan. Furthermore, it provides a highly effective initialization for fine-tuning, accelerating target domain adaptation and outperforming real-world initialization baselines by up to 40% in a low-data regime.

Figures

Figures reproduced from arXiv: 2607.06483 by the authors.

Figure 1
Figure 1. Procedural generation of the synthetic pre-training dataset. Our pipeline deliberately [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Qualitative comparison of the evaluated domains. Real-world datasets (Rplan, Magic [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Qualitative zero-shot transfer (DPFM). Rows: target domains; columns: training sources [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Qualitative zero-shot transfer results (Gueze et al. [6]). Grid layout strictly follows [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Data-efficient fine-tuning and domain adaptation scaling (DPFM). We compare models initialized with different priors across varying amounts of target data. (a) Geometric per￾formance. (b) Topological performance. Across both metrics, our style-agnostic Synthetic prior …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages

  1. [1]

    Lifull dataset.http://https://www.nii.ac.jp/dsc/idr/en/lifull/

    work page

  2. [2]

    Stochastic Interpolants: A Unifying Framework for Flows and Diffusions

    Michael S Albergo, Nicholas M Boffi, and Eric Vanden-Eijnden. Stochastic interpolants: A unifying framework for flows and diffusions.arXiv preprint arXiv:2303.08797, 2023. 13

    work page Pith review arXiv 2023

  3. [3]

    Polydiffuse: Polygonal shape reconstruc- tion via guided set diffusion models

    Jiacheng Chen, Ruizhi Deng, and Yasutaka Furukawa. Polydiffuse: Polygonal shape reconstruc- tion via guided set diffusion models. InAdvances in Neural Information Processing Systems, volume 36, pages 1863–1888, 2023

    work page 2023

  4. [4]

    ProcTHOR: Large-Scale Embodied AI Using Procedural Generation

    Matt Deitke, Eli VanderBilt, Alvaro Herrasti, Luca Weihs, Jordi Salvador, Kiana Ehsani, Win- son Han, Eric Kolve, Ali Farhadi, Aniruddha Kembhavi, and Roozbeh Mottaghi. ProcTHOR: Large-Scale Embodied AI Using Procedural Generation. InNeurIPS, 2022. Outstanding Paper Award

    work page 2022

  5. [5]

    Domain-adversarial training of neural net- works.Journal of Machine Learning Research, 17(59):1–35, 2016

    Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette, Mario March, and Victor Lempitsky. Domain-adversarial training of neural net- works.Journal of Machine Learning Research, 17(59):1–35, 2016

    work page 2016

  6. [6]

    Floor plan recon- struction from sparse views: Combining graph neural network with constrained diffusion

    Arnaud Gueze, Matthieu Ospici, Damien Rohmer, and Marie-Paule Cani. Floor plan recon- struction from sparse views: Combining graph neural network with constrained diffusion. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1583–1592, 2023

    work page 2023

  7. [7]

    Deformable polyg- onal flow matching with informed priors and hierarchical graph constraints

    Arnaud Gueze, Matthieu Ospici, Damien Rohmer, and Marie-Paule Cani. Deformable polyg- onal flow matching with informed priors and hierarchical graph constraints. InAAAI 2026 Main Conference, Jan 2026

    work page 2026

  8. [8]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. In Advances in neural information processing systems, volume 33, pages 6840–6851, 2020

    work page 2020

Show all 30 references
  1. [9]

    Puzzlefusion: Unleashing the power of diffusion models for spatial puzzle solving

    Sepidehsadat Hosseini, Mohammad Amin Shabani, Saghar Irandoust, and Yasutaka Furukawa. Puzzlefusion: Unleashing the power of diffusion models for spatial puzzle solving. InThirty- seventh Conference on Neural Information Processing Systems, 2023

    2023

  2. [10]

    Graph2plan: learning floorplan generation from layout graphs.ACM Trans

    Ruizhen Hu, Zeyu Huang, Yuhan Tang, Oliver Van Kaick, Hao Zhang, and Hui Huang. Graph2plan: learning floorplan generation from layout graphs.ACM Trans. Graph., 39(4), August 2020

    2020

  3. [11]

    Cubicasa5k, March 2019

    Ahti Kalervo, Juha Ylioinas, Markus Häikiö, Antti Karhu, and Juho Kannala. Cubicasa5k, March 2019

    2019

  4. [12]

    Aria data tools

    Selcuk Karakas, Pierre Moulon, Wenqi Zhang, Nan Yang, Julian Straub, Lingni Ma, Zhaoyang Lv, Elizabeth Argall, Georges Berenger, Tanner Schmidt, Kiran Somasundaram, Vijay Baiyya, Philippe Bouttefroy, Geof Sawaya, Yang Lou, Eric Huang, Tianwei Shen, David Caruso, Bilal Souti, C...

    2022

  5. [13]

    Flow matching for generative modeling

    Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximilian Nickel, and Matthew Le. Flow matching for generative modeling. InThe Eleventh International Conference on Learning Representations, 2023

    2023

  6. [14]

    Flow straight and fast: Learning to generate and transfer data with rectified flow

    Xingchao Liu, Chengyue Gong, and Qiang Liu. Flow straight and fast: Learning to generate and transfer data with rectified flow. InThe Eleventh International Conference on Learning Representations (ICLR), 2023

    2023

  7. [15]

    Learning transferable features with deep adaptation networks

    Mingsheng Long, Yue Cao, Jianmin Wang, and Michael Jordan. Learning transferable features with deep adaptation networks. In Francis Bach and David Blei, editors,Proceedings of the 14 32nd International Conference on Machine Learning, volume 37 ofProceedings of Machine Learning...

