pith. sign in

arxiv: 2606.02379 · v2 · pith:BF6NBG5Rnew · submitted 2026-06-01 · 💻 cs.CV

Honey, I Shrunk the Arc de Triomphe!

Yuanbo Xiangli , Hanyu Chen , Xueqing Tsang , Noah Snavely This is my paper

Pith reviewed 2026-06-29 05:27 UTC · model grok-4.3

classification 💻 cs.CV
keywords metric scale estimationmonocular depth estimationscale collapsein-the-wild datasetMetricScenesdepth completiongeo-tagged imagesstereo imagery
0
0 comments X

The pith

A new in-the-wild dataset with scales from geo-tags and stereo baselines lets fine-tuning fix metric underestimation in monocular depth models.

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

Foundation models for metric monocular geometry estimation underestimate scales of distant landmarks and large scenes, a problem the authors trace to limited training data that lacks real-world diversity. Existing datasets rely on vehicle LiDAR, short indoor scans, or synthetic scenes without complex semantics. The paper introduces MetricScenes, assembled from internet photo collections and stereo imagery, where absolute scales come from geo-tagged metadata and known camera baselines, with depth maps refined by a two-stage Poisson completion process. Fine-tuning MoGe-2 on this dataset reduces scale-collapse in unconstrained scenes and improves metric accuracy while holding state-of-the-art results on standard benchmarks.

Core claim

The paper claims that scale-collapse arises from a training data bottleneck in current metric monocular models, and that curating MetricScenes from diverse web sources with absolute scales recovered from geo-tags and stereo baselines, together with Poisson-refined depths, supplies the missing signal; fine-tuning on it then delivers accurate metric scales for open-domain scenes without sacrificing benchmark performance.

What carries the argument

The MetricScenes dataset, which supplies metrically grounded depth maps for unconstrained scenes via scale recovery from geo-tagged metadata and stereo baselines.

If this is right

  • Fine-tuned models recover accurate metric scales for distant objects where prior models collapse.
  • Metric accuracy improves in unconstrained open-domain scenes.
  • Performance stays at state-of-the-art levels on existing benchmarks.
  • The two-stage Poisson completion produces higher-quality depth maps from the new data.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Data curation focused on metric grounding may matter more than further model scaling for scale-sensitive geometry tasks.
  • The approach could transfer to other scale-dependent applications such as outdoor augmented reality or large-scale mapping.
  • Combining MetricScenes with existing datasets might yield even stronger scale recovery across mixed environments.

Load-bearing premise

Absolute scale recovered from geo-tagged metadata and known stereo baselines is accurate and free of systematic bias for the collected in-the-wild scenes, and off-the-shelf pose and depth estimators produce reliable initial maps.

What would settle it

Direct comparison of metric scale error on large real-world distances, such as known landmark separations in test photos, before and after fine-tuning would show no reduction if the central claim is false.

Figures

Figures reproduced from arXiv: 2606.02379 by Hanyu Chen, Noah Snavely, Xueqing Tsang, Yuanbo Xiangli.

Figure 1
Figure 1. Figure 1: Scale-collapse in metric geometry estimation. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Metric depth from Internet photo collections. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Metric Depth from Stereo4D [11]. Top: Standard stereo matching [36, 37] often produces distorted geometry in poorly calibrated in-the-wild videos, as seen in the converging facades (magenta boxes). Among multi-view models [13, 20, 35], \pi ^3 [35] maintains the most robust geometry and sharp local details (cyan boxes). Bottom: We process stereoscopic sequences via \pi ^3 to obtain dense geometry and poses,… view at source ↗
Figure 4
Figure 4. Figure 4: Visual comparison of depth completion methods. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Overview of the two-stage depth completion pipeline. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Metrology of novel in-the-wild scenes. The first column shows images with measurements obtained via Google Map’s measuring tool. We merge WildMoGe and MoGe-2’s results into a single column to highlight the accurate scaling achieved by our training scheme. WildMoGe consistently recovers more accurate absolute scales across diverse landmarks, whereas MoGe-2 [33], DepthAnything v3 [20] and Metric3D v2 [10] ex… view at source ↗
Figure 7
Figure 7. Figure 7: Comparison on the standard scenes. We compare WildMoGe against MoGe-2 [33] on representative indoor and street-level scenes. In standard indoor and street contexts (Rows 1 & 2), WildMoGe provides scale estimates consistent with MoGe-2. On the ETH3D [27] courtyard scene (Row 3), WildMoGe achieves better accuracy, recovering a desk leg height of 71.6cm compared to the 72cm ground truth. This implies that Wil… view at source ↗
read the original abstract

