pith. sign in

arxiv: 2606.23298 · v1 · pith:ZVDXMBZSnew · submitted 2026-06-22 · 💻 cs.CV

Ocean4D: Generative Underwater 4D Reconstruction via Medium-Aware Video Diffusion

Yuqiang Huang , Yuxi Wang , Junyu Dong , Zhaoxiang Zhang This is my paper

Pith reviewed 2026-06-26 09:22 UTC · model grok-4.3

classification 💻 cs.CV
keywords underwater 4D reconstructionvideo diffusiongenerative modelmedium-aware denoisingcross-view consistencymonocular videoparticipating mediageometric conditioning
0
0 comments X

The pith

A generative video diffusion model produces consistent 4D underwater reconstructions from monocular input by building geometric conditions and handling medium effects implicitly.

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

The paper seeks to establish that underwater 4D reconstruction can be achieved through a diffusion-based generative process that accounts for light absorption, scattering, and dynamic particle motion. It introduces 4D-GCC to create geometrically consistent conditioning across frames and a Medium-Aware Block to adjust denoising for underwater light transport without separate physics equations. A reader would care because prior methods assume clear air conditions and static scenes, which causes drifting geometry and view mismatches when applied underwater. If the approach works, monocular video alone becomes sufficient to generate new camera trajectories with preserved global structure.

Core claim

Ocean4D builds 4D geometrically consistent conditioning with improved cross-frame coverage through 4D-GCC and performs implicit medium-aware denoising via the Medium-Aware Block inside the latent diffusion process, allowing generation of videos along target camera trajectories that preserve global structure and cross-view consistency despite absorption, backscatter, and dynamic water variations.

What carries the argument

The Medium-Aware Block, which performs implicit medium-aware denoising during latent diffusion to stabilize appearance under absorption and scattering.

If this is right

  • The method generates videos along target trajectories while preserving global structure and cross-view consistency.
  • It achieves state-of-the-art results on both dynamic and static underwater benchmarks.
  • It avoids sensitivity to drifting particles and dynamic distractors by dropping near-static assumptions.
  • No explicit physical modeling or additional medium inputs are required.

Where Pith is reading between the lines

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

  • The same implicit denoising strategy could extend to reconstruction in other participating media such as fog without redesigning the model.
  • Underwater robot mapping tasks might rely on this method using only forward-facing video rather than calibrated multi-camera rigs.
  • Longer sequences or higher turbidity levels could be tested to check whether the geometric conditioning remains stable over extended time.

Load-bearing premise

The Medium-Aware Block can implicitly handle absorption and scattering effects in denoising without needing explicit physical models or extra inputs.

What would settle it

Generated output videos that exhibit mismatched colors, structures, or drifting particles when rendered from different target camera paths under strong scattering conditions would falsify the consistency claim.

Figures

Figures reproduced from arXiv: 2606.23298 by Junyu Dong, Yuqiang Huang, Yuxi Wang, Zhaoxiang Zhang.

Figure 1
Figure 1. Figure 1: Ocean4D addresses the problem of underwater 4D reconstruction from a monocular video. Ocean4D constructs 4D geometrically consistent conditions and per￾forms medium-aware generation with a video diffusion model, improving cross-view consistency and visual stability under dynamic underwater media. representations from casually captured monocular videos, which would enable reliable visualization, analysis, a… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of Ocean4D. Given a monocular source video with camera poses, our method generates multi-view videos under target poses using a video diffusion model. We first apply 4D-GCC to extract a 4D geometric representation and con￾struct a geometry-consistent conditioning video by combining time-varying and time￾invariant renderings to enrich cross-frame information. The resulting conditioning to￾kens, tog… view at source ↗
Figure 3
Figure 3. Figure 3: Medium-Aware Block architecture. The Medium-Aware Block takes the conditioning tokens and reference tokens as inputs and first produces a medium to￾ken via cross-attention. Three lightweight adaptors map this token to medium-related parameters {ϕ a , ϕb , ϕ∞}. Depth-dependent components D dir and D back are injected through depth adaptors to modulate the direct and backscatter pathways. Two cross￾attention… view at source ↗
Figure 4
Figure 4. Figure 4: Qualitative results on dynamic underwater scenes comparing our method with TrajectoryCrafter [64] and ReCamMaster [4]. Red arrows highlight typical failure modes under novel views, including semantic misrecognition and hallucinations caused by insufficient geometry [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative comparison on NUSR. Red boxes highlight distractor regions with zoomed-in views. Our method achieves better reconstruction quality in both static regions and distractor regions [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative comparison on DRUVA. Red boxes highlight detail regions that are strongly affected by water disturbances, including the flag in A11 and the rope in A12 [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Hallucination analysis on dynamic scenes. [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Ablation visualization. Visual comparisons on static scenes, including Groundtruth, w/o Medium-Aware Block, w/o 4D-GCC, and w/o Medium-Aware Block & 4D-GCC. Red boxes highlight differences in reef textures and fine details [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
read the original abstract

