Pith. sign in

REVIEW 3 major objections 3 minor 50 references

SeqLoc: Beyond the Single Frame for Cross-View Geo-Localization in Feature-Sparse Scenes

T0 review · 3 major / 3 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read SeqLoc, an online test-time aggregation of pose likelihood volumes, recovers accurate poses in feature-sparse cross-view geo-localization, improving recall by over 50 percent without retraining.

desk verdict A useful benchmark and a surprisingly effective test-time fusion method; the main reservations are missing hyperparameters and unvalidated reference poses in the new benchmark. read the letter →

arxiv 2608.07835 v1 pith:DXLOEQWQ submitted 2026-08-08 cs.CV

Junwei Zheng , Yun Huang , Ruize Dai , Ruiping Liu , Yufan Chen , Kunyu Peng , Kailun Yang , Jiaming Zhang
show 3 more authors
Guangming Wang Olaf Wysocki Rainer Stiefelhagen
This is my paper · ORCID
classification cs.CV
keywords cross-viewgeo-localizationfeature-sparsescenesOpenStreetMapsequenceaggregationtest-timefusionposelikelihoodbenchmarkruralroads
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

Cross-view geo-localization against OpenStreetMap works in cities but collapses on rural roads, where a single image looks the same along the road and the pose likelihood spreads into a ridge. The paper establishes a new benchmark, CV-FSS, that isolates this failure mode, and proposes SeqLoc, an online test-time sequence aggregation that fuses pose likelihood volumes from past frames only. SeqLoc recursively maintains a log-belief volume, tempering each incoming likelihood by its entropy, mixing in a map-shaped recovery distribution, and refining the peak at sub-grid precision. On CV-FSS, SeqLoc raises position and orientation recall by over 50 percent relative to the same single-frame backbone, and it also improves localization on the urban CV-RHO benchmark. The method is training-free and backbone-agnostic, operating purely on the likelihood volumes that any OSM-based metric localizer outputs.

What carries the argument

The load-bearing object is the log-belief volume $b_t(x,\theta)$ over a grid of poses, updated recursively as $b_t = \log\mathrm{softmax}(\log p_t + \hat b_t)$ with $\hat b_t$ the odometry-warped previous belief. Entropy-Tempered Uncertainty (ETU) gives each incoming likelihood volume $p_t$ an inverse temperature $w_t = \max(1-\hat H_t, 10^{-3})$ where $\hat H_t$ is the entropy normalized by $\log M$, so diffuse frames are flattened. Map-Guided Relocalization (MGR) injects a recovery term $q(x) \propto 0.9\exp(-d(x)^2/(2\sigma^2)) + 0.1U$, with $d(x)$ the distance to the nearest OSM road, mixed in with weight $\varepsilon=0.01$ at every step. Peak-Anchored Smoothing (PAS) replaces the discrete argmax by a weighted mean of positions inside a window of radius $r=3$ around the belief peak. The recursion works because the log domain avoids underflow and the three components respectively prevent diffuse frames from flattening the belief, keep a recovery path open, and sharpen the final estimate.

What would settle it

Take a random sample of CV-FSS test panoramas, obtain independent high-precision poses such as RTK GPS or manually matched landmarks against aerial imagery, and recompute the reported recalls; if the label errors are comparable to the 1/3/5 m thresholds, the claimed 50%+ gains may not survive relabeling.

Watch

Extended reading notes

Core claim

The central claim is that the failure of OSM-based cross-view geo-localization in feature-sparse scenes is an ambiguity problem, not a depth or scale problem, and that this ambiguity can be resolved online by accumulating pose likelihoods along the route. SeqLoc maintains a log-belief volume over a discretized pose space, warping the previous belief forward by odometry and fusing each new likelihood in the log domain. Three components carry the argument: Entropy-Tempered Uncertainty exponentiates each likelihood volume by its normalized entropy so ambiguous frames barely move the belief; Map-Guided Relocalization adds a small road-shaped distribution at every step so a wrongly suppressed true pose can recover; Peak-Anchored Smoothing computes a local expectation around the belief peak for sub-grid position accuracy. The paper shows that with the RHO backbone, position recall at 5 m on CV-FSS rises from 10.6% to 75.0% and orientation recall at 3 degrees from 18.4% to 86.2%, and that the improvement holds on every region and every backbone tested.