Metric scale monocular geometry estimation has seen significant progress through large-scale data aggregation, yet current foundation models suffer from a persistent ''scale-collapse'' phenomenon: distant landmarks and vast landscapes are metrically underestimated. We hypothesize that this performance gap stems from a training data bottleneck, where existing metric-scale datasets are hardware-constrained to homogenous vehicle-captured LiDAR or short-range indoor scans, or consist of synthetic data that lacks the semantic complexity of the physical world. To bridge this gap, we curate a new metrically-grounded, in-the-wild dataset that we call MetricScenes, gathered from a variety of sources including Internet photo collections and stereo imagery. We estimate camera poses and initial depth maps for each scene using off-the-shelf methods, and recover absolute scale from geo-tagged metadata as well as known stereo camera baselines. We also improve the quality of depth maps derived from MetricScenes via a new two-stage Poisson completion method. Fine-tuning MoGe-2 on our dataset significantly mitigates scale-collapse and achieves superior metric accuracy in unconstrained, open-domain scenes while maintaining state-of-the-art performance on standard benchmarks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper claims that existing metric-scale monocular geometry models suffer from scale-collapse on distant objects due to training-data limitations, introduces the MetricScenes dataset curated from internet photos and stereo imagery with absolute scale recovered from geo-tags and known baselines plus a two-stage Poisson depth completion, and reports that fine-tuning MoGe-2 on this data mitigates scale-collapse while achieving superior metric accuracy in open-domain scenes and retaining SOTA on standard benchmarks.

Significance. If the recovered metric labels prove reliable, the work would be significant for addressing a persistent failure mode in foundation models for unconstrained scenes; the use of diverse in-the-wild sources and the Poisson completion step represent a concrete step beyond hardware-limited or synthetic datasets.

major comments (2)
  1. [MetricScenes construction and scale-recovery procedure] The central performance claims rest on the accuracy of absolute-scale labels recovered from geo-tagged metadata and stereo baselines after off-the-shelf pose/depth estimation. No quantitative error analysis, bias quantification, or cross-validation against independent references is supplied for these labels, despite known tens-of-meters errors in consumer geo-tags and degradation of off-the-shelf estimators in unconstrained scenes. This directly affects whether fine-tuning truly recovers metric geometry or merely fits the model's scale-collapse metric to the dataset's own error distribution.
  2. [Abstract and experimental claims] The abstract asserts 'superior metric accuracy' and 'significantly mitigates scale-collapse' yet supplies no numerical results, error bars, validation protocol, or comparison tables; without these the reader cannot assess whether the data actually supports the claims.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. We address each major comment below and will revise the manuscript accordingly to strengthen the presentation of our contributions.

read point-by-point responses

  1. Referee: [MetricScenes construction and scale-recovery procedure] The central performance claims rest on the accuracy of absolute-scale labels recovered from geo-tagged metadata and stereo baselines after off-the-shelf pose/depth estimation. No quantitative error analysis, bias quantification, or cross-validation against independent references is supplied for these labels, despite known tens-of-meters errors in consumer geo-tags and degradation of off-the-shelf estimators in unconstrained scenes. This directly affects whether fine-tuning truly recovers metric geometry or merely fits the model's scale-collapse metric to the dataset's own error distribution.

    Authors: We agree that a quantitative validation of the recovered scales is essential to substantiate the claims. The current manuscript describes the scale-recovery procedure but does not include explicit error metrics or cross-validation. In the revised version we will add a dedicated analysis section that quantifies scale-recovery error on subsets with independent references (e.g., scenes overlapping with accurate geo-tagged benchmarks or stereo baselines with known ground-truth distances), reports bias statistics, and discusses the impact of geo-tag noise. This addition will directly address whether the fine-tuning recovers true metric geometry rather than dataset-specific error patterns. revision: yes

  2. Referee: [Abstract and experimental claims] The abstract asserts 'superior metric accuracy' and 'significantly mitigates scale-collapse' yet supplies no numerical results, error bars, validation protocol, or comparison tables; without these the reader cannot assess whether the data actually supports the claims.

