From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching

arXiv:2607.26817 · cs.RO, cs.CV · Submitted 2026-07-29 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: We're looking at the paper: "From Uncertainty to Determinism".

Dev: Here's the gist of the paper. Visual Floorplan Localization (FLoc) has emerged as a promising solution for indoor localization by matching egocentric images against minimalist structural maps.

Rosa: This paper proposes a coarse-to-fine visual FLoc framework that progresses from uncertainty to determinism, bypassing intermediate ray-matching paradigms.

Dev: Problem Formulation and Motivation Visual FLoc is formulated as estimating the continuous conditional pose distribution p(p I, M) rather than deterministic regression.

Rosa: The optimal pose is theoretically defined by the maximum a posteriori (MAP) estimation of Eq.

Title and authors: Rosa: So we're looking at this paper called "From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching". It tackles how you can figure out where a camera is in a building by matching what it sees to the floorplan map.

Dev: It seems they are focusing on moving away from those old ray matching methods that generate sparse lines, because those methods create this information bottleneck before you even get to the actual localization part.

Taro: So instead of predicting these geometric rays, this paper proposes a different approach where the system directly estimates the entire continuous conditional pose distribution over the map, which is written as p of p given I and M.

Rosa: That's right, they are moving from trying to find one single best answer to modeling all possible answers simultaneously before refining them later.

Dev: They suggest a coarse-to-fine framework where the first stage uses a generative model over the entire floorplan to get a set of plausible candidate poses, which they call p k c.

Taro: And then in the second stage, they take those candidates and use local information to refine them into more precise sub-meter pose residuals.

Rosa: That sounds like they are tackling the problem by first getting a broad idea of where you might be before zooming in on the exact spot using local details.

Dev: The authors say this avoids the issue where visual observations that look identical might map to different locations because it doesn't force a single, deterministic ray prediction.

Taro: They mention that they use a conditional denoising network during the first stage to model this multimodal distribution, injecting Gaussian noise into the pose representation over time.

Rosa: It sounds like they are using diffusion models to explore that entire range of possibilities across the whole map at once.

Dev: The core idea is to bypass those heavy offline preprocessing steps that ray matching needs for its databases, because they propose a rendering-free generative localization framework called CF2Loc.

Taro: And the results on their benchmarks, like Sthree dee and ZInD, show they achieve better recall figures compared to older methods, specifically improving the zero point five meter recall from eleven point one percent down to forty-five point five percent on ZInD.

Rosa: That jump is significant because it shows that this method is more robust when dealing with those tricky cross-modal asymmetries in indoor scenes.

Dev: They also found that applying joint map-pose rotation augmentation really helped boost the recall numbers, pushing the zero point five meter recall up to fifty-eight point five percent and the one point zero meter recall to seventy three point two percent on those benchmarks.

Taro: The ablation studies confirm they aren't just averaging all those diffusion samples randomly; they have to preserve those dominant mode hypotheses in the distribution, otherwise it won't work well.

Rosa: And integrating that local refiner was crucial for getting that fine-grained accuracy, especially once they enabled local patch canonicalization for the crops.

Dev: The final result is that this approach provides state-of-the-art accuracy while also supporting real-time inference without needing any offline map preprocessing or test time look up tables.

Taro: It’s a scalable alternative to those traditional ray matching paradigms, which rely on dense precomputed databases for every map.

Rosa: So, to wrap up the paper "From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching", we see they've successfully moved from uncertainty to determinism by using a generative model first and then local refinement.

Dev: This framework parameterizes the entire multimodal pose posterior via an image conditioned pose diffusion model, which is then followed by a lightweight local refiner for precise residual regression.

Taro: The implication for autonomy is that you get a system that can handle the ambiguity of indoor environments by exploring multiple plausible global hypotheses and then snapping into place locally.

Rosa: This work offers a way to localize indoors without needing those resource-intensive ray matching databases, which is pretty important when you don't have time to prepare huge sets of data for every single map.

Dev: It really shifts the focus from deterministic regression to estimating the continuous conditional pose distribution, making the localization problem itself probabilistic and more realistic.

Taro: I think the big win here is that it shows how you can build a system that progresses from a very fuzzy guess across the whole room to a very precise measurement using localized floorplan crops.

Rosa: It sounds like they've given us a way to do this in real-time, which is what we need for actual robotics outside of controlled lab settings.

Dev: And while they claim it works without offline preprocessing, you have to remember that the performance still depends on how well that initial global proxy distribution captures the true multimodal nature of the scene.

Taro: That's a fair point; if the first stage misses a dominant mode, the refinement stage won't be able to fix it correctly.

Rosa: So, looking ahead at this paper, I think we should keep watching how they handle those real-world misbehaves when they move this out of the controlled environment.

Dev: And for now, this CF2Loc framework gives us a solid foundation that doesn't rely on precomputed structures for every floorplan.

Taro: We'll see if the system can gracefully recover when things get truly messy outside of the perfect test conditions they used in their evaluations.

The paper's summary: Rosa: So, we're talking about how this new visual floorplan localization system works by moving away from those old ray matching methods that create sparse lines in the first place.

Dev: It’s about modeling the entire continuous conditional pose distribution across a whole map directly, instead of just picking one single best answer.

Rosa: Right, so they're using this coarse-to-fine structure where you get a broad idea of where you might be across the whole floorplan first.

Dev: And then you use local information to refine those broad guesses into very precise sub-meter corrections by looking at small patches of the map.

Rosa: It means they bypass that information bottleneck before localization and they don't need those huge offline databases that ray matching requires for every single map.

Dev: The results on benchmarks like Sthree dee and ZInD are pretty strong, showing improvements in recall figures, like taking the five point one percent recall up to forty-five point five percent on ZInD.

Rosa: And they found that adding joint map and pose rotation augmentation helped push that recall even higher, up to fifty-eight point five percent for the small distance guesses.

Dev: That local refiner is key because it eliminates those structural ambiguities locally, which gives them better fine-grained accuracy without needing to do any test time lookups.

Rosa: It really shows how you can build a system that handles the ambiguity of indoor scenes by exploring multiple plausible global hypotheses and then snapping into place locally.

Dev: This approach shifts the problem from just predicting a single pose to estimating the entire multimodal posterior, which is much more realistic for real-world scenarios where things aren't perfectly clear.