Load-bearing premise

The evaluation rests on the Mapillary ground-truth poses, a fusion of structure-from-motion and GPS that the paper does not independently validate in these texture-poor rural scenes, so meter-level label errors would make the measured recalls unreliable.

Editorial extensions

If this is right

  • SeqLoc can be dropped onto any OSM-based metric CVGL backbone that outputs a pose likelihood volume, with no retraining and no extra learned parameters.
  • Online aggregation from past frames alone localizes every frame as it arrives, unlike offline fusion that needs future frames and scores only a reference frame.
  • On the feature-sparse CV-FSS benchmark, SeqLoc raises position and orientation recall by more than 50 percent relative to single-frame localization, and it also improves every city of the urban CV-RHO benchmark.
  • The entropy of a single-frame likelihood volume is a usable per-frame reliability signal: high-entropy frames tend to produce argmaxes far from the truth, and tempering them improves fusion.
  • SeqLoc stays robust to realistic odometry noise, with 5 m position recall dropping only from 84.0% to 77.1% from ideal to poor odometry on one CV-FSS region.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the same online log-belief recursion could be applied to any stream of pose hypotheses from other localizers, including satellite-image retrieval or place-recognition systems, wherever per-frame likelihoods are available.
  • Editorial inference: the map-shaped recovery floor points to a general design rule for recursive fusion—always keep a small, structurally informed probability floor so that a suppressed hypothesis can recover; a semantic or learned prior could replace the road-distance Gaussian in off-road settings.
  • Editorial inference: the paper's per-frame entropy analysis implies that high-entropy likelihood volumes are not just uninformative but actively harmful in naive fusion, so any sequence-localization method should gate contributions by an uncertainty measure; this is exactly what ETU does.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The paper addresses cross-view geo-localization (CVGL) in feature-sparse scenes. It introduces CV-FSS, a benchmark of sequential panoramas from five rural regions paired with OSM tiles, and proposes SeqLoc, an online test-time sequence aggregation method that recursively updates a log-belief pose volume. SeqLoc combines three components: Entropy-Tempered Uncertainty (ETU) to down-weight ambiguous frames, Map-Guided Relocalization (MGR) to inject a road-shaped recovery distribution, and Peak-Anchored Smoothing (PAS) for sub-grid pose refinement. Experiments on CV-FSS and CV-RHO with three backbones show large gains in position and orientation recall over single-frame localization, with ablations and a noisy-odometry robustness study in the supplementary material.

Significance. If the benchmark is reliable and the reported numbers are valid, SeqLoc is a useful, training-free, backbone-agnostic tool that can substantially improve OSM-based metric CVGL in rural areas, a regime where current methods are known to fail. The paper provides broad empirical support: three backbones, five rural regions, seven urban cities, ablations for all components, and a realistic odometry-noise analysis. The planned release of the benchmark and source code is a strength. However, the significance hinges on two currently unresolved issues: the validity of the Mapillary-derived ground-truth poses in exactly the feature-sparse scenes the benchmark targets, and the selection of key hyperparameters on the test sets rather than on a validation split.