    Authors: The abstract is intended as a concise summary; the detailed numerical results, error bars, validation protocols, and comparison tables appear in the experimental section of the manuscript. To improve readability we will revise the abstract to incorporate the key quantitative outcomes (e.g., specific reductions in scale-collapse error on open-domain scenes and benchmark retention figures) while preserving its brevity. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical dataset curation and fine-tuning

full rationale

The paper's central claim is an empirical result obtained by curating MetricScenes from external geo-tagged metadata and stereo baselines, running off-the-shelf pose/depth estimators, applying Poisson completion, and then fine-tuning MoGe-2. No equation, parameter fit, or self-citation is shown to reduce the reported mitigation of scale-collapse to a quantity defined inside the paper itself; the scale labels are presented as inputs derived from independent sources rather than fitted or renamed outputs. The derivation chain therefore remains self-contained against external benchmarks and does not match any of the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim depends on the accuracy of scale recovery from metadata and the reliability of initial depth maps produced by external tools; these are domain assumptions rather than derived quantities.

axioms (2)
  • domain assumption Off-the-shelf methods for camera pose estimation and initial depth maps are sufficiently accurate for the collected in-the-wild scenes.
    Invoked when constructing the dataset from internet photos and stereo imagery.
  • domain assumption Geo-tagged metadata and known stereo baselines supply unbiased absolute metric scale.
    Used to assign real-world scale to the scenes.

pith-pipeline@v0.9.1-grok · 5728 in / 1230 out tokens · 32051 ms · 2026-06-29T05:27:11.708620+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

45 extracted references · 12 canonical work pages · 9 internal anchors

  1. [1]

    ARKitScenes: A Diverse Real-World Dataset For 3D Indoor Scene Understanding Using Mobile RGB-D Data

    Baruch, G., Chen, Z., Dehghan, A., Dimry, T., Feigin, Y., Fu, P., Gebauer, T., Joffe, B., Kurz, D., Schwartz, A., Shulman, E.: ARKitScenes: A diverse real- world dataset for 3d indoor scene understanding using mobile rgb-d data. ArXiv abs/2111.08897(2021)

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  2. [2]

    ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth

    Bhat,S.F.,Birkl,R.,Wofk,D.,Wonka,P.,Müller,M.:Zoedepth:Zero-shottransfer by combining relative and metric depth. arXiv preprint arXiv:2302.12288 (2023)

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  3. [3]

    Depth Pro: Sharp Monocular Metric Depth in Less Than a Second

    Bochkovskii, A., Delaunoy, A., Germain, H., Santos, M., Zhou, Y., Richter, S.R., Koltun, V.: Depth pro: Sharp monocular metric depth in less than a second. arXiv (2024),https://arxiv.org/abs/2410.02073

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  4. [4]

    Butler, D.J., Wulff, J., Stanley, G.B., Black, M.J.: A naturalistic open source movie for optical flow evaluation. In: A. Fitzgibbon et al. (Eds.) (ed.) European Conf. on Computer Vision (ECCV). pp. 611–625. Part IV, LNCS 7577, Springer-Verlag (Oct 2012)

    2012

  5. [5]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Cai, R., Tung, J., Wang, Q., Averbuch-Elor, H., Hariharan, B., Snavely, N.: Dop- pelgangers: Learning to disambiguate images of similar structures. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 34–44 (2023)

    2023

  6. [6]

    Downs, L., Francis, A., Koenig, N., Kinman, B., Hickman, R., Reymann, K., McHugh, T.B., Vanhoucke, V.: Google scanned objects: A high-quality dataset of 3d scanned household items (2022)

    2022

  7. [7]

    2025 International Conference on 3D Vision (3DV) pp

    Duisterhof, B.P., Žust, L., Weinzaepfel, P., Leroy, V., Cabon, Y., Revaud, J.: Mast3r-sfm: A fully-integrated solution for unconstrained structure-from-motion. 2025 International Conference on 3D Vision (3DV) pp. 1–10 (2024),https: //api.semanticscholar.org/CorpusID:272988049

