Pith. sign in

REVIEW 4 major objections 7 minor 51 references

WAVE: Warp-Based View Guidance for Consistent Novel View Synthesis Using a Single Image

T0 review · 4 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read WAVE is a training-free method that improves view consistency in single-image novel view synthesis by warping the input view and using the warps to guide attention and reinitialize noise.

desk verdict A useful training-free consistency booster for diffusion NVS, with real gains on ground-truth and 3D-reconstruction metrics, but the self-introduced COLMAP/LPIPS evaluation needs transparency fixes before the headline numbers can be trusted. read the letter →

arxiv 2506.23518 v2 pith:TZDQ3GWU submitted 2025-06-30 cs.CV

Jiwoo Park , Tae Eun Choi , Youngjun Jun , Seong Jae Hwang This is my paper
classification cs.CV
keywords novelviewsynthesisconsistencydiffusionmodelsattentionmaskingnoiseinitialization3Dwarpingtraining-freecameraposeaccuracy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

WAVE claims that the view inconsistency of single-image diffusion models has two addressable causes: each novel view is generated independently, and the initial noise injects random variation. It proposes a training-free intervention that warps the input image to target camera poses using depth, then uses the warped images twice: their overlap masks steer a masked batch self-attention between views, and their low-frequency content rewrites the initial noise. The paper reports that adding these steps to existing diffusion models such as MegaScenes and ZeroNVS improves inter-view and input-view consistency metrics and camera-pose accuracy on MegaScenes, DTU, RE10K, and Mip-NeRF 360, at roughly the base model's inference cost. A reader should care because the method points to a consistency fix that transfers across models and strengthens downstream 3D rendering without additional training.

What carries the argument

The load-bearing object is the warped image set $\{W_1,\ldots,W_N\}$ obtained by 3D warping the input view with its depth to the target camera poses. From this set the method derives binary warped region masks $M$, a viewpoint-overlap IoU matrix $U$, and the reference range $[\mu_i-\sigma_i,\mu_i+\sigma_i]$ for each view, where $\mu_i$ and $\sigma_i$ are the mean and standard deviation of the IoU values for view $i$. Warp-guided adaptive attention (WGAA) multiplies the softmax attention map by the mask before aggregating values, so each generated view borrows features only from viewpoints whose warped regions overlap it; pose-aware noise initialization (PANI) fills the warped holes with noise, encodes the warped images, runs a DDPM forward step, extracts normalized low-frequency components, and adds back high-frequency randomness from fresh noise. These two mechanisms inject the same geometric prior once at the attention level and once at the noise level.

What would settle it

Generate an orbit around a known synthetic scene with ground-truth depth and large pose steps such as 60 degrees, run WAVE, and compare LPIPS-next, CLIPSIM-next, and rotation-angle error against the base model and against a variant that masks attention with the true warped regions; if WAVE does not beat the base model, or if visual attention-map peaks do not track the warped region masks at those large baselines, the spatial-correspondence assumption is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is that diffusion models trained for novel view synthesis fail at view consistency because they generate each viewpoint independently and because noise randomness perturbs appearance, and that both failures can be repaired by injecting geometry from a 3D warp of the input view. Warp-guided adaptive attention computes binary masks of the regions visible after warping, measures inter-view overlap with an IoU matrix, picks an adaptive reference range per view, and masks the decoder's batch self-attention so each generated view borrows features only from overlapping views. Pose-aware noise initialization fills warped-image holes with noise, encodes the warped images, runs a DDPM forward step, extracts normalized low-frequency components, and combines them with the high-frequency part of fresh noise. The paper argues that these two mechanisms make generated views more coherent with each other and with the input, and supports the claim with consistency metrics, camera-pose accuracy, and a 3D-rendering downstream evaluation across datasets and diffusion models.

Load-bearing premise

The method depends on the decoder attention of the diffusion U-Net preserving spatial position correspondence, so that the image-space masks from warping line up with the attention maps; if that correspondence breaks, especially under large viewpoint changes, the masks would hide the features that should be shared and the consistency gain would vanish.

