Pith. sign in

REVIEW 3 major objections 4 minor 96 references

X-Lens establishes that a 0.04B-parameter network can estimate metric depth from mixed fisheye and pinhole views, cutting AbsRel by 25.4% over the strongest baseline on a six-view heterogeneous benchmark while running at 22 FPS.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 06:08 UTC pith:JDDQKXCN

load-bearing objection Solid engineering and a big synthetic dataset, but the heterogeneous-camera claim is built on in-distribution synthetic numbers and zero-shot baselines. the 3 major comments →

arxiv 2607.12993 v2 pith:JDDQKXCN submitted 2026-07-14 cs.CV

X-Lens: Real-Time Metric Depth Estimation with Heterogeneous Cameras

Heng Zhou , Shuhong Liu , Yonghao He , Bohao Zhang , Fa Fu , Chenhui Hou , Xianbao Hou , Lijun Han
show 1 more author
Wei Sui
This is my paper
classification cs.CV
keywords metric depth estimationheterogeneous camerasfisheyepinholemulti-view depthreal-time inferencesynthetic datasetgeometry-aware attention
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper tries to prove that real-time metric depth estimation does not require giant reconstruction-oriented models or panorama stitching: a compact feed-forward network can natively fuse calibrated fisheye and pinhole views and still beat much larger systems on heterogeneous-camera accuracy. To do this, X-Lens reasons entirely in ray space, uses learnable calibration tokens to adapt each lens's distortion, and injects a Jacobian-based bias into cross-view attention so the model matches features using local projection geometry. The authors also release OmniScene, a large synthetic six-view dataset, and report superior accuracy on it plus competitive results on conventional fisheye-only and pinhole-only benchmarks. If the central claim holds, mixed-rig perception for robots, vehicles, and embodied systems becomes feasible at a fraction of the usual compute cost.

Core claim

X-Lens is a feed-forward network that takes a variable number of calibrated fisheye and pinhole views and predicts, per view, a dense normalized depth map, a confidence map, and a single global metric scalar; multiplying the scalar by the normalized depth yields metric depth. All internal geometry is expressed in ray space through a generic unprojection map that absorbs both pinhole intrinsics and fisheye distortion, avoiding panoramic resampling. Two modules carry the heterogeneous alignment: multi-view calibration tokens, injected layer-wise only for fisheye views, provide view-local lens correction, and a Jacobian-parameterized distortion bias, added to cross-view attention logits, favors

What carries the argument

The central object is the generic unprojection map G, which maps each pixel to a unit ray while absorbing per-camera calibration and camera type; every attention and positional operation then runs in this shared ray space rather than in pixel or pinhole-camera coordinates. Two mechanisms carry the heterogeneous-camera argument: (1) calibration tokens, learnable per network layer and per camera type, injected only for fisheye views and kept view-local in cross-attention, so lens distortion is absorbed without contaminating cross-view fusion; (2) a Jacobian distortion bias, computed from per-patch ray fields and their finite-difference Jacobians, added to cross-view attention logits to bias ma

Load-bearing premise

The headline heterogeneous-camera advantage rests on OmniScene-Full, a synthetic benchmark rendered with the same six-camera rig geometry, asset pipeline, and trajectory sampler that generated the model's training data, so the result may reflect in-distribution familiarity rather than genuine cross-lens generalization.

What would settle it

Take a real mixed fisheye-pinhole rig (roughly four 180-degree fisheye and two pinhole cameras) with LiDAR ground truth and intrinsics that differ from OmniScene's, and evaluate X-Lens zero-shot without any fine-tuning. If AbsRel degrades to near or above the level of strong monocular any-camera baselines, or if removing the Jacobian distortion bias makes no measurable difference, the claim that the geometry modules drive heterogeneous accuracy would be falsified.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If the central claim is correct, real-time metric depth on mixed fisheye-pinhole rigs is achievable with a 0.04B-parameter model, eliminating the need for panorama stitching or separate networks per camera type.
  • The 25.4% AbsRel improvement over the strongest baseline on OmniScene-Full, with 88.9% fewer parameters, suggests the geometry-aware modules add measurable accuracy rather than relying on scale.
  • The three-stage training recipe (pinhole pre-train, fisheye token adaptation, heterogeneous joint fine-tune) provides a transferable schedule for other heterogeneous sensor-fusion problems.
  • Because all reasoning is in ray space, the same architecture can ingest pinhole, fisheye, and 360-degree cameras without architectural changes, as the paper states.
  • The released OmniScene dataset, with scene-disjoint splits, provides a benchmark and training resource for future heterogeneous-camera depth research.

Where Pith is reading between the lines

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

  • A natural next test, left implicit by the paper, is zero-shot evaluation on a real mixed fisheye-pinhole rig with LiDAR ground truth; if transfer holds, X-Lens could serve as a cheap geometric encoder for robot policies, a direction the appendix begins to explore.
  • The Jacobian-bias idea is generic: any multi-camera transformer that knows pixel-to-ray mappings could inject similar local projection priors, potentially improving tasks such as occupancy prediction or 3D detection on heterogeneous rigs.
  • The heaviest dependency is the synthetic-to-real bridge; the paper's own limitation section admits a sim-to-real gap for extreme FOV, so the headline gain may shrink outside OmniScene-like intrinsics.
  • A testable extension is to ablate the calibration tokens and Jacobian bias on a real out-of-domain rig; if their contribution vanishes there, the gain may reflect training-distribution overlap rather than geometric conditioning.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper presents X-Lens, a 0.04B-parameter feed-forward model for metric depth estimation from a variable number of calibrated fisheye and pinhole views. The model combines a DINOv2/DPT backbone with three main components: layer- and camera-type-specific calibration tokens, a Jacobian-parameterized distortion bias injected into cross-view attention, and a confidence-weighted 'Scale Attention' pooled metric-scale head. Training is staged: (1) pinhole multi-view pretraining on 13 public datasets, (2) fisheye adaptation using only calibration tokens on the new OmniScene synthetic dataset, and (3) joint heterogeneous fine-tuning on OmniScene, KITTI360, and pinhole data. OmniScene contains about 266K six-view frames (four 180° fisheye plus two pinhole cameras) rendered from 103 Kujiale/UE scenes. Experiments cover fisheye-only (KITTI360, OmniScene-Single/Quad), pinhole-only (ETH3D, ScanNet++V2, OmniOcc), and heterogeneous (OmniScene-Full) settings. The headline claim is that X-Lens reduces AbsRel by 25.4% on OmniScene-Full over the strongest baseline (UniDAC) while using 88.9% fewer parameters and running at 22 FPS. The paper also reports competitive fisheye results and 41 FPS on KITTI360, plus a small downstream robotic-manipulation study.