    2025

  8. [8]

    2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) pp

    Fonder, M., Droogenbroeck, M.V.: Mid-air: A multi-modal dataset for extremely low altitude drone flights. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) pp. 553–562 (2019),https://api. semanticscholar.org/CorpusID:156052231

    2019

  9. [9]

    In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2020)

    Guizilini, V., Ambrus, R., Pillai, S., Raventos, A., Gaidon, A.: 3d packing for self- supervised monocular depth estimation. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2020)

    2020

  10. [10]

    IEEE Transactions on Pattern Analysis and Machine Intelligence46(12), 10579–10596 (2024)

    Hu, M., Yin, W., Zhang, C., Cai, Z., Long, X., Chen, H., Wang, K., Yu, G., Shen, C., Shen, S.: Metric3d v2: A versatile monocular geometric foundation model for zero-shot metric depth and surface normal estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence46(12), 10579–10596 (2024)

    2024

  11. [11]

    In: CVPR (2025)

    Jin,L.,Tucker,R.,Li,Z.,Fouhey,D.,Snavely,N.,Holynski,A.:Stereo4D:Learning How Things Move in 3D from Internet Stereo Videos. In: CVPR (2025)

    2025

  12. [12]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Jung, H., Ruhkamp, P., Zhai, G., Brasch, N., Li, Y., Verdie, Y., Song, J., Zhou, Y., Armagan, A., Ilic, S., et al.: On the importance of accurate geometry data for dense 3d vision tasks. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 780–791 (2023) 16 Y. Xiangli et al

    2023

  13. [13]

    In: International Con- ference on 3D Vision (3DV)

    Keetha, N., Müller, N., Schönberger, J., Porzi, L., Zhang, Y., Fischer, T., Knapitsch, A., Zauss, D., Weber, E., Antunes, N., Luiten, J., Lopez-Antequera, M., Bulò, S.R., Richardt, C., Ramanan, D., Scherer, S., Kontschieder, P.: MapA- nything: Universal feed-forward metric 3D reconstruction. In: International Con- ference on 3D Vision (3DV). IEEE (2026)

    2026

  14. [14]

    In: Leal-Taixé, L., Roth, S

    Koch, T., Liebel, L., Fraundorfer, F., Körner, M.: Evaluation of cnn-based single- image depth estimation methods. In: Leal-Taixé, L., Roth, S. (eds.) Proceedings of the European Conference on Computer Vision Workshops (ECCV-WS). pp. 331–348. Springer International Publishing (2019)

    2019

  15. [15]

    Computer Vision and Image Understanding (CVIU)191, 102877 (2020).https: //doi.org/10.1016/j.cviu.2019.102877

    Koch, T., Liebel, L., Körner, M., Fraundorfer, F.: Comparison of monocular depth estimation methods using geometrically relevant metrics on the ibims-1 dataset. Computer Vision and Image Understanding (CVIU)191, 102877 (2020).https: //doi.org/10.1016/j.cviu.2019.102877

    work page doi:10.1016/j.cviu.2019.102877 2020

  16. [16]

    In: European Conference on Computer Vision

    Leroy, V., Cabon, Y., Revaud, J.: Grounding image matching in 3d with mast3r. In: European Conference on Computer Vision. pp. 71–91. Springer (2024)

    2024

  17. [17]

    3182–3192 (2023), https://api.semanticscholar.org/CorpusID:263135139

    Li, Y., Jiang, L., Xu, L., Xiangli, Y., Wang, Z., Lin, D., Dai, B.: Matrixcity: A large-scalecitydatasetforcity-scaleneuralrenderingandbeyond.2023IEEE/CVF International Conference on Computer Vision (ICCV) pp. 3182–3192 (2023), https://api.semanticscholar.org/CorpusID:263135139

    2023

  18. [18]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2026)

    Li, Y., Xiangli, Y., Averbuch-Elor, H., Snavely, N., Cai, R.: Long-tail internet photo reconstruction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2026)

    2026

  19. [19]

    In: Computer Vision and Pattern Recognition (CVPR) (2018)

    Li, Z., Snavely, N.: Megadepth: Learning single-view depth prediction from internet photos. In: Computer Vision and Pattern Recognition (CVPR) (2018)