major comments (3)
  1. [Ablation Studies] The ground-truth poses are taken from Mapillary's fusion of structure-from-motion and GPS, but the paper provides no validation of these labels. This concern is load-bearing because the benchmark deliberately selects rural, feature-sparse scenes, where SfM is known to be most fragile. If the ground-truth poses contain meter-level or larger errors, the reported 'collapse' of single-frame methods and the gains from SeqLoc may be artifacts of label noise rather than true localization performance. The authors should provide some validation evidence, such as comparisons against independently available GPS, manual checks on a sample of frames, or error statistics showing that pose likelihoods concentrate near the reference poses in ambiguous scenes.
  2. [Ablation Studies, Tables 6 and 7] The MGR recovery weight epsilon and the PAS window radius r are fixed by evaluating on the Overgaard test region of CV-FSS and the Berlin test city of CV-RHO, and the same test sets are then used to report the headline numbers in Tables 2-4. This is a form of test-set tuning that can inflate the reported improvements and weaken the claim that the components generalize across scenes. The paper should either select these hyperparameters on a separate validation split (e.g., held-out routes per region) or show that the conclusions are stable across the full range of epsilon and r on all regions, not just on the two used for the current choice.
  3. [Map-Guided Relocalization, Eq. (6)] The Gaussian width sigma in the MGR recovery distribution q(x) is never assigned a value. Unlike epsilon and r, which are explicitly fixed in Section 'Implementation Details', sigma is left unspecified, making the method non-reproducible. The authors must provide the value used in all experiments and, if sigma was chosen using test-set performance, disclose this and move the selection to a validation split.
minor comments (3)
  1. [Abstract and Conclusion] The phrase 'improving both position and orientation recall by over 50%' is ambiguous: it could be read as a relative improvement of 50% (which is far exceeded in some thresholds, e.g., position recall at 5 m goes from 10.6% to 75.0%, a relative gain of roughly 600%) or as at least 50 percentage points. Please state the intended interpretation explicitly.
  2. [Introduction, Figure 3] The caption states 'Single-frame localization degrades drastically from the urban CV-RHO to the proposed feature-sparse CV-FSS' but the left panel compares different backbones' absolute recall on different datasets with different route lengths; it would help to add a note that the comparison is qualitative.
  3. [Methodology, Eq. (3)] The normalized entropy is defined as H(p_t)/log M, which is correct for a discrete distribution, but the text says it 'lies in [0,1]'; this holds only when using natural logarithms, which is presumably the case but should be stated.

Circularity Check

1 steps flagged · score 3.0 of 10

Test-set hyperparameter selection slightly inflates the reported recall, but the aggregation recursion is self-contained and the central claim survives all swept parameter values; no load-bearing self-citation.

  1. fitted input called prediction [Experiments → Ablation Studies, Table 6 and surrounding text ('Effect of the recovery weight ε of MGR with RHO on the Overgaard region of CV-FSS'); Implementation Details.]
    "Table 6 studies the recovery weight ε that controls how much map-shaped mass MGR injects into the belief at each step. ... Increasing ε from 0.01 to 0.05 lowers recall on all six metrics ... Since ε = 0.01 gives the best result throughout, we fix it for all scenes and backbones. ... We fix the MGR weight ε = 0.01 and the PAS radius r = 3 for all scenes and backbones."

    The MGR mixing weight ε is swept on the Overgaard test region of CV-FSS, and the value that maximizes that test region's recall is then fixed and used to produce the reported numbers in Tables 2-3. The Overgaard row, and the CV-FSS average that contains it, is therefore by construction the best value observed in the sweep rather than a held-out result. The distortion is bounded: the sweep moves 5 m recall only from 75.6% to 73.5%, and the headline 'over 50%' relative improvement holds at every swept ε, so the central claim is not statistically forced. Notably, the PAS radius r (Table 7) was deliberately fixed below the test-observed optimum ('conservative r = 3'), i.e., that choice is not in-sample maximization.