Significance. If the heterogeneous-camera claims hold beyond the exact training distribution, X-Lens would be a practically valuable contribution: a compact, real-time model with explicit metric scale, no auxiliary reconstruction heads, and native support for mixed fisheye/pinhole rigs. The method is presented clearly, the loss in Eq. (7) directly supervises the metric scale through the |m-hat - m|/m term, so there is no evident circularity, and the staged training design is well motivated. The ablation study in Table 4 is thorough and internally consistent, showing that each stage and each proposed module contributes within the evaluated distribution. The release of code and the OmniScene dataset is a useful resource. However, the central 'superior heterogeneous-camera metric depth accuracy' claim is currently supported only by an in-distribution synthetic benchmark against zero-shot baselines, and the paper's own limitations section concedes a sim-to-real gap for unseen fisheye/FOV configurations. The significance of the work therefore depends on additional matched-training or real-mixed-rig evidence.

major comments (3)
  1. [§5.2, Table 3; §4; §D] The central heterogeneous-camera claim is measured only on OmniScene-Full. The test scenes are scene-disjoint, but the test split uses the same rendering assets (Kujiale/UE), the same Kannala-Brandt 180° fisheye model, the same pinhole intrinsics, the same six-camera rig, and the same occupancy-waypoint trajectory sampler as the training split. The compared baselines (UniDAC, DepthAnyCamera, MapAnything) are applied zero-shot, with monocular baselines applied per view without cross-view fusion, while X-Lens is trained on OmniScene in Stages 2 and 3. The 25.4% AbsRel reduction (0.1368→0.1021) and 68.1% Scale AbsRel reduction may therefore largely reflect train-test distribution overlap and evaluation asymmetry rather than the calibration-token/Jacobian-bias design. The internal ablations in Table 4 show the modules help within this distribution, but they do not control for data access. Pl
  2. [§5.1, Table 2 (ETH3D)] On real pinhole ETH3D, X-Lens achieves AbsRel 0.0445 and δ1 0.9723, compared with MapAnything at 0.0228/0.9992 and VGGT-Omega at 0.0055/0.9994. This is a large gap on a real-world benchmark and contradicts the abstract's 'competitive performance on conventional pinhole-only settings.' The paper reports only Scale AbsRel as the headline there (0.1217, which is best), but the dense-depth metrics are substantially worse than strong pinhole baselines. This result should be honestly contextualized; as written, it undermines the claim that X-Lens preserves pinhole performance and raises concerns about zero-shot generalization beyond the synthetic training distribution.
  3. [§7 (Limitations)] The Limitations section explicitly states that for 'unseen fisheye lens models with extreme FOV configurations that deviate drastically from the training distribution, a noticeable sim-to-real gap remains,' and the model strictly requires ground-truth calibration. There is no real-world mixed fisheye+pinhole evaluation in the paper: KITTI360 is fisheye-only, OmniOcc is pinhole-only, and OmniScene-Full is synthetic. Since the paper's central claim is 'superior heterogeneous-camera metric depth accuracy,' the absence of any real mixed-rig test means the headline claim is unsupported outside the synthetic training distribution. At minimum, add a small real mixed-rig evaluation or explicitly reframe the headline as an in-domain synthetic result.
minor comments (4)
  1. [§5.2, Table 3] The comparison with MapAnything claims '96.7% fewer parameters,' but MapAnything is a general feed-forward reconstruction model solving a broader task (joint depth, point maps, poses) than X-Lens. Parameter-count comparisons across models with different output scopes should be accompanied by a task-scope qualifier.
  2. [§5, Implementation Details] The FPS numbers are measured on a single H100 without specifying batch size, memory, or whether the reported FPS includes preprocessing/calibration-token injection. Please provide a reproducible speed protocol.
  3. [§6, Table 4] The 'Stage-2 only' row for pinhole OmniOcc is identical to 'Stage-1 only' (0.1052/0.0717/0.2801). This is expected since Stage 2 uses only fisheye data, but it should be stated explicitly to avoid confusion.
  4. [§7, Limitations] The wording 'noticeable sim-to-real gap remains, causing a slight degradation in performance' is internally inconsistent: a 'noticeable gap' is hard to reconcile with 'slight degradation.' Please align the language with the actual evidence.

Circularity Check

0 steps flagged

No significant circularity; the central depth/scale derivation is directly supervised and the evaluation-protocol concerns do not reduce to a derivation-by-construction.

full rationale

The paper's claimed derivation chain is self-contained. Metric scale is a directly supervised output: the total loss in Eq. (7) includes the term λ_scale |m̂−m|/m, and normalized depth is supervised in Eq. (8) against mean-normalized ground truth; no predicted quantity is defined in terms of its own evaluation target or fitted from the reported error metrics. The Jacobian distortion bias and calibration tokens are computed from the input calibration ray field by Eqs. (1), (3)–(5), not from depth ground truth. The OmniScene-Full headline result is obtained on a held-out, scene-disjoint split of the synthetic dataset used during training, which is a legitimate if limited evaluation design; the in-distribution nature of the test split and the zero-shot baseline setup are benchmark-comparability concerns, not circular reductions. The paper's own Limitations section concedes sim-to-real gaps for extreme unseen fisheye configurations, which further supports reading this as an external-generalization limitation rather than a derivational circularity. The only overlapping-author citation that appears (VO-DP [48]) is used in an auxiliary downstream appendix as a baseline, not as a load-bearing justification for the central depth-accuracy claim. No self-citation uniqueness theorem, ansatz-importation chain, or fitted-parameter-renamed-as-prediction is present.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 4 invented entities

The paper contributes a learned architecture and a dataset; its central numerical claims rest on a set of hand-chosen loss weights, token counts, and a synthetic in-distribution benchmark. The architecture itself is a data-driven regressor, so the 'free parameters' are largely training hyperparameters rather than physical constants, and the strongest empirical circularity signal is the OmniScene-Full evaluation being from the same synthetic pipeline used for training.