Editorial extensions

If this is right

  • Applied to either MegaScenes or ZeroNVS, WAVE lowers LPIPS-next and raises CLIPSIM-next, so neighboring generated views become visually closer while the set stays aligned with the input view.
  • Camera rotation errors (Frobenius norm, rotation angle difference, and angular consistency) decrease, meaning the generated set reads as a coherent camera trajectory rather than unrelated images.
  • Feeding WAVE-generated images to 3D Gaussian splatting improves rendered PSNR and SSIM over the base model's images, so view consistency directly helps 3D reconstruction from a single image.
  • Because the method is training-free and adds no modules, existing single-image diffusion models can adopt it without changing weights and with only the warp operation as extra inference overhead.

Reading between the lines

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

  • An extension the paper leaves implicit is that the same warp-mask recipe would likely transfer to other attention-based generative models whose decoder self-attention preserves spatial correspondence, potentially covering multi-view object generation or camera-conditioned video.
  • Since PANI only injects low-frequency structure, the paper's results imply that a large share of observed view inconsistency is a low-frequency phenomenon in color and layout rather than in high-frequency detail; directly measuring the frequency profile of consistency errors would test this.
  • A testable extension is the paper's suggested autoregressive strategy: generate a nearby view range, warp from those generated views, and repeat; if it removes the large-baseline degradation, it would make the method practical for full 360-degree orbits.
  • The method tolerates degraded depth estimates, since the paper reports that using smaller Depth Anything models changes consistency only slightly, suggesting the warp prior does not need to be pixel-perfect to improve consistency.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 7 minor

Summary. The paper proposes WAVE, a training-free plug-in for single-image novel view synthesis diffusion models. WAVE consists of two components: warp-guided adaptive attention (WGAA), which applies warped-region masks to batch self-attention over adaptively selected reference views, and pose-aware noise initialization (PANI), which re-initializes the diffusion noise with low-frequency content of warped target views. The authors also introduce a view-consistency metric framework combining video-style LPIPS/CLIP similarity between adjacent/input views with COLMAP-based camera pose accuracy. Experiments are conducted on MegaScenes, DTU, RE10K, and Mip-NeRF 360, using MegaScenes and ZeroNVS as base diffusion models, with a 3D Gaussian Splatting downstream task and ablations. The headline claim is that WAVE improves view consistency across different diffusion models at roughly constant inference cost.

Significance. If the results hold, WAVE would offer a low-cost, model-agnostic way to improve consistency in single-image novel view synthesis without retraining or added modules. The paper has clear strengths: the method is training-free, the ablation study isolates the two components, the downstream 3D rendering experiment provides an independent use case, and a project page with code is provided. However, the evaluation is weakened by the use of self-introduced metrics that are not externally validated, an ad hoc COLMAP pose penalty, and several supplementary numbers that contradict the paper's broad claims of improved image quality. The significance is therefore conditional: the core idea is novel and plausible, but the evidence as presented does not yet establish that the gains represent true 3D view consistency rather than shared low-frequency appearance.

