pith. sign in

arxiv: 2605.30060 · v2 · pith:MGAGJO6Fnew · submitted 2026-05-28 · 💻 cs.CV

Towards Consistent Video Geometry Estimation

Zhu Yu , Jingnan Gao , Runmin Zhang , Lingteng Qiu , Zhengyi Zhao , Rui Peng , Yichao Yan , Kejie Qiu
show 4 more authors
Siyu Zhu Zilong Dong Si-Yuan Cao Hui-Liang Shen
This is my paper

Pith reviewed 2026-06-29 08:00 UTC · model grok-4.3

classification 💻 cs.CV
keywords video geometry estimationdepth estimationsurface normal estimationpoint map estimationtransformer architecturetemporal consistencydata refinementfoundation model
0
0 comments X

The pith

ViGeo recovers dense and temporally consistent video geometry with one plain transformer that adapts attention patterns at test time.

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

The paper introduces ViGeo as a feed-forward foundation model that produces spatially dense and temporally consistent geometry estimates, including depth, surface normals, and point maps, from video sequences. It achieves this with a plain transformer using dynamic chunking attention that trains on both bidirectional and causal contexts so the model can switch between streaming, full-sequence, and long-video inference without retraining. A completion-based data refinement framework trains a teacher model on sparse noisy labels to generate higher-quality dense targets by exploiting video context. A sympathetic reader would care because reliable consistent geometry from ordinary video is a prerequisite for stable 3D scene understanding in changing environments.

Core claim

ViGeo is a feed-forward foundation model for recovering spatially dense and temporally consistent geometry from video sequences. Built upon a plain transformer architecture without task-specific architectural modifications, ViGeo supports streaming, full-sequence, and long-video inference within a unified model. The key design is dynamic chunking attention, which exposes the model to both bidirectional and causal temporal contexts during training and allows it to adapt its attention pattern at test time without retraining. To improve supervision quality, the authors introduce a completion-based data refinement framework that trains a video depth completion teacher conditioning on sparse and

What carries the argument

Dynamic chunking attention, which exposes the model to both bidirectional and causal temporal contexts during training and allows adaptation of the attention pattern at test time without retraining.

If this is right

  • The same trained model can switch between online streaming depth estimation and offline bidirectional processing without retraining.
  • Surface normal and point map predictions are generated alongside depth within one forward pass.
  • Long-video sequences maintain geometric consistency using the adapted attention pattern.
  • Training targets refined by the teacher model improve supervision quality over raw annotations.
  • State-of-the-art results are obtained using only public datasets across the listed tasks.

Where Pith is reading between the lines

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

  • The unified attention mechanism could reduce the engineering overhead of maintaining separate models for different video lengths in production systems.
  • Consistent normal and point map outputs may directly feed into downstream tasks such as video-based 3D reconstruction or SLAM without additional alignment steps.
  • The refinement approach might generalize to other sparse supervision settings in video, such as optical flow or instance segmentation.
  • If the dynamic chunking pattern proves stable, similar attention designs could appear in other sequence models that must support both causal and non-causal inference.

Load-bearing premise

The completion-based data refinement framework produces dense, temporally coherent, and geometrically reliable training targets from sparse and noisy annotations.

What would settle it

If a model trained directly on the original sparse noisy annotations matches or exceeds ViGeo's temporal consistency scores on long-video benchmarks, the contribution of the refinement framework would be falsified.

Figures

Figures reproduced from arXiv: 2605.30060 by Hui-Liang Shen, Jingnan Gao, Kejie Qiu, Lingteng Qiu, Rui Peng, Runmin Zhang, Si-Yuan Cao, Siyu Zhu, Yichao Yan, Zhengyi Zhao, Zhu Yu, Zilong Dong.

Figure 1
Figure 1. Figure 1: ViGeo is a unified feed-forward foundation model for video geometry estimation. It predicts temporally consistent depth, surface normals, and dense point maps from raw video frames. With dynamic chunking attention, the same trained model seamlessly switches between full-sequence reconstruction and streaming inference without retraining. Abstract This work presents ViGeo, a feed-forward foundation model for… view at source ↗
Figure 2
Figure 2. Figure 2: Benchmark comparison with previous state-of-the-art methods. Video geometry estimation is a fundamental prob￾lem in computer vision, supporting applications such as robotics [38], augmented reality [61], au￾tonomous navigation [1], and video editing [11]. These applications require geometry that is both spatially accurate and temporally consistent over long video sequences. Despite recent progress, achievi… view at source ↗
Figure 3
Figure 3. Figure 3: Architecture overview of ViGeo. Built upon a plain Transformer with dynamic chunking attention, ViGeo supports full-sequence, streaming, and long-video inference within a unified model and predicts temporally consistent depth, surface normals, and point maps. models [88, 58], we devise a data engine based on multi-view depth completion, fully leveraging the strengths from both images and sparse measurement… view at source ↗
Figure 4
Figure 4. Figure 4: Visualization of our completion-based data refinement pipeline. Given an RGB video [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative results on 3D reconstruction. Our method yields more accurate and robust [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Additional point cloud visualizations. ViGeo produces accurate and realistic reconstructions [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Qualitative results for video depth estimation. Compared with existing methods, ViGeo [PITH_FULL_IMAGE:figures/full_fig_p016_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Qualitative results for monocular depth estimation. Compared with existing methods, [PITH_FULL_IMAGE:figures/full_fig_p017_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Qualitative ablation of our data refinement pipeline. Our full pipeline effectively resolves [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Qualitative results of the data refinement pipeline. We visualize the sparse raw measure [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗
read the original abstract