free parameters (4)
  • Loss weights lambda_depth, lambda_grad, lambda_local, lambda_scale, lambda_mask, lambda_conf = 1.0, 1.0, 0.5, 1.0, 0.2, 1.0
    Hand-chosen hyperparameters balancing depth, gradient, local, scale, mask, and confidence objective terms (Eq. 7); no sensitivity analysis is reported.
  • Calibration token count per layer K = 16
    Number of learnable inject-attend-drop tokens per layer per camera type (Sec. 3.2); no ablation over token count.
  • Scale Attention confidence discard ratio = 25%
    Lowest-confidence 25% of pixels are discarded before confidence-weighted pooling in Eq. (6); threshold chosen without reported sensitivity study.
  • Stage-3 view-count sampling range = [6,8]
    Random view-count range during heterogeneous joint training (Sec. 5); influences generalization to variable camera configurations.
axioms (4)
  • domain assumption Ground-truth camera calibration (intrinsics, extrinsics, camera type) is available for every input view.
    All geometric conditioning uses the unprojection map G in Eq. (1); the model never predicts calibration, and the Limitations section states this reliance explicitly.
  • domain assumption OmniScene fisheye projections follow the Kannala-Brandt model with a fixed 180-degree FOV and a single radial profile.
    The synthetic dataset's camera setup uses Kannala-Brandt fisheye lenses (Sec. 4); the Limitations concede that unseen extreme-FOV lens models cause a sim-to-real gap.
  • domain assumption DINOv2/DPT backbones supply transferable image features for metric depth regression.
    The network is built on pre-trained DINOv2 and DPT heads (Sec. 3.2); no first-principles justification is offered, and the paper leverages them as unexamined strong priors.
  • domain assumption A single global scalar m-hat suffices to convert normalized depth to metric depth for the whole rig.
    Factored prediction in Eq. (2) assumes one per-rig metric scale, with no per-camera scale drift; this is a design choice that could fail if rig calibration or synchronization have residual errors.
invented entities (4)
  • OmniScene dataset no independent evidence
    purpose: 266K-frame synthetic six-camera heterogeneous depth training and evaluation corpus
    Central benchmark for the headline claim; dataset is described but not linked or released in the paper, so there is no external falsifiable handle yet.
  • Jacobian Distortion Bias no independent evidence
    purpose: Cross-attention bias B(k) computed from patch-ray Jacobians to align fisheye and pinhole features
    Internal mechanism ablated in Table 4; no independent test outside the paper.
  • Multi-View Calibration Tokens no independent evidence
    purpose: Layer- and camera-type-specific learnable tokens injected into attention to absorb lens distortion
    Internal mechanism adapted from monocular calibration tokens; no external falsifiable prediction.
  • Scale Attention no independent evidence
    purpose: Confidence-gated pooling over spatially reliable pixels to regress global metric scale
    Internal mechanism with a hand-chosen 25% confidence discard threshold; no independent validation of the threshold.

pith-pipeline@v1.3.0-alltime-deepseek · 23687 in / 13816 out tokens · 114173 ms · 2026-08-02T06:08:22.377902+00:00 · methodology

0 comments
read the original abstract

We present X-lens, a compact feed-forward model for metric depth estimation from a variable number of calibrated fisheye and pinhole views. To support real-time downstream perception, X-lens is built around a geometry-aware heterogeneous camera formulation with two key components. Learnable calibration tokens provide a coarse alignment between fisheye and pinhole projective spaces, while a Jacobian-parameterized distortion bias injected into cross-attention models local projection changes and promotes cross-camera consistency, enabling robust generalization with only 0.04B parameters and up to 41 FPS. The model predicts dense depth together with a global metric scale, avoiding auxiliary reconstruction targets that increase computation and optimization complexity. To learn such cross-camera generalization at scale and depth, X-lens is trained on multiple public datasets and OmniScene, our newly released large-scale synthetic dataset containing approximately 266K synchronized six-view frames, 1.7M individual images, and 103 indoor and outdoor scenes. Extensive experiments on both real-world and synthetic indoor and outdoor datasets demonstrate superior heterogeneous-camera metric depth accuracy, reducing AbsRel by 25.4\% on OmniScene-Full over the strongest baseline while using 88.9\% fewer parameters, with competitive performance on conventional fisheye-only and pinhole-only settings.

Figures

Figures reproduced from arXiv: 2607.12993 by Bohao Zhang, Chenhui Hou, Fa Fu, Heng Zhou, Lijun Han, Shuhong Liu, Wei Sui, Xianbao Hou, Yonghao He.

Figure 1
Figure 1. Figure 1: Teaser illustration of X-Lens. The top row shows metric point clouds projected from calibrated six-view heterogeneous cameras [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Radar-chart comparison across fisheye, pinhole, and heterogeneous-camera benchmarks. X-Lens consistently delivers strong [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Overview of the proposed X-Lens pipeline. Heterogeneous multi-view inputs are represented through generic camera rays and [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Overview of OmniScene. The dataset covers diverse scene categories, including urban, nature, public, and sci-fi environments. [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative results from one fisheye view and one pinhole view in the six-view OmniScene-Full setting. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative results on the real-world OmniOcc dataset and real-world fisheye scenes using X-Lens. The figure shows fisheye-only [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Point-cloud comparison on the synthetic OmniScene-Full setting with six multi-view inputs, consisting of four fisheye and two [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Qualitative ablation on the six-view heterogeneous OmniScene-Full setting. We show predictions from one fisheye view and one [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Visualization of the bias-correction magnitude from [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Qualitative comparisons on OmniScene scenes with a six-camera heterogeneous rig (two pinhole cameras and four fisheye [PITH_FULL_IMAGE:figures/full_fig_p023_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Additional qualitative comparisons on OmniScene scenes under the same six-camera heterogeneous setting (two pinhole [PITH_FULL_IMAGE:figures/full_fig_p024_11.png] view at source ↗

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

96 extracted references · 42 linked inside Pith

  1. [1]

    FishRoPE: Projective rotary position embeddings for omnidirectional visual perception.arXiv preprint arXiv:2604.10391, 2026

    Rahul Ahuja, Mudit Jain, Bala Murali Manoghar Sai Sudhakar, Venkatraman Narayanan, Pratik Likhar, Varun Ravi Kumar, and Senthil Yogamani. FishRoPE: Projective rotary position embeddings for omnidirectional visual perception.arXiv preprint arXiv:2604.10391, 2026. 6, 7