Rosa: It’s interesting because it doesn't rely on those heavy preprocessing steps or those test time lookup tables that ray matching needs.

Dev: The challenge then becomes making sure that initial global proxy distribution actually captures the true multimodal nature of the scene, otherwise the refinement stage won't be able to fix it correctly.

Rosa: So, what this means for us is a scalable way to localize indoors without needing those resource-intensive ray matching databases for every single map.

Dev: And this moves the focus from deterministic regression to estimating that continuous conditional pose distribution, making the localization problem itself probabilistic and more realistic.

Rosa: We'll see how this performs when it gets out of the controlled lab setting, especially with those real-world misbehaves Taro is testing for.

The paper's improvements: Tom: So, we're looking at how they're changing the approach by focusing on direct conditional pose distribution estimation instead of those old ray representations.

Rosa: That’s right, they're bypassing that information bottleneck before localization and they aren't using those huge offline databases that ray matching requires for every single map.

Dev: They introduce this coarse-to-fine structure where you get a broad idea of where you might be across the whole floorplan first.

Rosa: And then you use local information to refine those broad guesses into very precise sub-meter corrections by looking at small patches of the map.

Dev: The local refiner is crucial because it eliminates those structural ambiguities locally, which gives them better fine-grained accuracy without needing to do any test time lookups.

Rosa: It means they’re not just guessing a single point; they're modeling all the possibilities and then zooming in on the right one using local context.

Dev: This method parameterizes that entire multimodal posterior via an image conditioned pose diffusion model, which is then followed by that lightweight local refiner for precise residual regression.

Rosa: It really shows how you can build a system that handles the ambiguity of indoor scenes by exploring multiple plausible global hypotheses and then snapping into place locally.

Dev: This approach shifts the problem from just predicting a single pose to estimating the entire multimodal posterior, which is much more realistic for real-world scenarios where things aren't perfectly clear.

Rosa: It’s interesting because it doesn't rely on those heavy preprocessing steps or those test time lookup tables that ray matching needs.

Dev: The challenge then becomes making sure that initial global proxy distribution actually captures the true multimodal nature of the scene, otherwise the refinement stage won't be able to fix it correctly.

Rosa: So, what this means for us is a scalable way to localize indoors without needing those resource-intensive ray matching databases for every single map.

Dev: This moves the focus from deterministic regression to estimating that continuous conditional pose distribution, making the localization problem itself probabilistic and more realistic.

Rosa: We'll see how this performs when it gets out of the controlled lab setting, especially with those real-world misbehaves Taro is testing for.

Conclusion: Rosa: So, to wrap up "From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching," they've shown how you can go from just guessing a general location to a very precise measurement by using this generative framework first and then local refinement.

Dev: It’s interesting because it completely bypasses the need for those heavy offline preprocessing steps or test time lookups that ray matching requires for every single map.

Rosa: Exactly, so you get this system that can handle the ambiguity of indoor scenes by exploring multiple plausible global hypotheses and then snapping into place locally.

Dev: This moves the focus from just predicting a single pose to estimating that continuous conditional pose distribution, which is much more realistic for real-world scenarios where things aren't perfectly clear.

Rosa: We'll see how this performs when it gets out of the controlled lab setting, especially with those real-world misbehaves Taro is testing for.

Dev: The challenge then becomes making sure that initial global proxy distribution actually captures the true multimodal nature of the scene, otherwise the refinement stage won't be able to fix it correctly.

Rosa: It’s a scalable way to localize indoors without needing those resource-intensive ray matching databases for every single map.

Dev: This moves the focus from deterministic regression to estimating that continuous conditional pose distribution, making the localization problem itself probabilistic and more realistic.

Rosa: We'll see how this performs when it gets out of the controlled lab setting, especially with those real-world misbehaves Taro is testing for.

Dev: The safety case here is pretty strong because it handles uncertainty directly in the modeling stage instead of relying on a single deterministic guess.

Rosa: It sounds like a solid foundation that doesn't rely on precomputed structures for every floorplan, which is pretty important when you don't have time to prepare huge sets of data for every single map.

Dev: And if we can get this kind of direct modeling working reliably in real-time, it opens the door for more robust systems in mobile robotics where the environment is constantly changing.

Rosa: Yeah, I think this paper provides a really tangible method for how we can tackle that continuous pose distribution problem without being stuck with the limitations of older ray matching techniques.

Dev: Exactly, and we'll see if this system can gracefully recover when things get truly messy outside of the perfect test conditions they used in their evaluations.

Taro: I think what really stands out is how they address those real-world misbehaves during testing; it shows the system can handle uncertainty more gracefully than just averaging samples randomly.

Rosa: That’s true, and seeing how it performs when the world misbehaves outside of a lab setting is what we need to see next.

Dev: For now, this CF2Loc framework gives us a solid foundation that doesn't rely on precomputed structures for every floorplan.

Shiyong Meng, *Bolei Chen*, *Ping Zhong*, Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang

School of Computer Science and Engineering, Central South University

cs.RO, cs.CV

Submitted: 2026-07-29

Updated: 2026-11-16

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 91/100

The gist: Visual Floorplan Localization (FLoc) has emerged as a promising solution for indoor localization by matching egocentric images against minimalist structural maps.

Key concepts

Continuous Conditional Pose Distribution
Instead of guessing a single location, this concept models the entire range of possible poses as a continuous distribution. It represents all plausible locations for the camera based on both the visual input and the floorplan map, allowing for more robust uncertainty modeling than traditional methods.
Global Pose Diffusion Model
This is a generative model that uses progressive noise injection to estimate how a pose should look across the whole map. It learns to generate a diverse set of possible poses simultaneously, capturing the multimodal nature of localization without relying on intermediate ray calculations.
Local Pose Refiner
This component takes coarse guesses from the global model and then focuses only on small sections (crops) of the floorplan. It predicts precise, bounded errors (residuals) for translation and orientation within that local area, significantly improving accuracy by handling fine details.
Coarse-to-Fine Inference
This is a two-stage process where the system first generates broad hypotheses across the whole map (coarse stage) and then uses those hypotheses to refine them locally (fine stage). This hierarchical approach ensures both global context and precise local accuracy are captured effectively.