    2018

  20. [20]

    Depth Anything 3: Recovering the Visual Space from Any Views

    Lin, H., Chen, S., Liew, J.H., Chen, D.Y., Li, Z., Shi, G., Feng, J., Kang, B.: Depth anything 3: Recovering the visual space from any views. ArXivabs/2511.10647 (2025),https://api.semanticscholar.org/CorpusID:282992334

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  21. [21]

    In: Proc

    Mehl, L., Schmalfuss, J., Jahedi, A., Nalivayko, Y., Bruhn, A.: Spring: A high- resolutionhigh-detaildatasetandbenchmarkforsceneflow,opticalflowandstereo. In: Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2023)

    2023

  22. [22]

    In: ECCV (2012)

    Nathan Silberman, Derek Hoiem, P.K., Fergus, R.: Indoor segmentation and sup- port inference from rgbd images. In: ECCV (2012)

    2012

  23. [23]

    Piccinelli, L., Sakaridis, C., Yang, Y.H., Segu, M., Li, S., Abbeloos, W., Gool, L.V.: Unidepthv2: Universal monocular metric depth estimation made simpler (2025), https://arxiv.org/abs/2502.20110

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  24. [24]

    2021 IEEE/CVF International Conference on Com- puter Vision (ICCV) pp

    Roberts, M., Paczan, N.: Hypersim: A photorealistic synthetic dataset for holistic indoor scene understanding. 2021 IEEE/CVF International Conference on Com- puter Vision (ICCV) pp. 10892–10902 (2020),https://api.semanticscholar. org/CorpusID:226254406

    2021

  25. [25]

    In: Conference on Computer Vision and Pattern Recognition (CVPR) (2016)

    Schönberger, J.L., Frahm, J.M.: Structure-from-motion revisited. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2016)

    2016

  26. [26]

    In: European Conference on Computer Vision (ECCV) (2016)

    Schönberger, J.L., Zheng, E., Pollefeys, M., Frahm, J.M.: Pixelwise view selection for unstructured multi-view stereo. In: European Conference on Computer Vision (ECCV) (2016)

    2016

  27. [27]

    In: Conference on Computer Vision and Pattern Recognition (CVPR) (2019)

    Schöps, T., Sattler, T., Pollefeys, M.: BAD SLAM: Bundle adjusted direct RGB- D SLAM. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2019)

    2019

  28. [28]

    2020 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR) pp

    Sun, P., Kretzschmar, H., Dotiwalla, X., Chouard, A., Patnaik, V., Tsui, P., Guo, J., Zhou, Y., Chai, Y., Caine, B., Vasudevan, V., Han, W., Ngiam, J., Zhao, H., MetricScenes 17 Timofeev, A., Ettinger, S.M., Krivokon, M., Gao, A., Joshi, A., Zhang, Y., Shlens, J., Chen, Z., Anguelov, D.: Scalability in perception for autonomous driving: Waymo open dataset...

    2020

  29. [29]

    ArXiv abs/2406.11819(2024)

    Tung, J., Chou, G., Cai, R., Yang, G., Zhang, K., Wetzstein, G., Hariha- ran, B., Snavely, N.: Megascenes: Scene-level view synthesis at scale. ArXiv abs/2406.11819(2024)

    work page arXiv 2024

  30. [30]

    In: International Conference on 3D Vision (3DV) (2017)

    Uhrig, J., Schneider, N., Schneider, L., Franke, U., Brox, T., Geiger, A.: Sparsity invariant cnns. In: International Conference on 3D Vision (3DV) (2017)

    2017

  31. [31]

    CoRRabs/1908.00463(2019),http:// arxiv.org/abs/1908.00463

    Vasiljevic, I., Kolkin, N., Zhang, S., Luo, R., Wang, H., Dai, F.Z., Daniele, A.F., Mostajabi, M., Basart, S., Walter, M.R., Shakhnarovich, G.: DIODE: A Dense Indoor and Outdoor DEpth Dataset. CoRRabs/1908.00463(2019),http:// arxiv.org/abs/1908.00463

    work page arXiv 1908

  32. [32]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2025)

    Vuong, K., Ghosh, A., Ramanan, D., Narasimhan, S., Tulsiani, S.: Aerialmegadepth: Learning aerial-ground reconstruction and view synthesis. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2025)