This work presents ViGeo, a feed-forward foundation model for recovering spatially dense and temporally consistent geometry from video sequences. Built upon a plain transformer architecture without task-specific architectural modifications, ViGeo supports streaming, full-sequence, and long-video inference within a unified model. The key design is dynamic chunking attention, which exposes the model to both bidirectional and causal temporal contexts during training and allows it to adapt its attention pattern at test time without retraining. To improve supervision quality, we further introduce a completion-based data refinement framework. This framework trains a video depth completion teacher that conditions on sparse and noisy annotations and exploits video/multi-view context to produce dense, temporally coherent, and geometrically reliable training targets. Beyond depth and point maps, ViGeo also predicts surface normals within the same framework. Trained solely on public datasets, ViGeo achieves state-of-the-art performance across online, offline, and long-video depth estimation, surface normal estimation, and video point map estimation.

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

1 major / 1 minor

Summary. The paper presents ViGeo, a feed-forward foundation model based on a plain transformer for recovering spatially dense and temporally consistent geometry (depth, surface normals, point maps) from video. It introduces dynamic chunking attention to enable streaming, full-sequence, and long-video inference within one model. A completion-based data refinement framework trains a video depth completion teacher on sparse/noisy annotations to generate dense, temporally coherent training targets. Trained only on public datasets, ViGeo claims state-of-the-art results on online/offline/long-video depth estimation, surface normal estimation, and video point map estimation.

Significance. If substantiated, the work would advance video geometry estimation by offering a unified, architecture-agnostic foundation model with flexible inference modes. The dynamic chunking attention and public-data training are positive elements that could support broader adoption if the performance claims are rigorously validated.

major comments (1)
  1. [Data refinement framework] Data refinement framework (abstract and methods section): The SOTA claims depend on the teacher producing 'geometrically reliable' dense targets. No quantitative validation against independent dense ground truth is described, nor are details given on the teacher's loss beyond sparse-point conditioning or explicit multi-view consistency terms. This is load-bearing because unverified target quality could mean reported metric gains reflect annotation propagation rather than model capability.
minor comments (1)
  1. [Abstract] The abstract states the model 'exploits video/multi-view context' in the teacher but provides no implementation specifics or ablation on this component.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the thoughtful feedback on our data refinement framework. We address the concern point-by-point below and will incorporate additional details and validation in the revised manuscript to strengthen the presentation of the teacher model's target quality.

read point-by-point responses

  1. Referee: [Data refinement framework] Data refinement framework (abstract and methods section): The SOTA claims depend on the teacher producing 'geometrically reliable' dense targets. No quantitative validation against independent dense ground truth is described, nor are details given on the teacher's loss beyond sparse-point conditioning or explicit multi-view consistency terms. This is load-bearing because unverified target quality could mean reported metric gains reflect annotation propagation rather than model capability.

    Authors: We agree that the current manuscript provides insufficient quantitative validation of the teacher's dense outputs against independent dense ground truth and limited specifics on the full loss formulation. While the framework description emphasizes conditioning on sparse/noisy annotations and exploitation of video/multi-view context to generate coherent targets, we acknowledge this leaves open the possibility that gains partly reflect propagation of existing annotations. In the revision we will expand the methods section with: (1) the complete teacher loss, explicitly including any multi-view consistency or geometric regularization terms beyond sparse-point conditioning; (2) quantitative evaluation of teacher outputs on any available dense ground-truth subsets (e.g., selected sequences from datasets that provide both sparse and dense annotations); and (3) an ablation isolating the effect of the refined targets versus raw sparse supervision. These additions will allow readers to better assess whether the reported improvements stem from model capability or target quality. revision: yes