Terminology

Summary

Visual Floorplan Localization (FLoc) has emerged as a promising solution for indoor localization by matching egocentric images against minimalist structural maps. p. 0

Where the paper says it

cn, jxwang@mail. cn Abstract Visual Floorplan Localization (FLoc) has emerged as a promising solution for indoor localization by matching egocentric images against minimalist structural maps.

This paper proposes a coarse-to-fine visual FLoc framework that progresses from uncertainty to determinism, bypassing intermediate ray-matching paradigms. p. 0
Where the paper says it

Existing ray-matching-based methods tackle this by explicitly predicting sparse geometric or semantic rays, which inherently incurs information loss and demands resource-intensive preprocessing alongside exhaustive matching during inference. In this paper, we bypass the intermediate ray-matching paradigm and propose a coarse-to-fine visual FLoc framework that progresses from uncertainty to determinism.

Problem Formulation and Motivation

Visual FLoc is formulated as estimating the continuous conditional pose distribution p(p I, M) rather than deterministic regression. The optimal pose is theoretically defined by the maximum a posteriori (MAP) estimation of Eq. p. 2

Where the paper says it

To accommodate the intrinsic ambiguities of indoor structures, we move beyond deterministic regression and instead formulate localization as estimating the continuous conditional pose distribution p(p | I, M). The optimal pose is theoretically defined by the maximum a posteriori (MAP) estimation: p ∗ = arg max p p(p | I, M).

(1). The pronounced cross-modal asymmetry between feature-rich visual observations and minimalist 2D floorplans renders visual localization a highly ill-posed problem. p. 2
Where the paper says it

Motivation. The pronounced cross-modal asymmetry between feature-rich visual observations and minimalist 2D floorplans renders visual localization a highly ill-posed problem.

Existing deterministic regressors enforce a single-peak constraint on the output, leading to mode collapse and significant averaging errors in ambiguous regions. p. 2
Where the paper says it

Existing paradigms struggle with this multimodal nature. Deterministic regressors enforce a single-peak constraint on the output, leading to mode collapse and significant averaging errors in ambiguous regions.

Global Pose Diffusion

The global pose diffusion model directly estimates the continuous conditional pose distribution across the entire floorplan. This involves modeling the multimodal pose distribution through conditional diffusion, where Gaussian noise is progressively injected into the pose representation zt. p. 2

Where the paper says it

Following the standard DDPM (Ho, Jain, and Abbeel 2020) formulation, Gaussian noise is progressively injected into the pose representation: zt = √ α¯tz0 + √ 1 − α¯tϵ, ϵ ∼ N (0, I), (5) where z0 = p˜. A conditional denoising network predicts the injected noise from the noisy pose, diffusion timestep, global image descriptor, and observation-conditioned map tokens: ˆϵ = ϵθ(zt, t, FˆM, gI ).

A conditional denoising network predicts the injected noise from the noisy pose, diffusion timestep, global image descriptor, and observation-conditioned map tokens. p. 2
Where the paper says it

Following the standard DDPM (Ho, Jain, and Abbeel 2020) formulation, Gaussian noise is progressively injected into the pose representation: zt = √ α¯tz0 + √ 1 − α¯tϵ, ϵ ∼ N (0, I), (5) where z0 = p˜. A conditional denoising network predicts the injected noise from the noisy pose, diffusion timestep, global image descriptor, and observation-conditioned map tokens: ˆϵ = ϵθ(zt, t, FˆM, gI ).

Local Pose Refinement

The local refiner is introduced to predict bounded sub-meter pose residuals from candidate-centered floorplan crops. This refinement operates on localized floorplan crops while reusing the visual tokens extracted during the coarse stage. p. 3

Where the paper says it

2(a), the refiner operates on localized floorplan crops while reusing the visual tokens extracted during the coarse stage. Candidate-Centered Local Refinement.

The refiner predicts translation residuals ∆xˆk and orientation residuals ∆ˆθk using prediction heads Rϕ. p. 3
Where the paper says it

(9) Using the fused local representation, three lightweight prediction heads (unified as Rϕ) respectively estimate position logits l k = {l k j } Nc j=1, local heading vectors Vk = {v k j } Nc j=1, and a candidate confidence score sˆ k : (l k , V k , sˆ k ) = Rϕ(I, Mk c , p k c ). (10) Rather than directly regressing coordinates via global feature pooling, we compute translation residuals by constructing a spatial localization distribution over local map tokens: π k j = exp(l k j /τ ) P n exp(l k n/τ ) , ∆xˆ k = X j π k j uj , (11) where uj denotes the metric coordinate of token j in the candidate-local frame.

The final refined pose p kr is obtained by applying the predicted residual to the coarse candidate pose p kc. p. 3
Where the paper says it

Similarly, spatial heading vectors are aggregated via probability-weighted summation to determine the rotation residual: ∆ˆθ k = atan2   "X j π k j v k j # 1 , "X j π k j v k j # 2   . (12) The refined pose p k r is obtained by applying the predicted residual to the coarse candidate pose p k c ⊕ (∆xˆ k , ∆ˆθ k ).

Coarse-to-Fine Inference and Results

During inference, N pose particles are initialized from Gaussian noise and refined through 10 reverse diffusion steps to approximate the continuous multimodal pose posterior. Candidate modes are extracted via Kernel Density Estimation (KDE) to form coarse candidate modes. p. 4

Where the paper says it

To extract representative pose modes from the sample set, we estimate particle density over the pose manifold via Kernel Density Estimation (KDE): D(p) = 1 N XN i=1 KH(p − p i ), (17) where KH(·) denotes a product kernel in SE(2) space with bandwidth matrix H. The K density local maxima are extracted as coarse candidate modes {p k c } K k=1.

The top-ranked candidate p k r is output as the final localization pose. p. 4
Where the paper says it

Finally, candidate hypotheses are re-ranked based on their predicted confidence scores: k ∗ = arg max k∈{1,. ,K} sˆ k , (18) and the top-ranked candidate p k ∗ r is output as the final localization pose.

Comprehensive results on the S3D (full) and ZInD benchmarks demonstrate that our approach achieves state-of-the-art accuracy and robustness. p. 0
Where the paper says it

Our method effectively balances global multi-hypothesis tracking and local sub-meter refinement without requiring any offline map preprocessing or test-time lookup tables. Comprehensive results on the S3D (full) and ZInD benchmarks demonstrate that our approach achieves state-of-the-art accuracy and robustness.