Underwater 4D reconstruction remains challenging due to the coupling between degraded light transport in participating media and dynamic water variations. Most existing Methods are developed under in-air assumptions and do not explicitly account for underwater absorption and backscatter. Additionally, near-static assumptions make these approaches sensitive to drifting particles and dynamic distractors , leading to unstable geometry and inconsistent cross-view results. To address these issues, we propose a generative framework for underwater 4D reconstruction, named Ocean4D, which is built on two complementary components. Specifically, 4D-GCC constructs 4D geometrically consistent conditioning with improved cross-frame coverage, while the Medium-Aware Block performs implicit medium-aware denoising in the latent diffusion process to stabilize underwater appearance under absorption and scattering. Given a monocular video and target cameras, our method generates videos along the target trajectories while preserving global structure and cross-view consistency. Extensive experiments on both dynamic and static underwater benchmarks demonstrate state-of-the-art performance on underwater reconstruction.

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

0 major / 2 minor

Summary. The paper introduces Ocean4D, a generative framework for underwater 4D reconstruction from monocular video. It consists of two main components: 4D-GCC, which constructs 4D geometrically consistent conditioning with improved cross-frame coverage, and the Medium-Aware Block, which performs implicit medium-aware denoising in the latent diffusion process to stabilize underwater appearance under absorption and scattering. Given a monocular video and target cameras, the method generates videos along target trajectories while preserving global structure and cross-view consistency, and claims state-of-the-art performance on both dynamic and static underwater benchmarks.

Significance. If the central claims hold, the work could meaningfully advance underwater computer vision by demonstrating that video diffusion models can implicitly handle participating media effects and dynamic distractors without explicit physical modeling or additional inputs. This implicit stabilization approach, if effective, would be a useful design pattern for other challenging environments where light transport is degraded.

minor comments (2)
  1. The abstract contains a capitalization inconsistency: 'Most existing Methods' should read 'methods'.
  2. The abstract contains a typographical error with an extraneous space before the comma: 'distractors , leading'.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their summary of Ocean4D and for noting its potential to advance underwater computer vision through implicit handling of participating media. The recommendation is uncertain, yet the report contains no enumerated major comments. We therefore provide a general response below and stand ready to address any specific concerns the referee may wish to raise.

Circularity Check

0 steps flagged

No significant circularity in claimed derivation chain

full rationale

The abstract and description present Ocean4D as a proposed generative framework whose two components (4D-GCC and Medium-Aware Block) are introduced as architectural design choices to handle underwater media effects and consistency. No equations, fitted parameters, predictions, uniqueness theorems, or self-citations appear that would reduce any result to its own inputs by construction. The method is described at the level of high-level components and empirical performance claims rather than a derivation chain that collapses into self-definition or renamed fits. This is the common case of a self-contained engineering proposal without load-bearing circular steps.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No details available from abstract to identify specific free parameters, axioms, or invented entities; the framework relies on standard diffusion model assumptions and underwater medium effects as background.

pith-pipeline@v0.9.1-grok · 5705 in / 1079 out tokens · 19675 ms · 2026-06-26T09:22:50.830968+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

69 extracted references · 3 linked inside Pith

  1. [1]

    In: Proc

    Akkaynak, D., Treibitz, T.: A revised underwater image formation model. In: Proc. CVPR. pp. 6723–6732 (2018) 2, 4

    2018

  2. [2]

    In: Proc

    Akkaynak, D., Treibitz, T.: Sea-thru: A method for removing water from under- water images. In: Proc. CVPR. pp. 1682–1691 (2019) 2

    2019

  3. [3]

    arXiv preprint arXiv:2001.00330 (2020) 1

    Arain, B., Dayoub, F., Rigby, P., Dunbabin, M.: Close-proximity underwater terrain mapping using learning-based coarse range estimation. arXiv preprint arXiv:2001.00330 (2020) 1

    arXiv 2001

  4. [4]