Circularity Check

0 steps flagged

No significant circularity; claims rest on external benchmarks

full rationale

The provided abstract and description introduce ViGeo and a completion-based data refinement framework that generates training targets from public datasets' sparse annotations. Performance claims are evaluated on standard external metrics for depth, normals, and point maps across online/offline/long-video settings. No equations, self-citations, or derivations are shown that reduce any prediction or result to its own inputs by construction. The framework is presented as a training aid rather than a self-referential definition of success, and no uniqueness theorems or ansatzes are invoked via self-citation. The derivation chain is self-contained against public data and independent benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no information on free parameters, axioms, or invented entities.

pith-pipeline@v0.9.1-grok · 5725 in / 1048 out tokens · 33941 ms · 2026-06-29T08:00:36.370048+00:00 · methodology

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

94 extracted references · 23 canonical work pages · 8 internal anchors

  1. [1]

    En- hanced depth navigation through augmented reality depth mapping in patients with low vision

    Anastasios Nikolas Angelopoulos, Hossein Ameri, Debbie Mitra, and Mark Humayun. En- hanced depth navigation through augmented reality depth mapping in patients with low vision. Scientific reports, 9(1):11230, 2019

    2019

  2. [2]

    Estimating and exploiting the aleatoric uncertainty in surface normal estimation

    Gwangbin Bae, Ignas Budvytis, and Roberto Cipolla. Estimating and exploiting the aleatoric uncertainty in surface normal estimation. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 13137–13146, 2021

    2021

  3. [3]

    Rethinking inductive biases for surface normal estima- tion

    Gwangbin Bae and Andrew J Davison. Rethinking inductive biases for surface normal estima- tion. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9535–9545, 2024

    2024

  4. [4]

    ARKitscenes - a diverse real-world dataset for 3d indoor scene understanding using mobile RGB-d data

    Gilad Baruch, Zhuoyuan Chen, Afshin Dehghan, Tal Dimry, Yuri Feigin, Peter Fu, Thomas Gebauer, Brandon Joffe, Daniel Kurz, Arik Schwartz, and Elad Shulman. ARKitscenes - a diverse real-world dataset for 3d indoor scene understanding using mobile RGB-d data. In Advances in Neural Information Processing Systems Datasets and Benchmarks Track, 2021

    2021

  5. [5]

    Adabins: Depth estimation using adaptive bins

    Shariq Farooq Bhat, Ibraheem Alhashim, and Peter Wonka. Adabins: Depth estimation using adaptive bins. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4009–4018, 2021

    2021

  6. [6]

    Normalcrafter: Learning temporally consistent normals from video diffusion priors

    Yanrui Bin, Wenbo Hu, Haoyuan Wang, Xinya Chen, and Bing Wang. Normalcrafter: Learning temporally consistent normals from video diffusion priors. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 8330–8339, 2025

    2025

  7. [7]

    1–a model zoo for robust monocular relative depth estimation

    Reiner Birkl, Diana Wofk, and Matthias Müller. Midas v3. 1–a model zoo for robust monocular relative depth estimation.arXiv preprint arXiv:2307.14460, 2023

    work page arXiv 2023

  8. [8]

    Black, Priyanka Patel, Joachim Tesch, and Jinlong Yang

    Michael J. Black, Priyanka Patel, Joachim Tesch, and Jinlong Yang. BEDLAM: A synthetic dataset of bodies exhibiting detailed lifelike animated motion. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 8726–8737, 2023

    2023

  9. [9]

    Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets

    Andreas Blattmann, Tim Dockhorn, Sumith Kulal, Daniel Mendelevitch, Maciej Kilian, Do- minik Lorenz, Yam Levi, Zion English, Vikram V oleti, Adam Letts, et al. Stable video diffusion: Scaling latent video diffusion models to large datasets.arXiv preprint arXiv:2311.15127, 2023

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  10. [10]

    Transformerfusion: Monocular rgb scene reconstruction using transformers.Advances in Neural Information Processing Systems, 34:1403–1414, 2021

    Aljaz Bozic, Pablo Palafox, Justus Thies, Angela Dai, and Matthias Nießner. Transformerfusion: Monocular rgb scene reconstruction using transformers.Advances in Neural Information Processing Systems, 34:1403–1414, 2021

    2021

  11. [11]

    Pix2video: Video editing using image diffusion

    Duygu Ceylan, Chun-Hao P Huang, and Niloy J Mitra. Pix2video: Video editing using image diffusion. InProceedings of the IEEE/CVF international conference on computer vision, pages 23206–23217, 2023

    2023

  12. [12]

    Video depth anything: Consistent depth estimation for super-long videos

    Sili Chen, Hengkai Guo, Shengnan Zhu, Feihu Zhang, Zilong Huang, Jiashi Feng, and Bingyi Kang. Video depth anything: Consistent depth estimation for super-long videos. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 22831–22840, 2025