    Pith/arXiv arXiv 2026

  2. [2]

    Depth anything in 360: Towards scale invariance in the wild.arXiv preprint arXiv:2512.22819, 2025

    Hao Ai, Zhi Cao, Meixi Song, Yuxuan Liu, Haodong Li, Dizhe Zhang, Ming-Hsuan Yang, and Lu Qi. Depth anything in 360: Towards scale invariance in the wild.arXiv preprint arXiv:2512.22819, 2025. 2, 3

    arXiv 2025

  3. [3]

    Scenescript: Reconstructing scenes with an autoregressive structured language model

    Armen Avetisyan, Christopher Xie, Henry Howard-Jenkins, Tsun-Yi Yang, Samir Aroudj, Suvam Patra, Fuyang Zhang, Duncan Frost, Luke Holland, Campbell Orme, Jakob Engel, Edward Miller, Richard Newcombe, and Vasileios Balntas. Scenescript: Reconstructing scenes with an autoregressive structured language model. InEuropean Conference on Computer Vision (ECCV), 2024. 9

    2024

  4. [4]

    AdaBins: Depth estimation using adaptive bins

    Shariq Farooq Bhat, Ibraheem Alhashim, and Peter Wonka. AdaBins: Depth estimation using adaptive bins. InCVPR, 2021. 3

    2021

  5. [5]

    ZoeDepth: Zero-shot transfer by combining relative and metric depth.arXiv preprint arXiv:2302.12288, 2023

    Shariq Farooq Bhat, Reiner Birkl, Diana Wofk, Peter Wonka, and Matthias M ¨uller. ZoeDepth: Zero-shot transfer by combining relative and metric depth.arXiv preprint arXiv:2302.12288, 2023. 2, 3

    Pith/arXiv arXiv 2023

  6. [6]

    Richter, and Vladlen Koltun

    Aleksei Bochkovskii, Amael Delaunoy, Hugo Germain, Marcel Santos, Yichao Zhou, Stephan R. Richter, and Vladlen Koltun. Depth pro: Sharp monocular metric depth in less than a second.arXiv preprint arXiv:2410.02073, 2024. 3

    Pith/arXiv arXiv 2024

  7. [7]

    RT-2: Vision-language-action models transfer web knowledge to robotic control.arXiv preprint arXiv:2307.15818, 2023

    Anthony Brohan, Noah Brown, Justice Carbajal, Yevgen Chebotar, Xi Chen, Krzysztof Choromanski, Tianli Ding, Danny Driess, Avinava Dubey, Chelsea Finn, Pete Florence, Chuyuan Fu, Montse Gonzalez Arenas, Keerthana Gopalakrishnan, Kehang Han, Karol Hausman, Alex Herzog, Jasmine Hsu, Brian Ichter, Alex Irpan, Nikhil Joshi, Ryan Julian, Dmitry Kalashnikov, Yuh...

    Pith/arXiv arXiv 2023

  8. [8]

    Jake Bruce, Michael D. Dennis, Ashley Edwards, Jack Parker-Holder, Yuge Shi, Edward Hughes, Matthew Lai, Aditi Mavalankar, Richie Steigerwald, Chris Apps, Yusuf Aytar, Sarah Bechtle, Feryal Behbahani, Stephanie C. Y . Chan, Nicolas Heess, Laura Gonzalez, Simon Osindero, Sherjil Ozair, Scott Reed, Jingwei Zhang, Konrad Zolna, Jeff Clune, Tim Rocktaschel, S...

    Pith/arXiv arXiv 2024

  9. [9]

    D ´ej`a View: Looping transformers for multi-view 3d reconstruction

    Alessandro Burzio, Tobias Fischer, Sven Elflein, Qunjie Zhou, Riccardo de Lutio, Jiawei Ren, Jiahui Huang, Shengyu Huang, Marc Pollefeys, Laura Leal-Taix´e, Zan Gojcic, and Haithem Turki. D ´ej`a View: Looping transformers for multi-view 3d reconstruction. arXiv preprint arXiv:2605.30215, 2026. 4

    Pith/arXiv arXiv 2026

  10. [10]

    Robotwin 2.0: A scalable data generator and benchmark with strong domain randomization for robust bimanual robotic manipulation.ArXiv, 2025

    Tianxing Chen, Zanxin Chen, Baijun Chen, Zijian Cai, Yibin Liu, Qiwei Liang, Zixuan Li, Xianlian Lin, Yiheng Ge, Zhenyu Gu, Weiliang Deng, Yubin Guo, Tian Nian, Xuanbing Xie, Qiangyu Chen, Kailun Su, Tianling Xu, Guodong Liu, Mengkang Hu, Huan ang Gao, Kaixuan Wang, Zhixuan Liang, Yusen Qin, Xiaokang Yang, Ping Luo, and Yao Mu. Robotwin 2.0: A scalable da...

    2025

  11. [11]

    Youyu Chen, Junjun Jiang, Yueru Luo, Kui Jiang, Xianming Liu, Xu Yan, and Dave Zhenyu Chen. Reliev3R: Relieving feed-forward reconstruction from multi-view geometric annotations.Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 21860–21869, 2026. 4

    2026

  12. [12]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. InICLR, 2021. 6

    2021

  13. [13]

    Depth map prediction from a single image using a multi-scale deep network

    David Eigen, Christian Puhrsch, and Rob Fergus. Depth map prediction from a single image using a multi-scale deep network. In Advances in Neural Information Processing Systems, 2014. 3

    2014

  14. [14]

    Adarsh Ganesan, Sravan Puligilla, Ruicheng Zhang, Kamal Joshi, Sanja Fidler, and David F. Fouhey. UniDAC: Universal metric depth estimation for any camera. InCVPR, 2026. 3, 10, 11, 13, 14, S3

    2026

  15. [15]

    Extending foundational monocular depth estimators to fisheye cameras with calibration tokens

    Rit Gangopadhyay, Jung-Hee Kim, Xien Chen, Patrick Rim, Hyoungseob Park, and Alex Wong. Extending foundational monocular depth estimators to fisheye cameras with calibration tokens. InICCV, 2025. 6

    2025

  16. [16]

    Cl ´ement Godard, Oisin Mac Aodha, Michael Firman, and Gabriel J. Brostow. Digging into self-supervised monocular depth estima- tion. InICCV, 2019. 3

    2019

  17. [17]