    In: Proc

    Bai, J., Xia, M., Fu, X.,Wang, X.,Mu, L., Cao, J.,Liu, Z., Hu, H.,Bai, X., Wan, P., et al.: Recammaster: Camera-controlled generative rendering from a single video. In: Proc. ICCV. pp. 14834–14844 (2025) 2, 4, 9, 10, 12

    2025

  5. [5]

    In: Proc

    Barron, J.T., Mildenhall, B., Tancik, M., Hedman, P., Martin-Brualla, R., Srini- vasan, P.P.: Mip-nerf: A multiscale representation for anti-aliasing neural radiance fields. In: Proc. ICCV. pp. 5855–5864 (2021) 2, 3

    2021

  6. [6]

    In: Proc

    Barron, J.T., Mildenhall, B., Verbin, D., Srinivasan, P.P., Hedman, P.: Mip-nerf 360: Unbounded anti-aliased neural radiance fields. In: Proc. CVPR. pp. 5470–5479 (2022) 2, 3

    2022

  7. [7]

    In: Proc

    Barron, J.T., Mildenhall, B., Verbin, D., Srinivasan, P.P., Hedman, P.: Zip-nerf: Anti-aliased grid-based neural radiance fields. In: Proc. ICCV. pp. 19697–19705 (2023) 2, 3

    2023

  8. [8]

    arXiv preprint arXiv:2311.15127 (2023) 2, 4, 5

    Blattmann, A., Dockhorn, T., Kulal, S., Mendelevitch, D., Kilian, M., Lorenz, D., Levi, Y., English, Z., Voleti, V., Letts, A., et al.: Stable video diffusion: Scaling latent video diffusion models to large datasets. arXiv preprint arXiv:2311.15127 (2023) 2, 4, 5

    Pith/arXiv arXiv 2023

  9. [9]

    In: Proc

    Blattmann, A., Rombach, R., Ling, H., Dockhorn, T., Kim, S.W., Fidler, S., Kreis, K.: Align your latents: High-resolution video synthesis with latent diffusion models. In: Proc. CVPR. pp. 22563–22575 (2023) 2, 5

    2023

  10. [10]

    In: Proc

    Chen, A., Xu, Z., Zhao, F., Zhang, X., Xiang, F., Yu, J., Su, H.: Mvsnerf: Fast generalizable radiance field reconstruction from multi-view stereo. In: Proc. ICCV. pp. 14124–14133 (2021) 2, 3

    2021

  11. [11]

    arXiv preprint arXiv:2310.19512 (2023) 4, 5

    Chen, H., Xia, M., He, Y., Zhang, Y., Cun, X., Yang, S., Xing, J., Liu, Y., Chen, Q., Wang, X., et al.: Videocrafter1: Open diffusion models for high-quality video generation. arXiv preprint arXiv:2310.19512 (2023) 4, 5

    Pith/arXiv arXiv 2023

  12. [12]

    Computers & Graphics 123, 104025 (2024) 4

    Chen, L., Xiong, Y., Zhang, Y., Yu, R., Fang, L., Liu, D.: Sp-seanerf: Underwater neural radiance fields with strong scattering perception. Computers & Graphics 123, 104025 (2024) 4

    2024

  13. [13]

    In: Proc

    Chen, Y., Xu, H., Zheng, C., Zhuang, B., Pollefeys, M., Geiger, A., Cham, T.J., Cai, J.: Mvsplat: Efficient 3d gaussian splatting from sparse multi-view images. In: Proc. ECCV. pp. 370–386. Springer (2024) 2, 3

    2024

  14. [14]

    Symmetry16(8), 1010 (2024) 2, 4 16 Y

    Du, Y., Zhang, Z., Zhang, P., Sun, F., Lv, X.: Udr-gs: Enhancing underwater dynamic scene reconstruction with depth regularization. Symmetry16(8), 1010 (2024) 2, 4 16 Y. Huang et al

    2024

  15. [15]

    In: proc

    Duan, Y., Wei, F., Dai, Q., He, Y., Chen, W., Chen, B.: 4d-rotor gaussian splat- ting: towards efficient novel view synthesis for dynamic scenes. In: proc. ACM SIGGRAPH. pp. 1–11 (2024) 3