full rationale

SeqLoc's derivation chain is a deterministic recursion over per-frame pose likelihood volumes: Eq. (1) warps the prior by odometry, Eq. (2) fuses the tempered likelihood in log space (standard Bayesian filtering, explicitly credited to MCL), Eqs. (3)-(5) define ETU from the likelihood's own entropy, Eqs. (6)-(7) mix a map-shaped recovery prior, and Eq. (8) is a windowed expectation for sub-grid refinement. Ground-truth poses enter only at evaluation; no equation defines the output in terms of the target, so the recursion is not self-definitional. The benchmark construction is likewise not circular: the single-frame collapse on CV-FSS is measured for three backbones, two of them external (OrienterNet, OSMLoc), so the failure mode is not an artifact of the authors' own RHO backbone, and the self-cited CV-RHO numbers were re-run here rather than quoted. The one quotable by-construction element is the MGR weight ε: Table 6 sweeps ε on the Overgaard test region and the in-sample best value is then fixed and reported, making that region's row the sweep maximum by construction. The magnitude is small, and the central claim survives every swept value, so this is a mild in-sample selection rather than a forced prediction. No uniqueness theorem is imported, no ansatz is adopted by self-citation, and the recursive-fusion lineage is credited to independent prior work, so no self-citation is load-bearing. Non-circularity caveats that belong under correctness risk: (1) CV-FSS ground-truth poses come from Mapillary's unvalidated SfM+GPS fusion, which is fragile in exactly the texture-poor rural corridors the benchmark targets, so meter-level label error could manufacture both the measured single-frame collapse and part of SeqLoc's gain; (2) the main-paper relative motion Δt is taken from the reference poses (ideal odometry), as disclosed in the supplementary, with the synthetic-noise robustness test showing only graceful degradation; (3) the MGR Gaussian width σ (Eq. 6) is never given a value, a reproducibility gap rather than circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

SeqLoc introduces no new physical entities; it is an algorithm composed of hand-chosen constants and standard filtering assumptions. The free parameters and domain assumptions listed above are the main items the central claim rests on.

free parameters (6)
  • sigma (MGR Gaussian width) = not reported
    Width of the road-distance Gaussian in Eq. (6); required to compute the map-shaped recovery distribution and never given a value.
  • epsilon (MGR recovery weight) = 0.01
    Mixing weight in Eq. (7); fixed after ablation on Overgaard (Table 6).
  • r (PAS window radius) = 3
    Window radius in Eq. (8); fixed after ablation on Overgaard and Berlin (Table 7).
  • q mixture constant (0.9/0.1) = 0.9 road term, 0.1 uniform
    Hand-chosen balance in Eq. (6) between the map-shaped term and a uniform floor; not ablated.
  • entropy floor (10^-3) = 10^-3
    Lower bound on the ETU weight in Eq. (4) to keep all likelihoods strictly positive; chosen ad hoc.
  • Pose grid discretization = 0.5 m cells, 256 orientation bins
    Discretization used for all likelihood volumes; a design choice that affects both the belief update and the meaning of sub-grid refinement.
assumptions (5)
  • domain assumption Per-frame pose likelihoods are conditionally independent given the pose.
    Eq. (2) adds log-likelihoods, which is only a Bayesian update under conditional independence; adjacent panoramas are heavily correlated, so this is violated in practice but accepted as an approximation.
  • domain assumption Relative motion delta_t between frames is available and accurate.
    Eq. (1) warps the belief via T_delta_t; the headline experiments take delta_t from dataset reference poses (ideal odometry), and the supplementary tests noise afterward.
  • domain assumption Mapillary SfM/GPS fused ground truth is accurate in feature-sparse rural scenes.
    The CV-FSS benchmark and all evaluations rely on these poses; the paper does not validate them, and SfM is known to be fragile in texture-poor environments.
  • domain assumption The vehicle always travels on the OSM road network.
    MGR shapes the recovery distribution using distance to the nearest road; this is true for the benchmark routes but restricts general applicability to on-road localization.
  • domain assumption The backbone's likelihood entropy reflects true localization uncertainty.
    ETU uses normalized entropy to down-weight frames; if a backbone is overconfident or poorly calibrated, the weighting may not track true uncertainty, though the paper shows a correlation in Figure 15.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SeqLoc: Beyond the Single Frame for Cross-View Geo-Localization in Feature-Sparse Scenes." pith.science (2026). https://pith.science/paper/DXLOEQWQ