The method improves the 0. p. 5
Where the paper says it

When training with semantic floorplans, our method continues to establish new SOTA performance. Compared with the strongest semantic baseline, FoD, our framework improves recall on S3D (full) from 12.

5 m recall from 11. p. 4
Where the paper says it

2024; Grader and Averbuch-Elor 2025; Ye et al. 2026), we report recall metrics at spatial accuracy thresholds of 0.

1% (LASER) to 45. p. 4
Where the paper says it

1999) is the most popular framework for 2D localization on pure geometry maps. (3) LASER (Min et al.

5% on ZInD. p. 0
Where the paper says it

Our method effectively balances global multi-hypothesis tracking and local sub-meter refinement without requiring any offline map preprocessing or test-time lookup tables. Comprehensive results on the S3D (full) and ZInD benchmarks demonstrate that our approach achieves state-of-the-art accuracy and robustness.

Ablation Studies

Ablation studies confirm the necessity of preserving dominant mode hypotheses in a multimodal pose distribution rather than naively averaging diffusion samples. Applying joint map-pose rotation augmentation further boosts the 0. p. 6

Where the paper says it

This confirms the necessity of preserving dominant mode hypotheses in a multimodal pose distribution rather than naively averaging diffusion samples. Applying joint map-pose rotation augmentation further boosts the 0.

5 m recall to 58. p. 4
Where the paper says it

2024; Grader and Averbuch-Elor 2025; Ye et al. 2026), we report recall metrics at spatial accuracy thresholds of 0.

5% and the 1.0 m recall to 73. p. 4
Where the paper says it

2024; Grader and Averbuch-Elor 2025; Ye et al. 2026), we report recall metrics at spatial accuracy thresholds of 0.

2%. Integrating the local refiner consistently improves fine-grained localization accuracy, with gains becoming significantly more pronounced after enabling local patch canonicalization. p. 6
Where the paper says it

2%, indicating enhanced robustness to arbitrary global floorplan orientations. Integrating the local refiner consistently improves fine-grained localization accuracy, with gains becoming significantly more pronounced after enabling local patch canonicalization (rotation alignment).

The 5 m × 5 m crop yields the best accuracy by balancing spatial layout context with local unimodality. p. 7
Where the paper says it

Parametric Studies We conduct sensitivity analyses on the local map size. As shown in Table 4, a 5 m × 5 m crop yields the best accuracy by balancing spatial layout context with local unimodality—smaller patches lack context, whereas larger ones reintroduce structural ambiguities.

Conclusion

In this paper, we revisit visual floorplan localization (FLoc) from the perspective of direct conditional pose distribution estimation and propose a rendering-free, coarse-to-fine generative framework. Our method parameterizes the full multimodal pose posterior via an image-conditioned pose diffusion model, followed by a lightweight local refiner for precise residual regression. p. 7

Where the paper says it

Conclusion In this paper, we revisit visual floorplan localization (FLoc) from the perspective of direct conditional pose distribution estimation and propose a rendering-free, coarse-to-fine generative framework. Rather than relying on intermediate 1D ray prediction or exhaustive observation matching, our method parameterizes the full multimodal pose posterior via an image-conditioned pose diffusion model, followed by a lightweight local refiner for precise residual regression.

Extensive evaluations on the S3D (full) and ZInD benchmarks demonstrate that our approach consistently establishes new SOTA accuracy while supporting real-time inference without requiring offline map preprocessing. p. 7
Where the paper says it

This coarse-to-fine formulation seamlessly bridges global uncertainty modeling with local unimodal determinism, enabling robust multi-hypothesis tracking and fine-grained pose estimation within a unified architecture applicable to both geometric and semantic floorplans. Extensive evaluations on the S3D (full) and ZInD benchmarks demonstrate that our approach consistently establishes new SOTA accuracy while supporting real-time inference without requiring offline map preprocessing.

This work provides a scalable, rendering-free alternative to traditional ray-matching paradigms. p. 7
Where the paper says it

Extensive evaluations on the S3D (full) and ZInD benchmarks demonstrate that our approach consistently establishes new SOTA accuracy while supporting real-time inference without requiring offline map preprocessing. We hope this work provides a scalable, rendering-free alternative to traditional ray-matching paradigms and inspires further research in generative crossmodal spatial reasoning.

The paper's code will be publicly released.

References

Arandjelović, R.; Gronat, P. p. 7

Where the paper says it

References Arandjelović, R. ; Gronat, P.

; Torii, A. p. 7
Where the paper says it

; Gronat, P. ; Torii, A.

; Pajdla, T. p. 7
Where the paper says it

; Torii, A. ; Pajdla, T.

; and Sivic, J. p. 7
Where the paper says it

; Pajdla, T. ; and Sivic, J.

2016. NetVLAD: CNN Architecture for Weakly Supervised Place Recognition. p. 7
Where the paper says it

; and Sivic, J. NetVLAD: CNN Architecture for Weakly Supervised Place Recognition.

In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 5297–5307. p. 7
Where the paper says it

NetVLAD: CNN Architecture for Weakly Supervised Place Recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 5297– 5307.

Balntas, V. p. 0

Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

; Li, S.; and Prisacariu, V. p. 0
Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

p. 0
Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

p. 0
Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

A. 2018. RelocNet: Continuous Metric Learning Relocalisation Using Neural Nets. p. 7
Where the paper says it

; and Prisacariu, V. RelocNet: Continuous Metric Learning Relocalisation Using Neural Nets.

In Proceedings of the European Conference on Computer Vision (ECCV), 751–767. p. 7
Where the paper says it

RelocNet: Continuous Metric Learning Relocalisation Using Neural Nets. In Proceedings of the European Conference on Computer Vision (ECCV), 751–767.

Brachmann, E. p. 1

Where the paper says it

2019; Sattler, Leibe, and Kobbelt 2017), or direct pose regression using deep networks (Kendall, Grimes, and Cipolla 2015; Walch et al. 2017; Brachmann et al.

; Krull, A. p. 7
Where the paper says it

Brachmann, E. ; Krull, A.

; Nowozin, S. p. 7
Where the paper says it

; Krull, A. ; Nowozin, S.

; Shotton, J. p. 7
Where the paper says it

; Nowozin, S. ; Shotton, J.