    2025

  13. [13]

    Flashdepth: Real-time streaming video depth estimation at 2k resolution

    Gene Chou, Wenqi Xian, Guandao Yang, Mohamed Abdelfattah, Bharath Hariharan, Noah Snavely, Ning Yu, and Paul Debevec. Flashdepth: Real-time streaming video depth estimation at 2k resolution. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 9638–9648, 2025

    2025

  14. [14]

    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, et al. An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929, 2020

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  15. [15]

    Depth map prediction from a single image using a multi-scale deep network.Advances in Neural Information Processing Systems, 27, 2014

    David Eigen, Christian Puhrsch, and Rob Fergus. Depth map prediction from a single image using a multi-scale deep network.Advances in Neural Information Processing Systems, 27, 2014

    2014

  16. [16]

    Dens3r: A foundation model for 3d geometry prediction.arXiv preprint arXiv:2507.16290, 2025

    Xianze Fang, Jingnan Gao, Zhe Wang, Zhuo Chen, Xingyu Ren, Jiangjing Lyu, Qiaomu Ren, Zhonglei Yang, Xiaokang Yang, Yichao Yan, and Chengfei Lyu. Dens3r: A foundation model for 3d geometry prediction.arXiv preprint arXiv:2507.16290, 2025

    work page arXiv 2025

  17. [17]

    An instance-centric panoptic occupancy prediction benchmark for autonomous driving.arXiv preprint arXiv:2603.27238, 2026

    Yi Feng, Zizhan Guo, Yu Ma, Hanli Wang, Rui Fan, et al. An instance-centric panoptic occupancy prediction benchmark for autonomous driving.arXiv preprint arXiv:2603.27238, 2026

    work page arXiv 2026

  18. [18]

    Deep ordinal regression network for monocular depth estimation

    Huan Fu, Mingming Gong, Chaohui Wang, Kayhan Batmanghelich, and Dacheng Tao. Deep ordinal regression network for monocular depth estimation. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2002–2011, 2018

    2002

  19. [19]

    Geowizard: Unleashing the diffusion priors for 3d geometry estimation from a 20 single image

    Xiao Fu, Wei Yin, Mu Hu, Kaixuan Wang, Yuexin Ma, Ping Tan, Shaojie Shen, Dahua Lin, and Xiaoxiao Long. Geowizard: Unleashing the diffusion priors for 3d geometry estimation from a 20 single image. InEuropean Conference on Computer Vision, pages 241–258. Springer, 2024

    2024

  20. [20]

    More: 3d visual geometry reconstruction meets mixture-of- experts.arXiv preprint arXiv:2510.27234, 2025

    Jingnan Gao, Zhe Wang, Xianze Fang, Xingyu Ren, Zhuo Chen, Shengqi Liu, Yuhao Cheng, Jiangjing Lyu, Xiaokang Yang, and Yichao Yan. More: 3d visual geometry reconstruction meets mixture-of-experts.arXiv preprint arXiv:2510.27234, 2025

    work page arXiv 2025

  21. [21]

    Vision meets robotics: The kitti dataset.The International Journal of Robotics Research, 32(11):1231–1237, 2013

    Andreas Geiger, Philip Lenz, Christoph Stiller, and Raquel Urtasun. Vision meets robotics: The kitti dataset.The International Journal of Robotics Research, 32(11):1231–1237, 2013

    2013

  22. [22]

    Towards zero- shot scale-aware monocular depth estimation

    Vitor Guizilini, Igor Vasiljevic, Dian Chen, Rares, Ambrus, , and Adrien Gaidon. Towards zero- shot scale-aware monocular depth estimation. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 9233–9243, 2023

    2023

  23. [23]

    arXiv preprint arXiv:2409.18124 (2024)

    Jing He, Haodong Li, Wei Yin, Yixun Liang, Leheng Li, Kaiqiang Zhou, Hongbo Zhang, Bingbing Liu, and Ying-Cong Chen. Lotus: Diffusion-based visual foundation model for high-quality dense prediction.arXiv preprint arXiv:2409.18124, 2024

    work page arXiv 2024

  24. [24]

    Metric3d v2: A versatile monocular geomet- ric 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, Chunhua Shen, and Shaojie Shen. Metric3d v2: A versatile monocular geomet- ric foundation model for zero-shot metric depth and surface normal estimation.arXiv preprint arXiv:2404.15506, 2024

    work page arXiv 2024

  25. [25]

    Depthcrafter: Generating consistent long depth sequences for open-world videos.arXiv preprint arXiv:2409.02095, 2024