@misc{pith2026260807835,
  author       = {Pith},
  title        = {Pith review of: SeqLoc: Beyond the Single Frame for Cross-View Geo-Localization in Feature-Sparse Scenes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DXLOEQWQ}},
  note         = {Machine review of arXiv:2608.07835}
}
read the original abstract

Cross-View Geo-Localization (CVGL) with OpenStreetMap (OSM) performs well in structure-rich urban environments but collapses in feature-sparse scenes such as rural roads. To study this failure mode, in this work, we introduce CV-FSS, a benchmark that pairs sequential panoramas from five rural regions with aligned OSM maps, on which single-frame methods degrade drastically. We then propose SeqLoc, an online test-time sequence aggregation mechanism that recursively maintains a log-belief volume with three key components: (1) Entropy-Tempered Uncertainty (ETU) tempers each incoming pose likelihood volume by its normalized entropy; (2) Map-Guided Relocalization (MGR) mixes a map-shaped recovery distribution into the belief so that a suppressed true pose can recover; (3) Peak-Anchored Smoothing (PAS) derives the final pose at sub-grid precision. Extensive experiments on CV-FSS and CV-RHO demonstrate that SeqLoc outperforms single-frame localization by a large margin, improving both position and orientation recall by over 50%. The benchmark and source code will be made publicly available.

Figures

Figures reproduced from arXiv: 2608.07835 by the authors.

Figure 1
Figure 1. Single-frame OSM-based Cross-View Geo-Localization is unreliable in feature-sparse scenes. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Data samples from CV-FSS. The three views are [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 4
Figure 4. Overview of the five regions in CV-FSS. On the OSM of each region, training and test routes are marked in blue and [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figures from the paper (9 more)
Figure 5
Figure 5. Figure 5: Offline vs. online test-time sequence aggregation. [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]
Figure 7
Figure 7. Figure 7: MGR recovers a suppressed true pose. Curves are [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]
Figure 8
Figure 8. Figure 8: PAS refines the pose to sub-grid precision. [PITH_FULL_IMAGE:figures/full_fig_p005_8.png]
Figure 9
Figure 9. Figure 9: Qualitative results on the five regions of CV-FSS. [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]
Figure 11
Figure 11. Figure 11: Per-region entropy analysis on the Tirol region of [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]
Figure 12
Figure 12. Figure 12: Per-region entropy analysis on the Bad Herrenalb [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]
Figure 13
Figure 13. Figure 13: Per-region entropy analysis on the Speicher region [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]
Figure 15
Figure 15. Figure 15: Single-frame localization error against the normalized entropy [PITH_FULL_IMAGE:figures/full_fig_p013_15.png]
Figure 16
Figure 16. Figure 16: More data samples from the five regions of CV-FSS. From top to bottom, the rows correspond to the Neuanspach, [PITH_FULL_IMAGE:figures/full_fig_p013_16.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 40 canonical work pages

  1. [1]

    IEEE Transactions on intelligent vehicles , volume=

    A survey on map-based localization techniques for autonomous vehicles , author=. IEEE Transactions on intelligent vehicles , volume=. 2022 , publisher=

    work page 2022

  2. [2]

    Computer Science Review , volume=

    Mobile robot localization: Current challenges and future prospective , author=. Computer Science Review , volume=. 2024 , publisher=

    work page 2024

  3. [3]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Scene-agnostic pose regression for visual localization , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  4. [4]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Orienternet: Visual localization in 2d public maps with neural matching , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  5. [5]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    RHO: Robust Holistic OSM-Based Metric Cross-View Geo-Localization , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  6. [6]

    Artificial intelligence , volume=

    Robust Monte Carlo localization for mobile robots , author=. Artificial intelligence , volume=. 2001 , publisher=

    work page 2001

  7. [7]