    2024

  16. [16]

    arXiv preprint arXiv:2403.20309 (2024) 2, 3

    Fan, Z., Cong, W., Wen, K., Wang, K., Zhang, J., Ding, X., Xu, D., Ivanovic, B., Pavone, M., Pavlakos, G., et al.: Instantsplat: Sparse-view gaussian splatting in seconds. arXiv preprint arXiv:2403.20309 (2024) 2, 3

    arXiv 2024

  17. [17]

    In: Proc

    Feng, C., Yu, W., Cheng, X., Tang, Z., Zhang, J., Yuan, L., Tian, Y.: Ae-nerf: Augmenting event-based neural radiance fields for non-ideal conditions and larger scenes. In: Proc. AAAI. vol. 39, pp. 2924–2932 (2025) 3

    2025

  18. [18]

    In: Proc

    Fu, X., Zhuang, P., Huang, Y., Liao, Y., Zhang, X.P., Ding, X.: A retinex-based enhancing approach for single underwater image. In: Proc. ICIP. pp. 4572–4576. IEEE (2014) 3

    2014

  19. [19]

    Advances in Neural Information Processing Systems (2024) 4

    Gao*, R., Holynski*, A., Henzler, P., Brussee, A., Martin-Brualla, R., Srinivasan, P.P., Barron, J.T., Poole*, B.: Cat3d: Create anything in 3d with multi-view dif- fusion models. Advances in Neural Information Processing Systems (2024) 4

    2024

  20. [20]

    International Journal of Naval Ar- chitecture and Ocean Engineering6(4), 840–866 (2014) 3

    Ghani, A.S.A., Isa, N.A.M.: Underwater image quality enhancement through rayleigh-stretching and averaging image planes. International Journal of Naval Ar- chitecture and Ocean Engineering6(4), 840–866 (2014) 3

    2014

  21. [21]

    Sensors16(8), 1174 (2016) 1

    Hernández, J.D., Istenič, K., Gracias, N., Palomeras, N., Campos, R., Vidal, E., Garcia, R., Carreras, M.: Autonomous underwater navigation and optical mapping in unknown natural environments. Sensors16(8), 1174 (2016) 1

    2016

  22. [22]

    PloS one15(3), e0230671 (2020) 1

    Hopkinson, B.M., King, A.C., Owen, D.P., Johnson-Roberson, M., Long, M.H., Bhandarkar, S.M.: Automated classification of three-dimensional reconstructions of coral reefs using convolutional neural networks. PloS one15(3), e0230671 (2020) 1

    2020

  23. [23]

    Journal of Marine Science and Engineering11(5), 949 (2023) 2

    Hu, K., Wang, T., Shen, C., Weng, C., Zhou, F., Xia, M., Weng, L.: Overview of underwater 3d reconstruction technology based on optical images. Journal of Marine Science and Engineering11(5), 949 (2023) 2

    2023

  24. [24]

    In: Proc

    Hu, W., Gao, X., Li, X., Zhao, S., Cun, X., Zhang, Y., Quan, L., Shan, Y.: Depthcrafter: Generating consistent long depth sequences for open-world videos. In: Proc. CVPR. pp. 2005–2015 (2025) 5, 10

    2005

  25. [25]

    In: Proc

    Huang, Z., He, Y., Yu, J., Zhang, F., Si, C., Jiang, Y., Zhang, Y., Wu, T., Jin, Q., Chanpaisit, N., et al.: Vbench: Comprehensive benchmark suite for video genera- tive models. In: Proc. CVPR. pp. 21807–21818 (2024) 10, 11

    2024

  26. [26]

    In: 2010 IEEE international conference on systems, man and cybernetics

    Iqbal, K., Odetayo, M., James, A., Salam, R.A., Talib, A.Z.H.: Enhancing the low quality images using unsupervised colour correction method. In: 2010 IEEE international conference on systems, man and cybernetics. pp. 1703–1709. IEEE (2010) 3

    2010

  27. [27]

    ACM Transactions on Graphics44(6), 1–16 (2025) 3, 9, 11, 12, 13

    Jiang, L., Mao, Y., Xu, L., Lu, T., Ren, K., Jin, Y., Xu, X., Yu, M., Pang, J., Zhao, F., et al.: Anysplat: Feed-forward 3d gaussian splatting from unconstrained views. ACM Transactions on Graphics44(6), 1–16 (2025) 3, 9, 11, 12, 13

    2025

  28. [28]