    3D packing for self-supervised monocular depth estimation

    Vitor Guizilini, Rares Ambrus, Sudeep Pillai, Allan Raventos, and Adrien Gaidon. 3D packing for self-supervised monocular depth estimation. InCVPR, 2020. 3

    2020

  18. [18]

    Lotus: Diffusion-based visual foundation model for high-quality dense prediction.arXiv preprint arXiv:2409.18124, 2024

    Jing He, Haodong Li, Sili Chen, Jingkai Wang, Zhenyu Li, Xiaokang Chen, and Lei Zhang. Lotus: Diffusion-based visual foundation model for high-quality dense prediction.arXiv preprint arXiv:2409.18124, 2024. 3

    Pith/arXiv arXiv 2024

  19. [19]

    Metric3D v2: A versatile monocular geometric foundation model for zero-shot metric depth and surface normal estimation.arXiv preprint arXiv:2404.15506, 2024

    Mu Hu, Wei Yin, Chi Zhang, Zhipeng Cai, Xiaoxiao Long, Hao Chen, Kaixuan Wang, Gang Yu, and Chunhua Shen. Metric3D v2: A versatile monocular geometric foundation model for zero-shot metric depth and surface normal estimation.arXiv preprint arXiv:2404.15506, 2024. 3, 10, 11

    Pith/arXiv arXiv 2024

  20. [20]

    Yeh, and Alexander G

    Yuan-Ting Hu, Jiahong Wang, Raymond A. Yeh, and Alexander G. Schwing. SAIL-VOS 3D: A synthetic dataset and baselines for object detection and 3d mesh reconstruction from video data. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021. 9

    2021

  21. [21]

    Deepmvs: Learning multi-view stereopsis

    Po-Han Huang, Kevin Matzen, Johannes Kopf, Narendra Ahuja, and Jia-Bin Huang. Deepmvs: Learning multi-view stereopsis. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 9

    2018

  22. [22]

    MVSAnywhere: Zero-shot multi-view stereo.arXiv preprint arXiv:2503.22430,

    Sergio Izquierdo, Mohamed Sayed, Michael Firman, Guillermo Garcia-Hernando, Daniyar Turmukhambetov, Javier Civera, Oisin Mac Aodha, Gabriel Brostow, and Jamie Watson. MVSAnywhere: Zero-shot multi-view stereo.arXiv preprint arXiv:2503.22430,

    Pith/arXiv arXiv

  23. [23]

    UniFuse: Unidirectional fusion for 360 panorama depth estimation

    Hualie Jiang, Zheng Sheng, Siyu Zhu, Zilong Dong, and Rui Huang. UniFuse: Unidirectional fusion for 360 panorama depth estimation. InICRA, 2021. 2, 3

    2021

  24. [24]

    AnySplat: Feed-forward 3d gaussian splatting from unconstrained views.arXiv preprint arXiv:2505.23716, 2025

    Lihan Jiang, Yucheng Mao, Linning Xu, Tao Lu, Kerui Ren, Yichen Jin, Xudong Xu, Mulin Yu, Jiangmiao Pang, Feng Zhao, et al. AnySplat: Feed-forward 3d gaussian splatting from unconstrained views.arXiv preprint arXiv:2505.23716, 2025. 4

    arXiv 2025

  25. [25]

    Barron, Noah Snavely, and Aleksander Holynski

    Haian Jin, Rundi Wu, Tianyuan Zhang, Ruiqi Gao, Jonathan T. Barron, Noah Snavely, and Aleksander Holynski. ZipMap: Linear- time stateful 3d reconstruction with test-time training.arXiv preprint arXiv:2603.04385, 2026. 4

    Pith/arXiv arXiv 2026

  26. [26]

    Wid3R: Wide field-of-view 3d recon- struction via camera model conditioning.arXiv preprint arXiv:2602.05321, 2026

    Dongki Jung, Jaehoon Choi, Adil Qureshi, Somi Jeong, Dinesh Manocha, and Suyong Yeon. Wid3R: Wide field-of-view 3d recon- struction via camera model conditioning.arXiv preprint arXiv:2602.05321, 2026. 4

    arXiv 2026

  27. [27]

    Dynamicstereo: Consistent dynamic depth from stereo videos.CVPR, 2023

    Nikita Karaev, Ignacio Rocco, Benjamin Graham, Natalia Neverova, Andrea Vedaldi, and Christian Rupprecht. Dynamicstereo: Consistent dynamic depth from stereo videos.CVPR, 2023. 9

    2023

  28. [28]

    MapAnything: Universal feed-forward metric 3d reconstruction.arXiv preprint arXiv:2509.13414, 2025

    Nikhil Keetha, Norman M ¨uller, Johannes Sch ¨onberger, Lorenzo Porzi, Yuchen Zhang, Tobias Fischer, Arno Knapitsch, Duncan Zauss, Ethan Weber, Nelson Antunes, Jonathon Luiten, Manuel Lopez-Antequera, Samuel Rota Bul `o, Christian Richardt, Deva Ramanan, Sebastian Scherer, and Peter Kontschieder. MapAnything: Universal feed-forward metric 3d reconstructio...

    Pith/arXiv arXiv 2025

  29. [29]

    OpenVLA: An open-source vision-language-action model.arXiv preprint arXiv:2406.09246, 2024

    Moo Jin Kim, Karl Pertsch, Siddharth Karamcheti, Ted Xiao, Ashwin Balakrishna, Suraj Nair, Rafael Rafailov, Ethan Foster, Grace Lam, Pannag Sanketi, Quan Vuong, Thomas Kollar, Benjamin Burchfiel, Russ Tedrake, Dorsa Sadigh, Chelsea Finn, and Percy Liang. OpenVLA: An open-source vision-language-action model.arXiv preprint arXiv:2406.09246, 2024. 2

    Pith/arXiv arXiv 2024

  30. [30]

    From big to small: Multi-scale local planar guidance for monocular depth estimation

    Jin Han Lee, Myung-Kyu Han, Dong Wook Ko, and Il Hong Suh. From big to small: Multi-scale local planar guidance for monocular depth estimation. InarXiv preprint arXiv:1907.10326, 2019. 3

    Pith/arXiv arXiv 1907

  31. [31]