    Wenbo Hu, Xiangjun Gao, Xiaoyu Li, Sijie Zhao, Xiaodong Cun, Yong Zhang, Long Quan, and Ying Shan. Depthcrafter: Generating consistent long depth sequences for open-world videos. arXiv preprint arXiv:2409.02095, 2024

    work page arXiv 2024

  26. [26]

    Deepmvs: Learning multi-view stereopsis

    Po-Han Huang, Kevin Matzen, Johannes Kopf, Narendra Ahuja, and Jia-Bin Huang. Deepmvs: Learning multi-view stereopsis. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2821–2830, 2018

    2018

  27. [27]

    On the importance of accurate geometry data for dense 3d vision tasks

    HyunJun Jung, Patrick Ruhkamp, Guangyao Zhai, Nikolas Brasch, Yitong Li, Yannick Verdie, Jifei Song, Yiren Zhou, Anil Armagan, Slobodan Ilic, et al. On the importance of accurate geometry data for dense 3d vision tasks. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 780–791, 2023

    2023

  28. [28]

    Dynamicstereo: Consistent dynamic depth from stereo videos

    Nikita Karaev, Ignacio Rocco, Benjamin Graham, Natalia Neverova, Andrea Vedaldi, and Christian Rupprecht. Dynamicstereo: Consistent dynamic depth from stereo videos. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023

    2023

  29. [29]

    Repurposing diffusion-based image generators for monocular depth estimation

    Bingxin Ke, Anton Obukhov, Shengyu Huang, Nando Metzger, Rodrigo Caye Daudt, and Kon- rad Schindler. Repurposing diffusion-based image generators for monocular depth estimation. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9492–9502, 2024

    2024

  30. [30]

    arXiv preprint arXiv:2505.09358 (2025)

    B Ke, K Qu, T Wang, N Metzger, S Huang, B Li, A Obukhov, and K Schindler. Marigold: Affordable adaptation of diffusion-based image generators for image analysis.arXiv preprint arXiv:2505.09358, 2025

    work page arXiv 2025

  31. [31]

    Mapanything: Universal feed-forward metric 3d reconstruction

    Nikhil Varma Keetha, Norman Müller, Johannes Schönberger, Lorenzo Porzi, Yuchen Zhang, Tobias Fischer, Arno Knapitsch, Duncan Zauss, Ethan Weber, Nelson Antunes, Jonathon Luiten, Manuel Lopez-Antequera, Samuel Rota Bulò, Christian Richardt, Deva Ramanan, Sebastian Scherer, and Peter Kontschieder. Mapanything: Universal feed-forward metric 3d reconstructio...

    2026

  32. [32]

    STream3r: Scalable sequential 3d re- construction with causal transformer

    Yushi LAN, Yihang Luo, Fangzhou Hong, Shangchen Zhou, Honghua Chen, Zhaoyang Lyu, Bo Dai, Shuai Yang, Chen Change Loy, and Xingang Pan. STream3r: Scalable sequential 3d re- construction with causal transformer. InInternational Conference on Learning Representations, 2026

    2026

  33. [33]

    Grounding image matching in 3d with mast3r

    Vincent Leroy, Yohann Cabon, and Jérôme Revaud. Grounding image matching in 3d with mast3r. InEuropean Conference on Computer Vision, pages 71–91, 2024

    2024

  34. [34]

    Matrixcity: A large-scale city dataset for city-scale neural rendering and beyond

    Yixuan Li, Lihan Jiang, Linning Xu, Yuanbo Xiangli, Zhenzhi Wang, Dahua Lin, and Bo Dai. Matrixcity: A large-scale city dataset for city-scale neural rendering and beyond. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 3205–3215, 2023

    2023

  35. [35]

    Lightwheelocc: A 3d occupancy synthetic dataset in autonomous driving

    LightwheelAI and LightwheelOcc contributors. Lightwheelocc: A 3d occupancy synthetic dataset in autonomous driving. https://github.com/OpenDriveLab/LightwheelOcc, 2024

    2024

  36. [36]

    Chen, Zhenyu Li, Yang Zhao, Sida Peng, Hengkai Guo, Xiaowei Zhou, Guang Shi, Jiashi Feng, and Bingyi Kang

    Haotong Lin, Sili Chen, Jun Hao Liew, Donny Y . Chen, Zhenyu Li, Yang Zhao, Sida Peng, Hengkai Guo, Xiaowei Zhou, Guang Shi, Jiashi Feng, and Bingyi Kang. Depth anything 3: Recovering the visual space from any views. InInternational Conference on Learning Representations, 2026

    2026

  37. [37]

    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, et al. Dl3dv-10k: A large-scale scene dataset for deep learning-based 3d vision. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 22160–22169, 2024. 21

    2024

  38. [38]