    2015

  8. [16]

    A large dataset to train convolutional networks for disparity, optical flow, and scene flow estimation

    Nikolaus Mayer, Eddy Ilg, Philip Hausser, Philipp Fischer, Daniel Cremers, Alexey Dosovitskiy, and Thomas Brox. A large dataset to train convolutional networks for disparity, optical flow, and scene flow estimation. InProceedings of the IEEE Conference on Computer Vision and P...

    2016

  9. [17]

    Scenenet rgb-d: Can 5m synthetic images beat generic imagenet pre-training on indoor segmentation? 2017

    John McCormac, Ankur Handa, Stefan Leutenegger, and Andrew J.Davison. Scenenet rgb-d: Can 5m synthetic images beat generic imagenet pre-training on indoor segmentation? 2017

    2017

  10. [18]

    House- gan: Relational generative adversarial networks for graph-constrained house layout generation

    Nelson Nauata, Kai-Hung Chang, Chin-Yi Cheng, Greg Mori, and Yasutaka Furukawa. House- gan: Relational generative adversarial networks for graph-constrained house layout generation. InComputer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceed...

    2020

  11. [19]

    House-gan++: Generative adversarial layout refinement network towards intelligent computational agent for professional architects

    Nelson Nauata, Sepidehsadat Hosseini, Kai-Hung Chang, Hang Chu, Chin-Yi Cheng, and Yasutaka Furukawa. House-gan++: Generative adversarial layout refinement network towards intelligent computational agent for professional architects. InProceedings of the IEEE/CVF Conference on ...

    2021

  12. [20]

    Learning deep object detectors from 3d models

    Xingchao Peng, Baochen Sun, Karim Ali, and Kate Saenko. Learning deep object detectors from 3d models. InProceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), ICCV ’15, page 1278–1286, USA, 2015. IEEE Computer Society

    2015

  13. [21]

    Housediffusion: Vector floorplan generation via a diffusion model with discrete and continuous denoising

    Mohammad Amin Shabani, Sepidehsadat Hosseini, and Yasutaka Furukawa. Housediffusion: Vector floorplan generation via a diffusion model with discrete and continuous denoising. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5466–5475, 2023

    2023

  14. [22]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. In International Conference on Learning Representations, 2021

    2021

  15. [23]

    Matthias Standfest, Michael Franzen, Yvonne Schröder, Luis Gonzalez Medina, Yarilo Vil- lanueva Hernandez, Jan Hendrik Buck, Yen-Ling Tan, Milena Niedzwiecka, and Rachele Colmegna. Swiss dwellings: A large dataset of apartment models including aggregated geolocation-based simu...

    2022

  16. [24]

    Return of frustratingly easy domain adaptation

    Baochen Sun, Jiashi Feng, and Kate Saenko. Return of frustratingly easy domain adaptation. InAAAI, 2016

    2016

  17. [25]

    Domain randomization for transferring deep neural networks from simulation to the real world

    Josh Tobin, Rachel Fong, Alex Ray, Jonas Schneider, Wojciech Zaremba, and Pieter Abbeel. Domain randomization for transferring deep neural networks from simulation to the real world. In2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), page 23–30....

    2017

  18. [26]

    Training deep networks with synthetic data: Bridging the reality gap by domain randomization

    Jonathan Tremblay, Aayush Prakash, David Acuna, Mark Brophy, Varun Jampani, Cem Anil, Thang To, Eric Cameracci, Shaad Boochoon, and Stan Birchfield. Training deep networks with synthetic data: Bridging the reality gap by domain randomization. 04 2018. 15

    2018

  19. [27]

    Msd: A benchmark dataset for floor plan generation of building complexes

    Casper van Engelenburg, Fatemeh Mostafavi, Emanuel Kuhn, Yuntae Jeon, Michael Franzen, Matthias Standfest, Jan van Gemert, and Seyran Khademi. Msd: A benchmark dataset for floor plan generation of building complexes. In Aleš Leonardis, Elisa Ricci, Stefan Roth, Olga Russakovsk...

    2024

  20. [28]

    Data-driven interior plan generation for residential buildings.ACM Transactions on Graphics (SIGGRAPH Asia), 38(6), 2019

    Wenming Wu, Xiao-Ming Fu, Rui Tang, Yuhan Wang, Yu-Hao Qi, and Ligang Liu. Data-driven interior plan generation for residential buildings.ACM Transactions on Graphics (SIGGRAPH Asia), 38(6), 2019

    2019

  21. [29]

    Structured3d: A large photo-realistic dataset for structured 3d modeling

    Jia Zheng, Junfei Zhang, Jing Li, Rui Tang, Shenghua Gao, and Zihan Zhou. Structured3d: A large photo-realistic dataset for structured 3d modeling. InProceedings of The European Conference on Computer Vision (ECCV), 2020

    2020

  22. [30]

    Domain generalization: A survey.IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(4):4396–4415, 2023

    Kaiyang Zhou, Ziwei Liu, Yu Qiao, Tao Xiang, and Chen Change Loy. Domain generalization: A survey.IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(4):4396–4415, 2023. 16

    2023

Pith tools

Reviewed July 8, 2026 · model on record in the stance chip above.