    ACM Trans

    Kerbl, B., Kopanas, G., Leimkühler, T., Drettakis, G., et al.: 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph.42(4), 139–1 (2023) 2, 3, 9, 11, 12

    2023

  29. [29]

    In: Proc

    Levy, D., Peleg, A., Pearl, N., Rosenbaum, D., Akkaynak, D., Korman, S., Treibitz, T.: Seathru-nerf: Neural radiance fields in scattering media. In: Proc. CVPR. pp. 56–65 (2023) 3

    2023

  30. [30]

    In: Proc

    Li, H., Song, W., Xu, T., Elsig, A., Kulhanek, J.: Watersplatting: Fast underwater 3d scene reconstruction using gaussian splatting. In: Proc. 3DV. pp. 969–978. IEEE (2025) 2, 4, 9, 11, 12, 13 Ocean4D: Generative Underwater 4D Reconstruction 17

    2025

  31. [31]

    In: Proc

    Li, J., Li, D., Xiong, C., Hoi, S.: Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation. In: Proc. ICML. pp. 12888–12900. PMLR (2022) 7

    2022

  32. [32]

    Springer (2024) 2, 3

    Liang, Z., Zhang, Q., Hu, W., Zhu, L., Feng, Y., Jia, K.: Analytic-splatting: Anti- aliased3dgaussiansplattingviaanalyticintegration.In:Proc.ECCV.pp.281–297. Springer (2024) 2, 3

    2024

  33. [33]

    arXiv preprint arXiv:2412.00131 (2024) 4

    Lin, B., Ge, Y., Cheng, X., Li, Z., Zhu, B., Wang, S., He, X., Ye, Y., Yuan, S., Chen, L., et al.: Open-sora plan: Open-source large video generation model. arXiv preprint arXiv:2412.00131 (2024) 4

    Pith/arXiv arXiv 2024

  34. [34]

    In: Proc

    Lin, C.H., Ma, W.C., Torralba, A., Lucey, S.: Barf: Bundle-adjusting neural radi- ance fields. In: Proc. ICCV. pp. 5741–5751 (2021) 3

    2021

  35. [35]

    In: Proc

    Liu, R., Wu, R., Van Hoorick, B., Tokmakov, P., Zakharov, S., Vondrick, C.: Zero- 1-to-3: Zero-shot one image to 3d object. In: Proc. ICCV. pp. 9298–9309 (2023) 4

    2023

  36. [36]

    arXiv preprint arXiv:2411.00239 (2024) 2, 4

    Liu, S., Lu, J., Gu, Z., Li, J., Deng, Y.: Aquatic-gs: A hybrid 3d representation for underwater scenes. arXiv preprint arXiv:2411.00239 (2024) 2, 4

    arXiv 2024

  37. [37]

    Commu- nications of the ACM65(1), 99–106 (2021) 2

    Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., Ng, R.: Nerf: Representing scenes as neural radiance fields for view synthesis. Commu- nications of the ACM65(1), 99–106 (2021) 2

    2021

  38. [38]

    In: Proc

    Müller, N., Schwarz, K., Rössle, B., Porzi, L., Bulo, S.R., Nießner, M., Kontschieder, P.: Multidiff: Consistent novel view synthesis from a single image. In: Proc. CVPR. pp. 10258–10268 (2024) 4

    2024

  39. [39]

    Frontiers in Environmental Science10, 1044706 (2022) 1

    de Oliveira, L.M.C., Lim, A., Conti, L.A., Wheeler, A.J.: High-resolution 3d map- ping of cold-water coral reefs using machine learning. Frontiers in Environmental Science10, 1044706 (2022) 1

    2022

  40. [40]

    In: Proc

    Park, K., Sinha, U., Barron, J.T., Bouaziz, S., Goldman, D.B., Seitz, S.M., Martin- Brualla, R.: Nerfies: Deformable neural radiance fields. In: Proc. ICCV. pp. 5865– 5874 (2021) 3

    2021

  41. [41]

    In: Proc

    Pumarola, A., Corona, E., Pons-Moll, G., Moreno-Noguer, F.: D-nerf: Neural ra- diance fields for dynamic scenes. In: Proc. CVPR. pp. 10318–10327 (2021) 3

    2021

  42. [42]

    Journal of Machine Learning Research21(140), 1–67 (2020) 7

    Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., Liu, P.J.: Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research21(140), 1–67 (2020) 7