; Michel, F. p. 7
Where the paper says it

; Shotton, J. ; Michel, F.

; Gumhold, S. p. 7
Where the paper says it

; Michel, F. ; Gumhold, S.

; and Rother, C. p. 7
Where the paper says it

; Gumhold, S. ; and Rother, C.

2017. Dsac-differentiable ransac for camera localization. p. 7
Where the paper says it

; and Rother, C. Dsac-differentiable ransac for camera localization.

In Proceedings of the IEEE conference on computer vision and pattern recognition, 6684–6692. p. 7
Where the paper says it

Dsac-differentiable ransac for camera localization. In Proceedings of the IEEE conference on computer vision and pattern recognition, 6684–6692.

Chen, B.; Kang, J.; Yang, H.; Zhong, P. p. 0

Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

p. 0
Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

; and Wang, J. p. 0
Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

p. 0
Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

2025. Perspective from a Higher Dimension: Can 3D Geometric Priors Help Visual Floorplan Localization? p. 7
Where the paper says it

; and Wang, J. Perspective from a Higher Dimension: Can 3D Geometric Priors Help Visual Floorplan Localization?

In Proceedings of the 33rd ACM International Conference on Multimedia (ACM MM), 7181–7190. p. 7
Where the paper says it

Perspective from a Higher Dimension: Can 3D Geometric Priors Help Visual Floorplan Localization? In Proceedings of the 33rd ACM International Conference on Multimedia (ACM MM), 7181–7190.

Chen, B.; Yan, S.; Cui, Y.; Kang, J.; Zhong, P.; and Wang, J. 2026a. Perspective from a Broader Context: Can Room Style Knowledge Help Visual Floorplan Localization? p. 7

Where the paper says it

; and Wang, J. Perspective from a Broader Context: Can Room Style Knowledge Help Visual Floorplan Localization?

In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, 2787–2795. p. 7
Where the paper says it

Perspective from a Broader Context: Can Room Style Knowledge Help Visual Floorplan Localization? In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, 2787–2795.

Chen, C.; Wang, R.; Vogel, C. p. 7

Where the paper says it

In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, 2787–2795. ; Vogel, C.

; and Pollefeys, M. p. 7
Where the paper says it

; Vogel, C. ; and Pollefeys, M.

2024. F3Loc: Fusion and Filtering for Floorplan Localization. p. 7
Where the paper says it

; and Pollefeys, M. F 3Loc: Fusion and Filtering for Floorplan Localization.

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 18029–18038. p. 7
Where the paper says it

F 3Loc: Fusion and Filtering for Floorplan Localization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 18029–18038.

Chen, K.; Huang, Y. p. 0

Where the paper says it

Compared with dense 3D models, floorplans are lightweight, easy to maintain, and readily available in built environments, making visual FLoc particularly appealing for autonomous robot navigation (Li et al. 2025; Huang et al.

; An, D.; He, J.; Su, Y.; Liu, J.; Liu, N.; and Wang, L. p. 0
Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

2026b. FloorPlan-VLN: A New Paradigm for Floor Plan Guided Vision-Language Navigation. p. 7
Where the paper says it

; and Wang, L. FloorPlan-VLN: A New Paradigm for Floor Plan Guided Vision-Language Navigation.

In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, 14610–14618. p. 7
Where the paper says it

FloNa: Floor Plan Guided Embodied Visual Navigation. p. 7

Where the paper says it

; and Liu, F. p. 0

Where the paper says it

Introduction Visual Floorplan Localization (FLoc) (Chen et al. 2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al.

FloNa: Floor Plan Guided Embodied Visual Navigation.

In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, 14610–14618.

p. 7
Where the paper says it

FloNa: Floor Plan Guided Embodied Visual Navigation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, 14610–14618.

Cho, J.; Cai, R.; and Averbuch-Elor, H. p. 1

Where the paper says it

2026; Chen et al. 2025, 2026a; Grader and Averbuch-Elor 2025; Ye et al.

p. 1
Where the paper says it

2026; Chen et al. 2025, 2026a; Grader and Averbuch-Elor 2025; Ye et al.

2026. SceneAligner: 3D-Grounded Floorplan Localization in the Wild. p. 7
Where the paper says it

; and Averbuch-Elor, H. SceneAligner: 3D-Grounded Floorplan Localization in the Wild.

arXiv preprint arXiv:2605. p. 7
Where the paper says it

SceneAligner: 3D-Grounded Floorplan Localization in the Wild. arXiv preprint arXiv:2605.

22581.

Cruz, S.; Hutchcroft, W. p. 7

Where the paper says it

arXiv preprint arXiv:2605. ; Hutchcroft, W.

; Li, Y.; Khosravan, N. p. 7
Where the paper says it

; Hutchcroft, W. ; Khosravan, N.

p. 7
Where the paper says it

; Hutchcroft, W. ; Khosravan, N.

; Boyadzhiev, I. p. 7
Where the paper says it

; Khosravan, N. ; Boyadzhiev, I.

; and Kang, S. p. 7
Where the paper says it

; Boyadzhiev, I. ; and Kang, S.

B. 2021. Zillow Indoor Dataset: Annotated Floor Plans With 360° Panoramas and 3D Room Layouts. p. 7
Where the paper says it

; and Kang, S. Zillow Indoor Dataset: Annotated Floor Plans With 360° Panoramas and 3D Room Layouts.

IEEE.

Dellaert, F. p. 4

Where the paper says it

Its observation model aims to learn the similarity between an image and the corresponding map patch. (2) MCL (Dellaert et al.

; Fox, D.; Burgard, W. p. 7
Where the paper says it

Dellaert, F. ; Burgard, W.

; and Thrun, S. p. 7
Where the paper says it

; Burgard, W. ; and Thrun, S.

1999. Monte Carlo Localization for Mobile Robots. p. 7
Where the paper says it

; and Thrun, S. Monte Carlo Localization for Mobile Robots.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), volume 2, 1322–1328. p. 7
Where the paper says it

Monte Carlo Localization for Mobile Robots. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), volume 2, 1322–1328.

Grader, Y.; and Averbuch-Elor, H. 2025. Supercharging Floorplan Localization with Semantic Rays. p. 7

Where the paper says it

; and Averbuch-Elor, H. Supercharging Floorplan Localization with Semantic Rays.