    Geometry-aware 4D Video Generation for Robot Manipulation

    Zeyi Liu, Shuang Li, Eric Cousineau, Siyuan Feng, Benjamin Burchfiel, and Shuran Song. Geometry-aware 4d video generation for robot manipulation.arXiv preprint arXiv:2507.01099, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  39. [39]

    Align3r: Aligned monocular depth estimation for dynamic videos

    Jiahao Lu, Tianyu Huang, Peng Li, Zhiyang Dou, Cheng Lin, Zhiming Cui, Zhen Dong, Sai-Kit Yeung, Wenping Wang, and Yuan Liu. Align3r: Aligned monocular depth estimation for dynamic videos. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 22820–22830, 2025

    2025

  40. [40]

    Consistent video depth estimation.ACM Transactions on Graphics, 39(4):71–1, 2020

    Xuan Luo, Jia-Bin Huang, Richard Szeliski, Kevin Matzen, and Johannes Kopf. Consistent video depth estimation.ACM Transactions on Graphics, 39(4):71–1, 2020

    2020

  41. [41]

    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 Andrés Bruhn. Spring: A high-resolution high-detail dataset and benchmark for scene flow, optical flow and stereo. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023

    2023

  42. [42]

    DINOv2: Learning Robust Visual Features without Supervision

    Maxime Oquab, Timothée Darcet, Théo 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.arXiv preprint arXiv:2304.07193, 2023

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  43. [43]

    Refusion: 3d reconstruction in dynamic environments for rgb-d cameras exploiting residuals

    Emanuele Palazzolo, Jens Behley, Philipp Lottes, Philippe Giguere, and Cyrill Stachniss. Refusion: 3d reconstruction in dynamic environments for rgb-d cameras exploiting residuals. In IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 7855–7862, 2019

    2019

  44. [44]

    Aria digital twin: A new benchmark dataset for egocentric 3d machine perception

    Xiaqing Pan, Nicholas Charron, Yongqian Yang, Scott Peters, Thomas Whelan, Chen Kong, Omkar Parkhi, Richard Newcombe, and Yuheng (Carl) Ren. Aria digital twin: A new benchmark dataset for egocentric 3d machine perception. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 20133–20143, 2023

    2023

  45. [45]

    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), pages 20524–20531. IEEE, 2025

    2025

  46. [46]

    UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler

    Luigi Piccinelli, Christos Sakaridis, Yung-Hsu Yang, Mattia Segu, Siyuan Li, Wim Abbeloos, and Luc Van Gool. Unidepthv2: Universal monocular metric depth estimation made simpler. arXiv preprint arXiv:2502.20110, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  47. [47]

    Unidepth: Universal monocular metric depth estimation

    Luigi Piccinelli, Yung-Hsu Yang, Christos Sakaridis, Mattia Segu, Siyuan Li, Luc Van Gool, and Fisher Yu. Unidepth: Universal monocular metric depth estimation. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10106–10116, 2024

    2024

  48. [48]

    Xiaojuan Qi, Zhengzhe Liu, Renjie Liao, Philip HS Torr, Raquel Urtasun, and Jiaya Jia. Geonet++: Iterative geometric neural network with edge-aware refinement for joint depth and surface normal estimation.IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(2):969–984, 2020

    2020

  49. [49]

    Vision transformers for dense prediction

    René Ranftl, Alexey Bochkovskiy, and Vladlen Koltun. Vision transformers for dense prediction. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 12179– 12188, 2021

    2021

  50. [50]

    René 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, 44(3):1623–1637, 2020

    2020

  51. [51]

    Hypersim: A photorealistic synthetic dataset for holistic indoor scene understanding

    Mike Roberts, Jason Ramapuram, Anurag Ranjan, Atulit Kumar, Miguel Angel Bautista, Nathan Paczan, Russ Webb, and Joshua M Susskind. Hypersim: A photorealistic synthetic dataset for holistic indoor scene understanding. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 10912–10922, 2021

    2021

  52. [52]

    High- resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High- resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    2022

  53. [53]

    The synthia dataset: A large collection of synthetic images for semantic segmentation of urban scenes

    German Ros, Laura Sellart, Joanna Materzynska, David Vazquez, and Antonio M Lopez. The synthia dataset: A large collection of synthetic images for semantic segmentation of urban scenes. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 3234–3243, 2016

    2016

  54. [54]

    Structure-from-motion revisited

    Johannes L Schonberger and Jan-Michael Frahm. Structure-from-motion revisited. InProceed- ings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4104–4113, 2016

    2016

  55. [55]

    Learning temporally consistent video depth from video diffusion priors.arXiv preprint arXiv:2406.01493, 2024