    2020

  43. [43]

    In: Proc

    Ren,X.,Shen,T.,Huang,J.,Ling,H.,Lu,Y.,Nimier-David,M.,Müller,T.,Keller, A., Fidler, S., Gao, J.: Gen3c: 3d-informed world-consistent video generation with precise camera control. In: Proc. CVPR. pp. 6121–6132 (2025) 2, 4

    2025

  44. [44]

    In: Proc

    Sargent, K., Li, Z., Shah, T., Herrmann, C., Yu, H.X., Zhang, Y., Chan, E.R., Lagun, D., Fei-Fei, L., Sun, D., et al.: Zeronvs: Zero-shot 360-degree view synthesis from a single image. In: Proc. CVPR. pp. 9420–9429 (2024) 4

    2024

  45. [45]

    In: OCEANS 2023-MTS/IEEE US Gulf Coast

    Sethuraman, A.V., Ramanagopal, M.S., Skinner, K.A.: Waternerf: Neural radiance fields for underwater scenes. In: OCEANS 2023-MTS/IEEE US Gulf Coast. pp. 1–

    2023

  46. [46]

    In: Proc

    Song, J., Meng, C., Ermon, S.: Denoising diffusion implicit models. In: Proc. ICLR (2020) 4

    2020

  47. [47]

    IEEE Transactions on Visualization and Computer Graphics29(5), 2732–2742 (2023) 3

    Song, L., Chen, A., Li, Z., Chen, Z., Chen, L., Yuan, J., Xu, Y., Geiger, A.: Nerf- player: A streamable dynamic scene representation with decomposed neural radi- ance fields. IEEE Transactions on Visualization and Computer Graphics29(5), 2732–2742 (2023) 3

    2023

  48. [48]

    arXiv preprint arXiv:2601.09499 (2026) 6, 9 18 Y

    Sucar, E., Insafutdinov, E., Lai, Z., Vedaldi, A.: V-dpm: 4d video reconstruction with dynamic point maps. arXiv preprint arXiv:2601.09499 (2026) 6, 9 18 Y. Huang et al

    arXiv 2026

  49. [49]

    In: Proc

    Tang, Y., Zhu, C., Wan, R., Xu, C., Shi, B.: Neural underwater scene representa- tion. In: Proc. CVPR. pp. 11780–11789 (2024) 3, 9, 12, 13

    2024

  50. [50]

    In: Proc

    Van Hoorick, B., Wu, R., Ozguroglu, E., Sargent, K., Liu, R., Tokmakov, P., Dave, A., Zheng, C., Vondrick, C.: Generative camera dolly: Extreme monocular dynamic novel view synthesis. In: Proc. ECCV. pp. 313–331. Springer (2024) 2, 4, 5

    2024

  51. [51]

    In: Proc

    Varghese, N., Kumar, A., Rajagopalan, A.: Self-supervised monocular underwater depth recovery, image restoration, and a real-sea video dataset. In: Proc. ICCV. pp. 12248–12258 (2023) 9, 12, 13

    2023

  52. [52]

    IEEE Transactions on Pattern Analysis and Machine Intelligence47(11), 9426–9437 (2024) 3

    Verbin, D., Hedman, P., Mildenhall, B., Zickler, T., Barron, J.T., Srinivasan, P.P.: Ref-nerf: Structured view-dependent appearance for neural radiance fields. IEEE Transactions on Pattern Analysis and Machine Intelligence47(11), 9426–9437 (2024) 3

    2024

  53. [53]

    In: Proc

    Wang, H., Agapito, L.: 3d reconstruction with spatial memory. In: Proc. 3DV. pp. 78–89. IEEE (2025) 3

    2025

  54. [54]

    In: Proc

    Wang, J., Chen, M., Karaev, N., Vedaldi, A., Rupprecht, C., Novotny, D.: Vggt: Visual geometry grounded transformer. In: Proc. CVPR. pp. 5294–5306 (2025) 3, 9

    2025

  55. [55]

    In: Proc

    Wang, Q., Zhang, Y., Holynski, A., Efros, A.A., Kanazawa, A.: Continuous 3d perception model with persistent state. In: Proc. CVPR. pp. 10510–10522 (2025) 3

    2025

  56. [56]

    In: Proc

    Wang, S., Leroy, V., Cabon, Y., Chidlovskii, B., Revaud, J.: Dust3r: Geometric 3d vision made easy. In: Proc. CVPR. pp. 20697–20709 (2024) 3