    2025

  33. [33]

    MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details

    Wang, R., Xu, S., Dong, Y., Deng, Y., Xiang, J., Lv, Z., Sun, G., Tong, X., Yang, J.: Moge-2: Accurate monocular geometry with metric scale and sharp details. arXiv preprint arXiv:2507.02546 (2025),https://arxiv.org/abs/2507.02546

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  34. [34]

    2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) pp

    Wang, W., Zhu, D., Wang, X., Hu, Y., Qiu, Y., Wang, C., Hu, Y., Kapoor, A., Scherer, S.A.: Tartanair: A dataset to push the limits of visual slam. 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) pp. 4909–4916 (2020),https://api.semanticscholar.org/CorpusID:214727835

    2020

  35. [35]

    $\pi^3$: Permutation-Equivariant Visual Geometry Learning

    Wang, Y., Zhou, J., Zhu, H., Chang, W., Zhou, Y., Li, Z., Chen, J., Pang, J., Shen, C., He, T.: Pi3: Permutation-equivariant visual geometry learning. arXiv preprint arXiv:2507.13347 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  36. [36]

    In: European Conference on Computer Vision

    Wang, Y., Lipson, L., Deng, J.: Sea-raft: Simple, efficient, accurate raft for optical flow. In: European Conference on Computer Vision. pp. 36–54. Springer (2024)

    2024

  37. [37]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Wen, B., Trepte, M., Aribido, J., Kautz, J., Gallo, O., Birchfield, S.: Foundation- stereo: Zero-shot stereo matching. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 5249–5260 (2025)

    2025

  38. [38]

    Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting

    Wilson, B., Qi, W., Agarwal, T., Lambert, J., Singh, J., Khandelwal, S., Pan, B., Kumar, R., Hartnett, A., Pontes, J.K., Ramanan, D., Hays, J.: Argoverse 2: Next generation datasets for self-driving perception and forecasting. ArXiv abs/2301.00493(2023)

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  39. [39]

    Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsing

    Wrenninge, M., Unger, J.: Synscapes: A photorealistic synthetic dataset for street scene parsing. ArXivabs/1810.08705(2018),https://api.semanticscholar. org/CorpusID:53047282

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  40. [40]

    Xiangli, Y., Cai, R., Chen, H., Byrne, J., Snavely, N.: Doppelgangers++: Improved visual disambiguation with geometric 3d features (2025)

    2025

  41. [41]

    In: NeurIPS (2024)

    Yang, L., Kang, B., Huang, Z., Zhao, Z., Xu, X., Feng, J., Zhao, H.: Depth anything v2. In: NeurIPS (2024)

    2024

  42. [42]

    2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp

    Yao, Y., Luo, Z., Li, S., Zhang, J., Ren, Y., Zhou, L., Fang, T., Quan, L.: Blend- edmvs: A large-scale dataset for generalized multi-view stereo networks. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 1787–1796 (2019),https://api.semanticscholar.org/CorpusID:208248003

    2020

  43. [43]

    2023 IEEE/CVF International Conference on Computer Vi- 18 Y

    Yeshwanth, C., Liu, Y.C., Nießner, M., Dai, A.: Scannet++: A high-fidelity dataset of 3d indoor scenes. 2023 IEEE/CVF International Conference on Computer Vi- 18 Y. Xiangli et al. sion (ICCV) pp. 12–22 (2023),https://api.semanticscholar.org/CorpusID: 261064784

    2023

  44. [44]

    2023 IEEE/CVF International Conference on Computer Vision (ICCV) pp

    Yin, W., Zhang, C., Chen, H., Cai, Z., Yu, G., Wang, K., Chen, X., Shen, C.: Metric3d: Towards zero-shot metric 3d prediction from a single image. 2023 IEEE/CVF International Conference on Computer Vision (ICCV) pp. 9009–9019 (2023),https://api.semanticscholar.org/CorpusID:259991083

    2023

  45. [45]

    2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp

    Yin, W., Zhang, J., Wang, O., Niklaus, S., Mai, L., Chen, S., Shen, C.: Learning to recover 3d scene shape from a single image. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 204–213 (2020),https: //api.semanticscholar.org/CorpusID:229298063

    2021