    Jiahao Shao, Yuanbo Yang, Hongyu Zhou, Youmin Zhang, Yujun Shen, Matteo Poggi, and Yiyi Liao. Learning temporally consistent video depth from video diffusion priors.arXiv preprint arXiv:2406.01493, 2024

    work page arXiv 2024

  56. [56]

    Indoor segmentation and support inference from rgbd images

    Nathan Silberman, Derek Hoiem, Pushmeet Kohli, and Rob Fergus. Indoor segmentation and support inference from rgbd images. InProceedings of the European Conference on Computer 22 Vision, pages 746–760, 2012

    2012

  57. [57]

    Scalability in perception for autonomous driving: Waymo open dataset

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Yuning Chai, Benjamin Caine, Vijay Vasudevan, Wei Han, Jiquan Ngiam, Hang Zhao, Aleksei Timofeev, Scott Ettinger, Maxim Krivokon, Amy Gao, Aditya Joshi, Yu Zhang, Jonathon Shlens, Zhifeng Chen, and Dragomir Anguelov. Scalability in perception...

    2020

  58. [58]

    Masked depth modeling for spatial perception.arXiv preprint arXiv:[2601.17895], 2026

    Bin Tan, Changjiang Sun, Xiage Qin, Hanat Adai, Zelin Fu, Tianxiang Zhou, Han Zhang, Yinghao Xu, Xing Zhu, Yujun Shen, et al. Masked depth modeling for spatial perception.arXiv preprint arXiv:2601.17895, 2026

    work page arXiv 2026

  59. [59]

    Mv-dust3r+: Single-stage scene reconstruction from sparse views in 2 seconds

    Zhenggang Tang, Yuchen Fan, Dilin Wang, Hongyu Xu, Rakesh Ranjan, Alexander Schwing, and Zhicheng Yan. Mv-dust3r+: Single-stage scene reconstruction from sparse views in 2 seconds. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5283–5293, 2025

    2025

  60. [60]

    Droid-slam: Deep visual slam for monocular, stereo, and rgb-d cameras.Advances in neural information processing systems, 34:16558–16569, 2021

    Zachary Teed and Jia Deng. Droid-slam: Deep visual slam for monocular, stereo, and rgb-d cameras.Advances in neural information processing systems, 34:16558–16569, 2021

    2021

  61. [61]

    Depth from motion for smartphone ar.ACM Transactions on Graphics, 37(6):1–19, 2018

    Julien Valentin, Adarsh Kowdle, Jonathan T Barron, Neal Wadhwa, Max Dzitsiuk, Michael Schoenberg, Vivek Verma, Ambrus Csaszar, Eric Turner, Ivan Dryanovski, et al. Depth from motion for smartphone ar.ACM Transactions on Graphics, 37(6):1–19, 2018

    2018

  62. [62]

    3D Reconstruction with Spatial Memory

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

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  63. [63]

    Vggt: Visual geometry grounded transformer

    Jianyuan Wang, Minghao Chen, Nikita Karaev, Andrea Vedaldi, Christian Rupprecht, and David Novotny. Vggt: Visual geometry grounded transformer. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5294–5306, 2025

    2025

  64. [64]

    Vggsfm: Visual geometry grounded deep structure from motion

    Jianyuan Wang, Nikita Karaev, Christian Rupprecht, and David Novotny. Vggsfm: Visual geometry grounded deep structure from motion. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 21686–21697, 2024

    2024

  65. [65]

    From editor to dense geometry estimator.arXiv preprint arXiv:2509.04338, 2025

    JiYuan Wang, Chunyu Lin, Lei Sun, Rongying Liu, Lang Nie, Mingxing Li, Kang Liao, Xiangxiang Chu, and Yao Zhao. From editor to dense geometry estimator.arXiv preprint arXiv:2509.04338, 2025

    work page arXiv 2025

  66. [66]

    Flow-motion and depth network for monocular stereo and beyond.arXiv preprint arXiv:1909.05452, 2019

    Kaixuan Wang and Shaojie Shen. Flow-motion and depth network for monocular stereo and beyond.arXiv preprint arXiv:1909.05452, 2019

    work page arXiv 1909

  67. [67]

    Continuous 3d perception model with persistent state

    Qianqian Wang, Yifei Zhang, Aleksander Holynski, Alexei A Efros, and Angjoo Kanazawa. Continuous 3d perception model with persistent state. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10510–10522, 2025

    2025

  68. [68]

    Moge: Unlocking accurate monocular geometry estimation for open-domain images with optimal training supervision

    Ruicheng Wang, Sicheng Xu, Cassie Dai, Jianfeng Xiang, Yu Deng, Xin Tong, and Jiaolong Yang. Moge: Unlocking accurate monocular geometry estimation for open-domain images with optimal training supervision. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5261–5271, 2025

    2025

  69. [69]