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 27116–27125. p. 7
Where the paper says it

Supercharging Floorplan Localization with Semantic Rays. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 27116–27125.

Ho, J.; Jain, A.; and Abbeel, P. p. 2

Where the paper says it

To circumvent boundary discontinuities associated with angular regression, the camera pose is parameterized as p˜ = (x, y, cos θ,sin θ), where orientation is recovered using atan2(·) during inference. Following the standard DDPM (Ho, Jain, and Abbeel 2020) formulation, Gaussian noise is progressively injected into the pose representation: zt = √ α¯tz0 + √ 1 − α¯tϵ, ϵ ∼ N (0, I), (5) where z0 = p˜.

2020. Denoising Diffusion Probabilistic Models. p. 7
Where the paper says it

; and Abbeel, P. Denoising Diffusion Probabilistic Models.

In Larochelle, H. p. 7
Where the paper says it

Denoising Diffusion Probabilistic Models. In Larochelle, H.

; Ranzato, M. p. 7
Where the paper says it

In Larochelle, H. ; Ranzato, M.

; Hadsell, R. p. 7
Where the paper says it

; Ranzato, M. ; Hadsell, R.

; Balcan, M. p. 7
Where the paper says it

; Hadsell, R. ; Balcan, M.

; and Lin, H. p. 7
Where the paper says it

; Balcan, M. ; and Lin, H.

, eds., Advances in Neural Information Processing Systems, volume 33, 6840–6851. p. 7
Where the paper says it

; and Lin, H. , Advances in Neural Information Processing Systems, volume 33, 6840–6851.

Howard-Jenkins, H. p. 1

Where the paper says it

End-to-End Visual FLoc End-to-end visual FLoc directly learns a mapping from egocentric observations and floorplans to camera poses without explicit geometric matching. Representative methods, such as LaLaLoc (Howard-Jenkins, Ruiz-Sarmiento, and Prisacariu 2021), LaLaLoc++ (Howard-Jenkins and Prisacariu 2022), and PF-Net (Karkus, Hsu, and Lee 2018), learn joint image–floorplan representations for direct pose regression or particle-filter-based localization.

p. 1
Where the paper says it

End-to-End Visual FLoc End-to-end visual FLoc directly learns a mapping from egocentric observations and floorplans to camera poses without explicit geometric matching. Representative methods, such as LaLaLoc (Howard-Jenkins, Ruiz-Sarmiento, and Prisacariu 2021), LaLaLoc++ (Howard-Jenkins and Prisacariu 2022), and PF-Net (Karkus, Hsu, and Lee 2018), learn joint image–floorplan representations for direct pose regression or particle-filter-based localization.

; and Prisacariu, V. A. 2022. LaLaLoc++: Global Floor Plan Comprehension for Layout Localisation in Unvisited Environments. p. 7
Where the paper says it

; and Prisacariu, V. LaLaLoc++: Global Floor Plan Comprehension for Layout Localisation in Unvisited Environments.

In Proceedings of the European Conference on Computer Vision (ECCV), 693–709. p. 7
Where the paper says it

LaLaLoc++: Global Floor Plan Comprehension for Layout Localisation in Unvisited Environments. In Proceedings of the European Conference on Computer Vision (ECCV), 693– 709.

Howard-Jenkins, H.; Ruiz-Sarmiento, J. p. 1

Where the paper says it

End-to-End Visual FLoc End-to-end visual FLoc directly learns a mapping from egocentric observations and floorplans to camera poses without explicit geometric matching. Representative methods, such as LaLaLoc (Howard-Jenkins, Ruiz-Sarmiento, and Prisacariu 2021), LaLaLoc++ (Howard-Jenkins and Prisacariu 2022), and PF-Net (Karkus, Hsu, and Lee 2018), learn joint image–floorplan representations for direct pose regression or particle-filter-based localization.

-R.; and Prisacariu, V. A. 2021. LaLaLoc: Latent Layout Localisation in Dynamic, Unvisited Environments. p. 7
Where the paper says it

; and Prisacariu, V. LaLaLoc: Latent Layout Localisation in Dynamic, Unvisited Environments.

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 10107–10116. p. 7
Where the paper says it

LaLaLoc: Latent Layout Localisation in Dynamic, Unvisited Environments. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 10107–10116.

Huang, W.; Li, J.; Wan, Z.; Di, H.; Liang, W. p. 7

Where the paper says it

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 10107–10116. ; Liang, W.

p. 7
Where the paper says it

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 10107–10116. ; Liang, W.

; and Yang, Z. p. 0
Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

2025. Floor Plan-Guided Visual Navigation Incorporating Depth and Directional Cues. p. 7
Where the paper says it

; and Yang, Z. Floor Plan-Guided Visual Navigation Incorporating Depth and Directional Cues.

arXiv preprint arXiv:2511. p. 7
Where the paper says it

Floor Plan-Guided Visual Navigation Incorporating Depth and Directional Cues. arXiv preprint arXiv:2511.

01493.

Karkus, P.; Hsu, D.; and Lee, W. p. 1

Where the paper says it

End-to-End Visual FLoc End-to-end visual FLoc directly learns a mapping from egocentric observations and floorplans to camera poses without explicit geometric matching. Representative methods, such as LaLaLoc (Howard-Jenkins, Ruiz-Sarmiento, and Prisacariu 2021), LaLaLoc++ (Howard-Jenkins and Prisacariu 2022), and PF-Net (Karkus, Hsu, and Lee 2018), learn joint image–floorplan representations for direct pose regression or particle-filter-based localization.

S. 2018. Particle Filter Networks with Application to Visual Localization. p. 7
Where the paper says it

; and Lee, W. Particle Filter Networks with Application to Visual Localization.

In Proceedings of the 2nd Conference on Robot Learning (CoRL), volume 87 of Proceedings of Machine Learning Research, 169–178 PMLR. p. 7
Where the paper says it

Particle Filter Networks with Application to Visual Localization. In Proceedings of the 2nd Conference on Robot Learning (CoRL), volume 87 of Proceedings of Machine Learning Research, 169–178.

Kendall, A. p. 1

Where the paper says it

2016; Balntas, Li, and Prisacariu 2018), 2D–3D feature matching with Structure-from-Motion (SfM) models (Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017), or direct pose regression using deep networks (Kendall, Grimes, and Cipolla 2015; Walch et al.