    Computer , volume=

    Google street view: Capturing the world at street level , author=. Computer , volume=. 2010 , publisher=

    2010

  8. [8]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    Cvm-net: Cross-view matching network for image-based ground-to-aerial geo-localization , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    work page

Show all 50 references
  1. [9]

    CVPR , year=

    TransGeo: Transformer Is All You Need for Cross-view Image Geo-localization , author=. CVPR , year=

  2. [10]

    ICCV , year=

    Sample4Geo: Hard Negative Sampling For Cross-View Geo-Localisation , author=. ICCV , year=

  3. [11]

    ECCV , year=

    Statewide visual geolocalization in the wild , author=. ECCV , year=

  4. [12]

    CVPR , year=

    Where am I? Cross-view geo-localization with natural language descriptions , author=. CVPR , year=

  5. [13]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Lending orientation to neural networks for cross-view geo-localization , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  6. [14]

    Advances in Neural Information Processing Systems , volume=

    Spatial-aware feature aggregation for image based cross-view geo-localization , author=. Advances in Neural Information Processing Systems , volume=

  7. [15]

    Advances in Neural Information Processing Systems , volume=

    Cross-view geo-localization with layer-to-layer transformer , author=. Advances in Neural Information Processing Systems , volume=

  8. [16]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Vigor: Cross-view image geo-localization beyond one-to-one retrieval , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  9. [17]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Beyond cross-view image retrieval: Highly accurate vehicle localization using satellite image , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  10. [18]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Slicematch: Geometry-guided aggregation for cross-view pose estimation , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  11. [19]

    Advances in Neural Information Processing Systems , volume=

    Fine-grained cross-view geo-localization using a correlation-aware homography estimator , author=. Advances in Neural Information Processing Systems , volume=

  12. [20]

    IEEE Transactions on Pattern Analysis and Machine Intelligence , volume=

    Convolutional cross-view pose estimation , author=. IEEE Transactions on Pattern Analysis and Machine Intelligence , volume=. 2023 , publisher=

    2023

  13. [21]

    Xia, Zimin and Alahi, Alexandre , booktitle=. FG\^

  14. [22]

    European Conference on Computer Vision , pages=

    Visual cross-view metric localization with dense uncertainty estimates , author=. European Conference on Computer Vision , pages=. 2022 , organization=

    2022

  15. [23]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Uncertainty-aware vision-based metric cross-view geolocalization , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  16. [24]

    2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages=

    Maplocnet: Coarse-to-fine feature registration for visual re-localization in navigation maps , author=. 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages=. 2024 , organization=

    2024

  17. [25]

    Information Fusion , pages=

    OSMLoc: Single image-based visual localization in OpenStreetMap with fused geometric and semantic guidance , author=. Information Fusion , pages=. 2026 , publisher=

    2026

  18. [26]

    European Conference on Computer Vision , pages=

    Gama: Cross-view video geo-localization , author=. European Conference on Computer Vision , pages=. 2022 , organization=

    2022

  19. [27]

    Asian Conference on Computer Vision , pages=

    Cvlnet: Cross-view semantic correspondence learning for video-based camera localization , author=. Asian Conference on Computer Vision , pages=. 2022 , organization=

    2022

  20. [28]

    Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages=

    Cross-view image sequence geo-localization , author=. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages=

  21. [29]

    Proceedings 1999 IEEE international conference on robotics and automation (Cat

    Monte carlo localization for mobile robots , author=. Proceedings 1999 IEEE international conference on robotics and automation (Cat. No. 99CH36288C) , volume=. 1999 , organization=

    1999

  22. [30]

    2012 IEEE international conference on robotics and automation , pages=

    SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights , author=. 2012 IEEE international conference on robotics and automation , pages=. 2012 , organization=

    2012

  23. [31]

    Communications of the ACM , volume=

    Probabilistic robotics , author=. Communications of the ACM , volume=. 2002 , publisher=