    Grounding image matching in 3d with MASt3R.arXiv preprint arXiv:2406.09756, 2024

    Vincent Leroy, Yohann Cabon, and J ´erˆome Revaud. Grounding image matching in 3d with MASt3R.arXiv preprint arXiv:2406.09756, 2024. 2, 4

    Pith/arXiv arXiv 2024

  32. [32]

    Fisheyedistancenet: Self-supervised scale-aware distance estimation using monocular fisheye camera for autonomous driving.arXiv preprint arXiv:2112.13842, 2021

    Yuyan Li, Sheng Gu, Christoph Mayer, Luc Van Gool, and Radu Timofte. Fisheyedistancenet: Self-supervised scale-aware distance estimation using monocular fisheye camera for autonomous driving.arXiv preprint arXiv:2112.13842, 2021. 3

    Pith/arXiv arXiv 2021

  33. [33]

    CasOmniMVS: Cascade omnidirectional depth estimation with dynamic spherical sweeping.Applied Sciences, 2024

    Yuyan Li, Peng Wang, Lingjie Liu, and Wenping Wang. CasOmniMVS: Cascade omnidirectional depth estimation with dynamic spherical sweeping.Applied Sciences, 2024. 3

    2024

  34. [34]

    Megadepth: Learning single-view depth prediction from internet photos

    Zhengqi Li and Noah Snavely. Megadepth: Learning single-view depth prediction from internet photos. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 9

    2018

  35. [35]

    PatchRefiner v2: Fast and lightweight real-domain high-resolution metric depth estimation.arXiv preprint arXiv:2501.01121, 2025

    Zhenyu Li, Xuyang Wang, Bingyi Kang, Jiashi Feng, and Hengshuang Zhao. PatchRefiner v2: Fast and lightweight real-domain high-resolution metric depth estimation.arXiv preprint arXiv:2501.01121, 2025. 3

    Pith/arXiv arXiv 2025

  36. [36]

    Language as prior, vision as calibration: Metric scale recovery for monocular depth estimation.arXiv preprint arXiv:2601.01457, 2026

    Zhenyu Li, Xuyang Wang, Bingyi Kang, Jiashi Feng, and Hengshuang Zhao. Language as prior, vision as calibration: Metric scale recovery for monocular depth estimation.arXiv preprint arXiv:2601.01457, 2026. 3

    arXiv 2026

  37. [37]

    KITTI-360: A novel dataset and benchmarks for urban scene understanding in 2d and 3d

    Yiyi Liao, Jun Xie, and Andreas Geiger. KITTI-360: A novel dataset and benchmarks for urban scene understanding in 2d and 3d. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(3):3292–3310, 2023. 9, 10, 11, S3

    2023

  38. [38]

    Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang

    Haotong Lin, Sili Chen, Junhao Liew, Donny Y . Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang. Depth anything 3: Recovering the visual space from any views.arXiv preprint arXiv:2511.10647, 2025. 4, 8, 10, 12, 14

    Pith/arXiv arXiv 2025

  39. [39]

    Mix3R: Mixing feed-forward reconstruction and generative 3d priors for joint multi-view aligned 3d reconstruction and pose estimation.arXiv preprint arXiv:2605.03359, 2026

    Siyou Lin, Zhou Xue, Hongwen Zhang, Liang An, Dongping Li, Shaohui Jiao, and Yebin Liu. Mix3R: Mixing feed-forward reconstruction and generative 3d priors for joint multi-view aligned 3d reconstruction and pose estimation.arXiv preprint arXiv:2605.03359, 2026. 4

    Pith/arXiv arXiv 2026

  40. [40]

    Depth any panoramas: A foundation model for panoramic depth estimation

    Xin Lin, Meixi Song, Dizhe Zhang, Wenxuan Lu, Haodong Li, Bo Du, Ming-Hsuan Yang, Truong Nguyen, and Lu Qi. Depth any panoramas: A foundation model for panoramic depth estimation. InCVPR, 2026. 2, 3, 4

    2026

  41. [41]

    DL3DV-10K: A large-scale scene dataset for deep learning-based 3d vision

    Lu Ling, Yichen Sheng, Zhi Tu, Wentian Zhao, Cheng Xin, Kun Wan, Lantao Yu, Qianyu Guo, Zixun Yu, Yawen Lu, Xuanmao Li, Xingpeng Sun, Rohan Ashok, Aniruddha Mukherjee, Hao Kang, Xiangrui Kong, Gang Hua, Tianyi Zhang, Bedrich Benes, and Aniket Bera. DL3DV-10K: A large-scale scene dataset for deep learning-based 3d vision. InProceedings of the IEEE/CVF Conf...

    2024

  42. [42]

    Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. In International Conference on Learning Representations (ICLR), 2023. S1

    2023

  43. [43]

    Mapillary planet-scale depth dataset

    Manuel L ´opez Antequera, Pau Gargallo, Markus Hofinger, Samuel Rota Bul `o, Yubin Kuang, and Peter Kontschieder. Mapillary planet-scale depth dataset. InEuropean Conference on Computer Vision (ECCV), 2020. 9

    2020

  44. [44]

    WideDepth: Millimeter-accurate benchmark for fisheye depth estimation.arXiv preprint arXiv:2605.24074, 2026

    Wenxuan Lu, Xin Lin, Meixi Song, Dizhe Zhang, Haodong Li, and Lu Qi. WideDepth: Millimeter-accurate benchmark for fisheye depth estimation.arXiv preprint arXiv:2605.24074, 2026. 3

    Pith/arXiv arXiv 2026

  45. [45]

    SimpleProc: Fully procedural synthetic data from simple rules for multi-view stereo

    Zeyu Ma, Alexander Raistrick, and Jia Deng. SimpleProc: Fully procedural synthetic data from simple rules for multi-view stereo. arXiv preprint arXiv:2604.04925, 2026. 4

    Pith/arXiv arXiv 2026

  46. [46]

    Spring: A high-resolution high-detail dataset and benchmark for scene flow, optical flow and stereo

    Lukas Mehl, Jenny Schmalfuss, Azin Jahedi, Yaroslava Nalivayko, and Andr ´es Bruhn. Spring: A high-resolution high-detail dataset and benchmark for scene flow, optical flow and stereo. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4981–4991, 2023. 9

    2023

  47. [47]

    Andreas Meuleman, HyunJun Jang, Dongju Kim, Dae Gyu Jeon, and Min H. Kim. Real-time sphere sweeping stereo from multiview fisheye images. InCVPR, 2021. 3