; Grimes, M. p. 1
Where the paper says it

2016; Balntas, Li, and Prisacariu 2018), 2D–3D feature matching with Structure-from-Motion (SfM) models (Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017), or direct pose regression using deep networks (Kendall, Grimes, and Cipolla 2015; Walch et al.

; and Cipolla, R. p. 1
Where the paper says it

2016; Balntas, Li, and Prisacariu 2018), 2D–3D feature matching with Structure-from-Motion (SfM) models (Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017), or direct pose regression using deep networks (Kendall, Grimes, and Cipolla 2015; Walch et al.

2015. Posenet: A convolutional network for real-time 6-dof camera relocalization. p. 7
Where the paper says it

; and Cipolla, R. Posenet: A convolutional network for real-time 6-dof camera relocalization.

In Proceedings of the IEEE international conference on computer vision, 2938–2946. p. 7
Where the paper says it

Posenet: A convolutional network for real-time 6-dof camera relocalization. In Proceedings of the IEEE international conference on computer vision, 2938–2946.

Li, J.; Huang, W. p. 0

Where the paper says it

Compared with dense 3D models, floorplans are lightweight, easy to maintain, and readily available in built environments, making visual FLoc particularly appealing for autonomous robot navigation (Li et al. 2025; Huang et al.

; Wang, Z.; Liang, W.; Di, H.; and Liu, F. 2025. FloNa: Floor Plan Guided Embodied Visual Navigation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, 14610–14618.

Liu, L.; Li, H.; and Dai, Y. p. 0

Where the paper says it

Introduction Visual Floorplan Localization (FLoc) (Chen et al. 2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al.

2017. Efficient Global 2D–3D Matching for Camera Localization in a Large-Scale 3D Map. p. 7
Where the paper says it

; and Dai, Y. Efficient Global 2D– 3D Matching for Camera Localization in a Large-Scale 3D

In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2372–2381. p. 8
Where the paper says it

In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2372–2381. Loshchilov, I. p. 5

Where the paper says it

All attention mechanisms employ four heads with a dropout rate of 0. The diffusion model is optimized via AdamW (Loshchilov and Hutter 2017) using a cosine noise schedule (T = 1000), a batch size of 8, an initial learning rate of 10−4 , a minimum learning rate of 10−5 , and a weight decay of 10−4 .

Loshchilov, I.; and Hutter, F. p. 5

Where the paper says it

All attention mechanisms employ four heads with a dropout rate of 0. The diffusion model is optimized via AdamW (Loshchilov and Hutter 2017) using a cosine noise schedule (T = 1000), a batch size of 8, an initial learning rate of 10−4 , a minimum learning rate of 10−5 , and a weight decay of 10−4 .

2017. Decoupled Weight Decay Regularization. p. 8
Where the paper says it

; and Hutter, F. Decoupled Weight Decay Regularization.

In International Conference on Learning Representations. p. 8
Where the paper says it

Decoupled Weight Decay Regularization. In International Conference on Learning Representations.

Min, Z.; Khosravan, N.; Bessinger, Z. p. 8

Where the paper says it

; Khosravan, N. ; Bessinger, Z.

; Narayana, M. p. 8
Where the paper says it

; Bessinger, Z. ; Narayana, M.

; Kang, S. B.; Dunn, E.; and Boyadzhiev, I. p. 7
Where the paper says it

; Khosravan, N. ; Boyadzhiev, I.

2022. LASER: LAtent SpacE Rendering for 2D Visual Localization. p. 8
Where the paper says it

; and Boyadzhiev, I. LASER: LAtent SpacE Rendering for 2D Visual Localization.

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 11122–11131. p. 8
Where the paper says it

LASER: LAtent SpacE Rendering for 2D Visual Localization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 11122–11131.

Sarlin, P.-E.; Cadena, C. p. 8

Where the paper says it

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 11122–11131. ; Cadena, C.

; Siegwart, R. p. 8
Where the paper says it

; Cadena, C. ; Siegwart, R.

; and Dymczyk, M. p. 8
Where the paper says it

; Siegwart, R. ; and Dymczyk, M.

2019. From Coarse to Fine: Robust Hierarchical Localization at Large Scale. p. 8
Where the paper says it

; and Dymczyk, M. From Coarse to Fine: Robust Hierarchical Localization at Large Scale.

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 12716–12725. p. 8
Where the paper says it

From Coarse to Fine: Robust Hierarchical Localization at Large Scale. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 12716–12725.

Sattler, T. p. 0

Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

; Leibe, B. p. 0
Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

; and Kobbelt, L. p. 0
Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

2017. Efficient and Effective Prioritized Matching for Large-Scale Image-Based Localization. p. 8
Where the paper says it

; and Kobbelt, L. Efficient and Effective Prioritized Matching for Large-Scale Image-Based Localization.

IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(9): 1744–1756. p. 8
Where the paper says it

Efficient and Effective Prioritized Matching for Large-Scale Image-Based Localization. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(9): 1744–1756.

Vaswani, A. p. 5

Where the paper says it

The floorplan encoder comprises a convolutional stem and the first two stages of ResNet-18, followed by two residual context blocks. The global diffusion denoising network incorporates two pose-to-map attention blocks (Vaswani et al.

; Shazeer, N. p. 8
Where the paper says it

Vaswani, A. ; Shazeer, N.

; Parmar, N. p. 8
Where the paper says it

; Shazeer, N. ; Parmar, N.

; Uszkoreit, J. p. 8
Where the paper says it

; Parmar, N. ; Uszkoreit, J.

; Jones, L. p. 8
Where the paper says it

; Uszkoreit, J. ; Jones, L.

; Gomez, A. p. 8
Where the paper says it

; Jones, L. ; Gomez, A.

N.; Kaiser, L. p. 8
Where the paper says it

; Gomez, A. ; Kaiser, L.

; and Polosukhin, I. p. 8
Where the paper says it

; Kaiser, L. ; and Polosukhin, I.

2017. Attention Is All You Need. p. 8
Where the paper says it

; and Polosukhin, I. Attention Is All You Need.

arXiv.

Walch, F.; Hazirbas, C. p. 8

Where the paper says it

Attention Is All You Need. ; Hazirbas, C.