    2002

  24. [32]

    arXiv preprint arXiv:1610.06475 , year=

    ORB-SLAM2: an open-source SLAM system for monocular, stereo and RGB-D cameras , author=. arXiv preprint arXiv:1610.06475 , year=

    arXiv

  25. [33]

    IEEE transactions on robotics , volume=

    Vins-mono: A robust and versatile monocular visual-inertial state estimator , author=. IEEE transactions on robotics , volume=. 2018 , publisher=

    2018

  26. [34]

    , author=

    LOAM: Lidar odometry and mapping in real-time. , author=. Robotics: Science and systems , pages=. 2014 , organization=

    2014

  27. [35]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    Deep residual learning for image recognition , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

  28. [36]

    3rd international conference on learning representations (ICLR 2015) , year=

    Very deep convolutional networks for large-scale image recognition , author=. 3rd international conference on learning representations (ICLR 2015) , year=

    2015

  29. [37]

    IEEE Transactions on robotics , volume=

    Past, present, and future of simultaneous localization and mapping: Toward the robust-perception age , author=. IEEE Transactions on robotics , volume=. 2016 , publisher=

    2016

  30. [38]

    arXiv preprint arXiv:2412.18852 , year=

    Cross-view image set geo-localization , author=. arXiv preprint arXiv:2412.18852 , year=

    arXiv

  31. [39]

    arXiv preprint arXiv:2607.15491 , year=

    Trajectory-aware Cross-view Geo-localization with Sequential Observations , author=. arXiv preprint arXiv:2607.15491 , year=

    arXiv

  32. [40]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Video geo-localization employing geo-temporal feature learning and gps trajectory smoothing , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  33. [41]

    European Conference on Computer Vision , pages=

    Garet: cross-view video geolocalization with adapters and auto-regressive transformers , author=. European Conference on Computer Vision , pages=. 2024 , organization=

    2024

  34. [42]

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , volume=

    Cv-cities: Advancing cross-view geo-localization in global cities , author=. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , volume=. 2024 , publisher=

    2024

  35. [43]

    2012 IEEE conference on computer vision and pattern recognition , pages=

    Are we ready for autonomous driving? the kitti vision benchmark suite , author=. 2012 IEEE conference on computer vision and pattern recognition , pages=. 2012 , organization=

    2012

  36. [44]

    2023 IEEE International Conference on Robotics and Biomimetics (ROBIO) , pages=

    Tightly-coupled LiDAR-visual SLAM based on geometric features for mobile agents , author=. 2023 IEEE International Conference on Robotics and Biomimetics (ROBIO) , pages=. 2023 , organization=

    2023

  37. [45]

    2009 IEEE conference on computer vision and pattern recognition , pages=

    Imagenet: A large-scale hierarchical image database , author=. 2009 IEEE conference on computer vision and pattern recognition , pages=. 2009 , organization=

    2009

  38. [46]

    Asian Conference on Computer Vision , pages=

    OneBEV: Using one panoramic image for bird’s-eye-view semantic mapping , author=. Asian Conference on Computer Vision , pages=. 2024 , organization=

    2024

  39. [47]

    IEEE Transactions on Pattern Analysis and Machine Intelligence , year=

    OneBEV++: Towards Unifying Bird's-Eye-View Semantic Mapping with Panoramas , author=. IEEE Transactions on Pattern Analysis and Machine Intelligence , year=

  40. [48]

    European Conference on Computer Vision , pages=

    Open panoramic segmentation , author=. European Conference on Computer Vision , pages=. 2024 , organization=

    2024

  41. [49]

    arXiv preprint arXiv:2411.16481 , year=

    Deformable mamba for wide field of view segmentation , author=. arXiv preprint arXiv:2411.16481 , year=

    arXiv

  42. [50]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    More than the Sum: Panorama-Language Models for Adverse Omni-Scenes , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.