    Moge-2: Accurate monocular geometry with metric scale and sharp details

    Ruicheng Wang, Sicheng Xu, Yue Dong, Yu Deng, Jianfeng Xiang, Zelong Lv, Guangzhong Sun, Xin Tong, and Jiaolong Yang. Moge-2: Accurate monocular geometry with metric scale and sharp details. InAdvances in Neural Information Processing Systems, 2025

    2025

  70. [70]

    Dust3r: Geometric 3d vision made easy

    Shuzhe Wang, Vincent Leroy, Yohann Cabon, Boris Chidlovskii, and Jérôme Revaud. Dust3r: Geometric 3d vision made easy. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 20697–20709, 2024

    2024

  71. [71]

    Tartanair: A dataset to push the limits of visual slam

    Wenshan Wang, Delong Zhu, Xiangwei Wang, Yaoyu Hu, Yuheng Qiu, Chen Wang, Yafei Hu, Ashish Kapoor, and Sebastian Scherer. Tartanair: A dataset to push the limits of visual slam. In 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 4909–4916, 2020

    2020

  72. [72]

    Neural video depth stabilizer

    Yiran Wang, Min Shi, Jiaqi Li, Zihao Huang, Zhiguo Cao, Jianming Zhang, Ke Xian, and Guosheng Lin. Neural video depth stabilizer. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 9466–9476, 2023

    2023

  73. [73]

    π3: Permutation-equivariant visual geometry learning

    Yifan Wang, Jianjun Zhou, Haoyi Zhu, Wenzheng Chang, Yang Zhou, Zizun Li, Junyi Chen, Jiangmiao Pang, Chunhua Shen, and Tong He. π3: Permutation-equivariant visual geometry learning. InInternational Conference on Learning Representations, 2026

    2026

  74. [74]

    Omniobject3d: Large-vocabulary 3d object dataset for realistic perception, reconstruction and generation

    Tong Wu, Jiarui Zhang, Xiao Fu, Yuxin Wang, Jiawei Ren, Liang Pan, Wayne Wu, Lei Yang, Jiaqi Wang, Chen Qian, Dahua Lin, and Ziwei Liu. Omniobject3d: Large-vocabulary 3d object dataset for realistic perception, reconstruction and generation. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023

    2023

  75. [75]

    Rgbd objects in the wild: Scaling real-world 3d object learning from rgb-d videos

    Hongchi Xia, Yang Fu, Sifei Liu, and Xiaolong Wang. Rgbd objects in the wild: Scaling real-world 3d object learning from rgb-d videos. InProceedings of the IEEE/CVF conference 23 on computer vision and pattern recognition, pages 22378–22389, 2024

    2024

  76. [76]

    Diffusion knows transparency: Repurposing video diffusion for transparent object depth and normal estimation.arXiv preprint arXiv:2512.23705, 2025

    Shaocong Xu, Songlin Wei, Qizhe Wei, Zheng Geng, Hong Li, Licheng Shen, Qianpu Sun, Shu Han, Bin Ma, Bohan Li, Chongjie Ye, Yuhang Zheng, Nan Wang, Saining Zhang, and Hao Zhao. Diffusion knows transparency: Repurposing video diffusion for transparent object depth and normal estimation.arXiv preprint arXiv:2512.23705, 2025

    work page arXiv 2025

  77. [77]

    Ge- ometrycrafter: Consistent geometry estimation for open-world videos with diffusion priors

    Tian-Xing Xu, Xiangjun Gao, Wenbo Hu, Xiaoyu Li, Song-Hai Zhang, and Ying Shan. Ge- ometrycrafter: Consistent geometry estimation for open-world videos with diffusion priors. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6632–6644, 2025

    2025

  78. [78]

    Depth any video with scalable synthetic data.arXiv preprint arXiv:2410.10815, 2024

    Honghui Yang, Di Huang, Wei Yin, Chunhua Shen, Haifeng Liu, Xiaofei He, Binbin Lin, Wanli Ouyang, and Tong He. Depth any video with scalable synthetic data.arXiv preprint arXiv:2410.10815, 2024

    work page arXiv 2024

  79. [79]

    Fast3r: Towards 3d reconstruction of 1000+ images in one forward pass

    Jianing Yang, Alexander Sax, Kevin J Liang, Mikael Henaff, Hao Tang, Ang Cao, Joyce Chai, Franziska Meier, and Matt Feiszli. Fast3r: Towards 3d reconstruction of 1000+ images in one forward pass. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 21924–21935, 2025

    2025

  80. [80]

    Depth anything: Unleashing the power of large-scale unlabeled data

    Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, and Hengshuang Zhao. Depth anything: Unleashing the power of large-scale unlabeled data. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2024

    2024

Showing first 80 references.