; Leal-Taixe, L. p. 8
Where the paper says it

; Hazirbas, C. ; Leal-Taixe, L.

; Sattler, T. p. 0
Where the paper says it

2024) has emerged as a promising and scalable paradigm for indoor localization by aligning egocentric RGB observations with lightweight 2D floorplans, bypassing the need for storageintensive 3D reconstructions (Liu, Li, and Dai 2017; Sarlin et al. 2019; Sattler, Leibe, and Kobbelt 2017) or large-scale image databases (Balntas, Li, and Prisacariu 2018; Arandjelović et al.

; Hilsenbeck, S. p. 8
Where the paper says it

; Sattler, T. ; Hilsenbeck, S.

; and Cremers, D. p. 8
Where the paper says it

; Hilsenbeck, S. ; and Cremers, D.

2017. Image-based localization using lstms for structured feature correlation. p. 8
Where the paper says it

; and Cremers, D. Image-based localization using lstms for structured feature correlation.

In Proceedings of the IEEE international conference on computer vision, 627–637. p. 8
Where the paper says it

Image-based localization using lstms for structured feature correlation. In Proceedings of the IEEE international conference on computer vision, 627–637.

Wüest, M.; Engelmann, F. p. 8

Where the paper says it

In Proceedings of the IEEE international conference on computer vision, 627–637. ; Engelmann, F.

; Miksik, O. p. 8
Where the paper says it

; Engelmann, F. ; Miksik, O.

; Pollefeys, M. p. 7
Where the paper says it

; Vogel, C. ; and Pollefeys, M.

; and Barath, D. p. 8
Where the paper says it

; Pollefeys, M. ; and Barath, D.

2026. UnLoc: Leveraging Depth Uncertainties for Floorplan Localization. p. 8
Where the paper says it

; and Barath, D. UnLoc: Leveraging Depth Uncertainties for Floorplan Localization.

In International Conference on Learning Representations (ICLR). p. 8
Where the paper says it

UnLoc: Leveraging Depth Uncertainties for Floorplan Localization. In International Conference on Learning Representations (ICLR).

Yang, L.; Kang, B.; Huang, Z. p. 0

Where the paper says it

Compared with dense 3D models, floorplans are lightweight, easy to maintain, and readily available in built environments, making visual FLoc particularly appealing for autonomous robot navigation (Li et al. 2025; Huang et al.

; Zhao, Z.; Xu, X.; Feng, J.; and Zhao, H. p. 8
Where the paper says it

; Huang, Z. ; and Zhao, H.

2024. Depth Anything V2. p. 5
Where the paper says it

Our framework is implemented in PyTorch and trained on a single NVIDIA RTX 5880 Ada GPU. The visual observation encoder utilizes the frozen ViTS backbone from Depth Anything V2 (Yang et al.

ArXiv, abs/2406. p. 8
Where the paper says it

Depth Anything V2. ArXiv, abs/2406.

09414.

Ye, K.; Chen, M.; Zhang, X. p. 8

Where the paper says it

ArXiv, abs/2406. ; Zhang, X.

; and Yang, X. p. 0
Where the paper says it

From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching Shiyong Meng, Bolei Chen∗ , Ping Zhong∗ , Yang Wan, Rongzhi Wang, Jiazhi Xia, Jianxin Wang School of Computer Science and Engineering, Central South University {xiaowugui1017, boleichen, ping. zhong, 8205230820, 7803240121, xiajiazhi}@csu.

2026. Fusion of Depth and Semantics for Probabilistic Floorplan Localization. p. 8
Where the paper says it

; and Yang, X. Fusion of Depth and Semantics for Probabilistic Floorplan Localization.

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 19413–19422. p. 8
Where the paper says it

Fusion of Depth and Semantics for Probabilistic Floorplan Localization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 19413– 19422.

Zheng, J.; Zhang, J. p. 8

Where the paper says it

ArXiv, abs/2406. ; Zhang, X.

; Li, J.; Tang, R.; Gao, S.; and Zhou, Z. p. 8
Where the paper says it

; Zhang, J. ; and Zhou, Z.

2020. Structured3d: A large photo-realistic dataset for structured 3d modeling. p. 8
Where the paper says it

; and Zhou, Z. Structured3d: A large photo-realistic dataset for structured 3d modeling.

In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX 16. p. 8
Where the paper says it

Structured3d: A large photo-realistic dataset for structured 3d modeling. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX 16, 519–535.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Direct pose distribution estimation allows systems to directly model the continuous conditional pose distribution p(p I, M) instead of relying on intermediate ray representations, which bypasses the information bottleneck prior to localization.

  2. The coarse-to-fine architecture enables robust global hypothesis generation by using an image conditioned pose diffusion model to parameterize the continuous multimodal pose distribution across the entire floorplan.

  3. The local pose refiner performs precise sub-meter correction by predicting bounded sub-meter pose residuals from candidate-centered floorplan crops, eliminating structural ambiguities locally.

  4. The framework achieves real-time execution and superior localization accuracy without requiring offline map preprocessing or test-time lookup tables, unlike ray matching methods.

Abstract

Visual Floorplan Localization (FLoc) has emerged as a promising solution for indoor localization by matching egocentric images against minimalist structural maps. However, due to cross-modal information asymmetry and repetitive indoor layouts, visual FLoc is fundamentally challenged by multimodal pose distributions, where visually identical observations map to distinct, spatially separated locations. Existing ray-matching-based methods tackle this by explicitly predicting sparse geometric or semantic rays, which inherently incur information loss and demand resource-intensive preprocessing alongside exhaustive matching during inference. In this paper, we bypass the intermediate ray-matching paradigm and propose a coarse-to-fine visual FLoc framework that progresses from uncertainty to determinism. In the coarse stage, we design an image-conditioned pose diffusion model to parameterize the continuous multimodal pose distribution, effectively routing stochastically initialized pose particles toward distinct candidate modes. In the refinement stage, we propose a localized refiner that predicts bounded sub-meter pose residuals from candidate-centered floorplan crops, where structural ambiguities are largely eliminated. Our method effectively balances global multi-hypothesis tracking and local sub-meter refinement without requiring any offline map preprocessing or test-time lookup tables. Comprehensive results on the S3D (full) and ZInD benchmarks demonstrate that our approach achieves state-of-the-art accuracy and robustness.

Sources

Related papers