    2024

  57. [57]

    In: Proc

    Wu, G., Yi, T., Fang, J., Xie, L., Zhang, X., Wei, W., Liu, W., Tian, Q., Wang, X.: 4d gaussian splatting for real-time dynamic scene rendering. In: Proc. CVPR. pp. 20310–20320 (2024) 3

    2024

  58. [58]

    In: Proc

    Wu, R., Gao, R., Poole, B., Trevithick, A., Zheng, C., Barron, J.T., Holynski, A.: Cat4d: Create anything in 4d with multi-view video diffusion models. In: Proc. CVPR. pp. 26057–26068 (2025) 4

    2025

  59. [59]

    In: Proc

    Wu, R., Mildenhall, B., Henzler, P., Park, K., Gao, R., Watson, D., Srinivasan, P.P., Verbin, D., Barron, J.T., Poole, B., et al.: Reconfusion: 3d reconstruction with diffusion priors. In: Proc. CVPR. pp. 21551–21561 (2024) 4

    2024

  60. [60]

    In: Proc

    Xie, Y., Kong, L., Chen, K., Zheng, Z., Yu, X., Yu, Z., Zheng, B.: Uveb: A large- scale benchmark and baseline towards real-world underwater video enhancement. In: Proc. CVPR. pp. 22358–22367 (2024) 9, 10

    2024

  61. [61]

    In: Proc

    Yang, D., Leonard, J.J., Girdhar, Y.: Seasplat: Representing underwater scenes with 3d gaussian splatting and a physically grounded image formation model. In: Proc. ICRA. pp. 7632–7638. IEEE (2025) 2, 4

    2025

  62. [62]

    In: Proc

    Yang, Z., Teng, J., Zheng, W., Ding, M., Huang, S., Xu, J., Yang, Y., Hong, W., Zhang, X., Feng, G., et al.: Cogvideox: Text-to-video diffusion models with an expert transformer. In: Proc. ICLR (2024) 4, 5, 7

    2024

  63. [63]

    Advancesin Neural Information ProcessingSystems34,4805–4815(2021) 2

    Yariv, L., Gu, J., Kasten, Y., Lipman, Y.: Volume rendering of neural implicit surfaces. Advancesin Neural Information ProcessingSystems34,4805–4815(2021) 2

    2021

  64. [64]

    In: Proc

    Yu, M., Hu, W., Xing, J., Shan, Y.: Trajectorycrafter: Redirecting camera trajec- tory for monocular videos via diffusion models. In: Proc. ICCV. pp. 100–111 (2025) 2, 4, 5, 6, 7, 9, 10, 11, 12, 13

    2025

  65. [65]

    In: Proc

    Yu, W., Fan, Y., Zhang, Y., Wang, X., Yin, F., Bai, Y., Cao, Y.P., Shan, Y., Wu, Y., Sun, Z., et al.: Nofa: Nerf-based one-shot facial avatar reconstruction. In: Proc. ACM SIGGRAPH. pp. 1–12 (2023) 3 Ocean4D: Generative Underwater 4D Reconstruction 19

    2023

  66. [66]

    IEEE Transactions on Pattern Analysis and Machine Intelligence (2025) 4

    Yu, W., Xing, J., Yuan, L., Hu, W., Li, X., Huang, Z., Gao, X., Wong, T.T., Shan, Y., Tian, Y.: Viewcrafter: Taming video diffusion models for high-fidelity novel view synthesis. IEEE Transactions on Pattern Analysis and Machine Intelligence (2025) 4

    2025

  67. [67]

    In: Proc

    Zhang, J., Herrmann, C., Hur, J., Jampani, V., Darrell, T., Cole, F., Sun, D., Yang, M.H.: Monst3r: A simple approach for estimating geometry in the presence of motion. In: Proc. ICLR (2024) 3

    2024

  68. [68]

    Neurocomputing245, 1–9 (2017) 3

    Zhang, S., Wang, T., Dong, J., Yu, H.: Underwater image enhancement via ex- tended multi-scale retinex. Neurocomputing245, 1–9 (2017) 3

    2017

  69. [69]

    In: Proc

    Zhu, Z., Fan, Z., Jiang, Y., Wang, Z.: Fsgs: Real-time few-shot view synthesis using gaussian splatting. In: Proc. ECCV. pp. 145–163. Springer (2024) 3

    2024