major comments (4)
  1. [Section 4.1 and Supp. C] The LPIPS-next/CLIPSIM-next metrics are not established view-consistency metrics, and PANI (Algorithm 1) injects the low-frequency component of each warped image into the initial noise of its corresponding target view. As a result, adjacent generated views are initialized to share low-frequency content by construction, and LPIPS/CLIP are known to be strongly influenced by low-frequency structure. The reported gains may therefore reflect this shared initialization rather than 3D-consistent geometry across viewpoints. No external validation of the metric framework on novel-view synthesis data is provided, and no error bars or significance tests are reported. This directly affects the central claim that WAVE improves view consistency.
  2. [Supp. D.2] The COLMAP pose-accuracy penalty is ad hoc. The text states that when COLMAP reconstructs only 2 of 19 cameras, the available poses are duplicated to 19 before computing Frobenius Norm, Rotation Angle Difference, and Angular Consistency. The frequency of such truncation is not reported per method, so the reader cannot determine how much of the pose-accuracy improvement is driven by this penalty. The authors themselves acknowledge that the duplication approach 'may not be suitable for all scenarios,' yet these metrics are central to Table 1 and to the camera-accuracy claims.
  3. [Supp. Table 5 vs. Section 4.1] The claim that WAVE 'improves performance over the baseline methods across overall metrics' is contradicted by several entries in Supp. Table 5. For example, on RE10K, MegaScenes+WAVE has SSIM 0.261 vs. 0.309 and LPIPS 0.516 vs. 0.494, both worse than MegaScenes; on the MegaScenes dataset, FID worsens from 13.55 to 15.03; and on Mip-NeRF 360, KID worsens from 0.009 to 0.011. Table 2 likewise shows ZeroNVS+WAVE FID worsening from 24.13 to 25.67. These inconsistencies must be resolved, and the image-quality claims need to be qualified accordingly.
  4. [Section 3.3 and Supp. B.1] WGAA relies on the assumption that the decoder self-attention preserves spatial position correspondence, so that image-space warped-region masks can be applied directly to attention maps. The only support is a qualitative attention-map visualization (Fig. 9), and the Limitations section admits sensitivity to large viewpoint changes. A quantitative validation of this assumption is needed, for example by measuring the alignment between attention maps and warped-region masks over varying viewpoint distances, or by ablating WGAA under large viewpoint changes. As written, a reader cannot tell whether WGAA's mechanism is responsible for the reported gains or whether those gains arise primarily from PANI.
minor comments (7)
  1. [Algorithm 1] The note in Algorithm 1 uses G(D0) but D0 is not defined; it should be G(D), the Gaussian filter applied to the input depth map.
  2. [Supp. E.1] The text contains a typo 'key-vale pairs'; it should be 'key-value pairs'.
  3. [Section 4.1] There is a missing space in 'VistaDream[42] and Viewcrafter[47]'; the bracket should follow a space after the model name.
  4. [Figure 4] The labels in Figure 4, especially in part (a), are cluttered and hard to read; please enlarge or reorganize the annotation.
  5. [Supp. D.1] The statement that varying the dropout ratio causes 'no significant performance changes' is not supported by any reported experiment; either add the ablation or remove the claim.
  6. [Supp. C.2] In the Frobenius Norm definition, the matrix A whose entries aij are used is not explicitly defined as the difference between estimated and ground-truth extrinsic matrices; please clarify the notation.
  7. [Supp. D.4] The RE10K sequence evaluation uses only 6 target views per video; the total number of evaluation videos and whether FID/KID are pooled across all frames should be stated.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: WAVE's consistency gains come from externally cited geometric and noise-initialization priors applied to task inputs, not from fitting or self-citation.

full rationale

WAVE's derivation chain is self-contained. WGAA (Section 3.3) and PANI (Section 3.4, Algorithm 1) are defined directly on the task inputs—input view, depth map, and target camera poses—via 3D warping, and the reported metrics are computed on the generated outputs against external baselines (ZeroNVS, MegaScenes, VistaDream), with RE10K additionally evaluated against ground-truth frames. No parameter is fitted to the evaluation metrics, and no load-bearing claim rests on a self-citation: the low-frequency noise-initialization idea is credited to prior external work [24,44], and the attention-masking concept is credited to [38,51]. The WGAA spatial-correspondence assumption is an empirical hypothesis validated qualitatively in Supp. B.1, not an assumption that includes the target result. The main risk is metric validity—PANI injects warped-image low-frequency content into every generated sample, which can inflate LPIPS-next, and the COLMAP duplication penalty in Supp. D.2 is ad hoc—but these are evaluation concerns, not circular derivation. The central claim is therefore not equivalent to its inputs by construction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The method introduces no new physical or mathematical entities; warped images and masks are derived from existing inputs (input image, depth, camera poses). The free parameters are implementation choices rather than fitted constants, and the axioms are domain assumptions about attention geometry, warping reliability, and the evaluation proxy.

free parameters (3)
  • attention dropout ratio = 0.2
    Fixed in experiments (Supp D.1); authors note no significant changes when varied, so not load-bearing.
  • DDPM noise level for noise initialization = 950
    Fixed noise level after VAE encode (Supp D.1); chosen based on prior work, not fitted to target metrics.
  • Gaussian low-pass filter parameters = not specified (borrowed from FreeInit/ConsistI2V)
    Used in Algorithm 1 steps 5-7; exact bandwidth is borrowed from prior work and not reported.
assumptions (5)
  • domain assumption Decoder self-attention preserves spatial position correspondence, so image-space warp masks can be applied directly to attention maps.
    Invoked in Sec 3.3 and validated only qualitatively in Supp B.1; if false, WGAA would corrupt rather than guide generation.
  • domain assumption 3D warped region masks computed from monocular depth and target poses measure true viewpoint overlap.
    Used for adaptive range selection (Sec 3.2) and attention masking (Sec 3.3); degrades for large viewpoint changes, acknowledged in Limitations.
  • domain assumption Low-frequency components of warped latents can be replaced or mixed with high-frequency noise without harming generation fidelity.
    Underlies pose-aware noise initialization (Sec 3.4, Algorithm 1), following FreeInit/ConsistI2V.
  • domain assumption COLMAP rotation estimates on generated images are a valid proxy for view consistency.
    Used for camera pose metrics (Supp D.2); missing correspondences are handled by duplicating poses, an ad hoc assumption the authors acknowledge.
  • standard math Standard FFT/IFFT, Gaussian filtering, and attention softmax behave as expected.
    Background math for Algorithm 1 and Eq. 2; no new mathematical claims.