    2021

  48. [48]

    V o-dp: Semantic-geometric adaptive diffusion policy for vision-only robotic manipulation.ArXiv, 2025

    Zehao Ni, Yonghao He, Lingfeng Qian, Jilei Mao, Fa Fu, Wei Sui, Hu Su, Junran Peng, Zhipeng Wang, and Bin He. V o-dp: Semantic-geometric adaptive diffusion policy for vision-only robotic manipulation.ArXiv, 2025. S1

    2025

  49. [49]

    Octo: An open-source generalist robot policy.arXiv preprint arXiv:2405.12213, 2024

    Octo Model Team, Dibya Ghosh, Homer Walke, Karl Pertsch, Kevin Black, Oier Mees, Sudeep Dasari, Joey Hejna, Charles Xu, Jianlan Luo, Tess Kreiman, You Liang Tan, Dorsa Chen, Pannag Sanketi, Quan Vuong, Ted Xiao, Dorsa Sadigh, Chelsea Finn, and Sergey Levine. Octo: An open-source generalist robot policy.arXiv preprint arXiv:2405.12213, 2024. 2

    Pith/arXiv arXiv 2024

  50. [50]

    DINOv2: Learning robust visual features without supervision.Transactions on Machine Learning Research (TMLR), 2024

    Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. DINOv2: Learning robust visual features without supervision.Transactions on Machine Learning Research (TMLR), 2024. 2

    2024

  51. [51]

    TartanGround: A large-scale dataset for ground robot perception and navigation

    Manthan Patel, Fan Yang, Yuheng Qiu, Cesar Cadena, Sebastian Scherer, Marco Hutter, and Wenshan Wang. TartanGround: A large-scale dataset for ground robot perception and navigation. In2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025. 9

    2025

  52. [52]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. InIEEE/CVF International Conference on Computer Vision (ICCV), 2023. S1

    2023

  53. [53]

    UniDepth: Universal monocular metric depth estimation

    Luigi Piccinelli, Christos Sakaridis, Yung-Hsu Yang, Mattia Segu, Siyuan Li, Luc Van Gool, and Fisher Yu. UniDepth: Universal monocular metric depth estimation. InCVPR, 2024. 2, 3

    2024

  54. [54]

    UniDepthV2: Universal monocular metric depth estimation made simpler.arXiv preprint arXiv:2502.20110, 2025

    Luigi Piccinelli, Yung-Hsu Yang, Christos Sakaridis, Mattia Segu, Siyuan Li, Luc Van Gool, and Fisher Yu. UniDepthV2: Universal monocular metric depth estimation made simpler.arXiv preprint arXiv:2502.20110, 2025. 3, 10, 11

    Pith/arXiv arXiv 2025

  55. [55]

    UniK3D: Universal camera monocular 3d estimation.arXiv preprint arXiv:2503.16591, 2025

    Luigi Piccinelli, Yung-Hsu Yang, Christos Sakaridis, Mattia Segu, Siyuan Li, Luc Van Gool, and Fisher Yu. UniK3D: Universal camera monocular 3d estimation.arXiv preprint arXiv:2503.16591, 2025. 3

    Pith/arXiv arXiv 2025

  56. [56]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision. InInternational Conference on Machine Learning (ICML), 2021. S1

    2021

  57. [57]

    Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020

    Ren ´e Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, and Vladlen Koltun. Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020. 3

    2020

  58. [58]

    Vision transformers for dense prediction

    Ren ´e Ranftl, Alexey Bochkovskiy, and Vladlen Koltun. Vision transformers for dense prediction. InICCV, 2021. 2, 3

    2021

  59. [59]

    Sch¨onberger, Silvano Galliani, Torsten Sattler, Konrad Schindler, Marc Pollefeys, and Andreas Geiger

    Thomas Sch ¨ops, Johannes L. Sch¨onberger, Silvano Galliani, Torsten Sattler, Konrad Schindler, Marc Pollefeys, and Andreas Geiger. A multi-view stereo benchmark with high-resolution images and multi-camera videos. InCVPR, 2017. 10, 11, 12, S3

    2017

  60. [60]

    PanoFormer: Panorama transformer for indoor 360 depth estimation

    Zhijie Shen, Chunyu Lin, Kang Liao, Lang Nie, Zishuo Zheng, and Yao Zhao. PanoFormer: Panorama transformer for indoor 360 depth estimation. InECCV, 2022. 2, 3

    2022

  61. [61]

    Oriane Sim’eoni, Huy V . V o, Maximilian Seitzer, Federico Baldassarre, Maxime Oquab, Cijo Jose, Vasil Khalidov, Marc Szafraniec, Seungeun Yi, Michael Ramamonjisoa, Francisco Massa, Daniel Haziza, Luca Wehrstedt, Jianyuan Wang, Timoth ´ee Darcet, Th ´eo Moutakanni, Leonel Sentana, Claire Roberts, Andrea Vedaldi, Jamie Tolan, John Brandt, Camille Couprie, ...

    2025

  62. [62]

    RoFormer: Enhanced transformer with rotary position embedding.Neurocomputing, 568:127063, 2024

    Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Bo Wen, and Yunfeng Liu. RoFormer: Enhanced transformer with rotary position embedding.Neurocomputing, 568:127063, 2024

    2024

  63. [63]

    HoHoNet: 360 indoor holistic understanding with latent horizontal features

    Cheng Sun, Ching-Yu Hsiao, Min Sun, and Hwann-Tzong Chen. HoHoNet: 360 indoor holistic understanding with latent horizontal features. InCVPR, 2021. 3

    2021

  64. [64]

    Distortion-aware convolutional filters for dense prediction in panoramic images

    Keisuke Tateno, Nassir Navab, and Federico Tombari. Distortion-aware convolutional filters for dense prediction in panoramic images. InECCV, 2018. 3

    2018

  65. [65]

    Smd-nets: Stereo mixture density networks

    Fabio Tosi, Yiyi Liao, Carolin Schmitt, and Andreas Geiger. Smd-nets: Stereo mixture density networks. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021. 9

    2021

  66. [66]

    Generative camera dolly: Extreme monocular dynamic novel view synthesis

    Basile Van Hoorick, Rundi Wu, Ege Ozguroglu, Kyle Sargent, Ruoshi Liu, Pavel Tokmakov, Achal Dave, Changxi Zheng, and Carl V ondrick. Generative camera dolly: Extreme monocular dynamic novel view synthesis. InEuropean Conference on Computer Vision (ECCV), 2024. 9

    2024

  67. [67]

    BiFuse: Monocular 360 depth estimation via bi- projection fusion

    Fu-En Wang, Yu-Hsuan Yeh, Min Sun, Wei-Chen Chiu, and Yi-Hsuan Tsai. BiFuse: Monocular 360 depth estimation via bi- projection fusion. InCVPR, 2020. 3

    2020

  68. [68]

    V oyager: An open-ended embodied agent with large language models

    Guanzhi Wang, Yuqi Xie, Yunfan Jiang, Ajay Mandlekar, Chaowei Xiao, Yuke Zhu, Linxi Fan, and Anima Anandkumar. V oyager: An open-ended embodied agent with large language models. InNeurIPS, 2023. 2

    2023

  69. [69]

    3d reconstruction with spatial memory.arXiv preprint arXiv:2408.16061, 2024

    Hengyi Wang and Lourdes Agapito. 3d reconstruction with spatial memory.arXiv preprint arXiv:2408.16061, 2024. 4

    Pith/arXiv arXiv 2024

  70. [70]

    VGGT: Visual geometry grounded transformer.arXiv preprint arXiv:2503.11651, 2025

    Jianyuan Wang, Minghao Chen, Nikita Karaev, Andrea Vedaldi, Christian Rupprecht, and David Novotny. VGGT: Visual geometry grounded transformer.arXiv preprint arXiv:2503.11651, 2025. 2, 4, 5, 10, 11, 12, S1

    Pith/arXiv arXiv 2025

  71. [71]

    VGGT-Ω: Scaling feed-forward 3d reconstruction.arXiv preprint arXiv:2605.15195, 2026

    Jianyuan Wang, Minghao Chen, Shangzhan Zhang, Nikita Karaev, Johannes Sch ¨onberger, Patrick Labatut, Piotr Bojanowski, David Novotny, Andrea Vedaldi, and Christian Rupprecht. VGGT-Ω: Scaling feed-forward 3d reconstruction.arXiv preprint arXiv:2605.15195, 2026. 2, 4, 10, 12

    Pith/arXiv arXiv 2026

  72. [72]

    OmniMVS: End-to-end learning for omnidirectional stereo matching.arXiv preprint arXiv:2203.10759, 2022

    Peng Wang, Lingjie Liu, Yuan Liu, Christian Theobalt, Taku Komura, and Wenping Wang. OmniMVS: End-to-end learning for omnidirectional stereo matching.arXiv preprint arXiv:2203.10759, 2022. 3

    Pith/arXiv arXiv 2022

  73. [73]

    Unsupervised OmniMVS: Efficient omnidirectional depth inference via establishing pseudo-stereo supervision.arXiv preprint arXiv:2302.09922, 2023

    Peng Wang, Yuan Liu, Lingjie Liu, Christian Theobalt, Taku Komura, and Wenping Wang. Unsupervised OmniMVS: Efficient omnidirectional depth inference via establishing pseudo-stereo supervision.arXiv preprint arXiv:2302.09922, 2023. 3

    Pith/arXiv arXiv 2023

  74. [74]

    MoGe: Unlocking accurate monocular geometry estimation for open-domain images with optimal training supervision.arXiv preprint arXiv:2410.19115, 2024

    Ruicheng Wang, Sicheng Chen, Jiayuan Xu, Xingyi Li, Zhengqi Li, Zhizheng Zhang, Juefei-Xu Liu, and Ziwei Liu. MoGe: Unlocking accurate monocular geometry estimation for open-domain images with optimal training supervision.arXiv preprint arXiv:2410.19115, 2024. 3, 8, 14

    Pith/arXiv arXiv 2024

  75. [75]

    MoGe-2: Accurate monocular geometry with metric scale and sharp details.arXiv preprint arXiv:2507.02546, 2025

    Ruicheng Wang, Sicheng Chen, Jiayuan Xu, Xingyi Li, Zhengqi Li, Zhizheng Zhang, Juefei-Xu Liu, and Ziwei Liu. MoGe-2: Accurate monocular geometry with metric scale and sharp details.arXiv preprint arXiv:2507.02546, 2025. 3, 8

    Pith/arXiv arXiv 2025

  76. [76]

    AnchorD: Metric grounding of monocular depth using factor graphs.arXiv preprint arXiv:2605.02667, 2026

    Ruicheng Wang, Sicheng Chen, Jiayuan Xu, Xingyi Li, Zhengqi Li, Zhizheng Zhang, Juefei-Xu Liu, and Ziwei Liu. AnchorD: Metric grounding of monocular depth using factor graphs.arXiv preprint arXiv:2605.02667, 2026. 3

    Pith/arXiv arXiv 2026

  77. [77]

    MetricAnything: Scaling metric depth pretraining with noisy heterogeneous sources.arXiv preprint arXiv:2601.22054, 2026

    Ruicheng Wang, Sicheng Chen, Jiayuan Xu, Xingyi Li, Zhengqi Li, Zhizheng Zhang, Juefei-Xu Liu, and Ziwei Liu. MetricAnything: Scaling metric depth pretraining with noisy heterogeneous sources.arXiv preprint arXiv:2601.22054, 2026. 3

    Pith/arXiv arXiv 2026

  78. [78]

    DUSt3R: Geometric 3d vision made easy

    Shuzhe Wang, Vincent Leroy, Yohann Cabon, Boris Chidlovskii, and Jerome Revaud. DUSt3R: Geometric 3d vision made easy. arXiv preprint arXiv:2312.14132, 2023. 2, 4, 5

    Pith/arXiv arXiv 2023

  79. [79]

    Yifan Wang, Jianjun Zhou, Haoyi Zhu, Wenzheng Chang, Yang Zhou, Zizun Li, Junyi Chen, Jiangmiao Pang, Chunhua Shen, and Tong He.π 3: Scalable permutation-equivariant visual geometry learning.arXiv preprint arXiv:2507.13347, 2025. 2, 4

    Pith/arXiv arXiv 2025

  80. [80]

    Omnimvs: End-to-end learning for omnidirectional stereo matching

    Changhee Won, Francois Rameau, Jinwoo Kim, and In So Kweon. Omnimvs: End-to-end learning for omnidirectional stereo matching. InICCV, 2019. 3

    2019

Showing first 80 references.