how reviews work

0 comments
Cite this review

Pith. "Pith review of WAVE: Warp-Based View Guidance for Consistent Novel View Synthesis Using a Single Image." pith.science (2026). https://pith.science/paper/TZDQ3GWU

@misc{pith2026250623518,
  author       = {Pith},
  title        = {Pith review of: WAVE: Warp-Based View Guidance for Consistent Novel View Synthesis Using a Single Image},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TZDQ3GWU}},
  note         = {Machine review of arXiv:2506.23518}
}
read the original abstract

Generating high-quality novel views of a scene from a single image requires maintaining structural coherence across different views, referred to as view consistency. While diffusion models have driven advancements in novel view synthesis, they still struggle to preserve spatial continuity across views. Diffusion models have been combined with 3D models to address the issue, but such approaches lack efficiency due to their complex multi-step pipelines. This paper proposes a novel view-consistent image generation method which utilizes diffusion models without additional modules. Our key idea is to enhance diffusion models with a training-free method that enables adaptive attention manipulation and noise reinitialization by leveraging view-guided warping to ensure view consistency. Through our comprehensive metric framework suitable for novel-view datasets, we show that our method improves view consistency across various diffusion models, demonstrating its broader applicability.

Figures

Figures reproduced from arXiv: 2506.23518 by the authors.

Figure 1
Figure 1. Teaser. We present WAVE, a training-free approach that enhances view consistency both among generated images and between the generated images and the input view. This approach is designed for novel view synthesis diffusion models using a input view. WAVE generates consistent images (lowest LPIPS-next) across views without extra modules, enabling roughly 67s of inference time for 19 images. Abstract Generating high-q… view at source ↗
Figure 2
Figure 2. View inconsistency in diffusion models. Generated novel views using the single image on the left are presented. WAVE maintains view consistency, whereas other diffusion-based models like MegaScenes [39] and ZeroNVS [28] struggle to pre￾serve consistency across viewpoints. As warped images inherently retain partial low-frequency details of the target view, they help alleviate noise random￾ness, ultimately enhancing v… view at source ↗
Figure 3
Figure 3. Effectiveness of adaptive warp-range selection. We present examples of applying our method alongside the conven￾tional batch self-attention method [38, 49] that references all view￾points. The differences in the two camera pose accuracy metrics, Angular Consistency and Frobenius Norm, show that the previous method generates images that are less accurately aligned with the camera poses than ours. 3. Methods We propos… view at source ↗
Figures from the paper (13 more)
Figure 4
Figure 4. Figure 4: Illustration of WAVE. Given an input view, depth map and continuous camera poses, our method generates scene images with smooth view transitions through three distinct processes: (a) adaptive warp-range selection utilizes warped region masks, from which the relevance b…
Figure 5
Figure 5. Figure 5: Analysis of warping methods. We present the genera￾tion results of the methods attempted to mitigate the negative im￾pact of missing regions in warped images : (a) not filling missing regions with noise, (b) only filling missing regions with noise, (c) our method (low-…
Figure 6
Figure 6. Figure 6: Overall comparison of methods. The size of the circles represents inference time. Applying our method improves perfor￾mance with comparable computation costs. stream task. Additional results and experimental settings can be found in the supplementary material. 4.1. Com…
Figure 7
Figure 7. Figure 7: Qualitative results. We provide samples for our method and baseline models [28, 39]. The results are obtained by generating images from an input view and continuous camera poses, selecting representative samples from them. Our method exhibits consistent images among ge…
Figure 8
Figure 8. Figure 8: Visual comparisons in ablation study. We compare of the generated results when PANI and WGAA are removed from our method, respectively. We generate images along an orbital tra￾jectory, ranging from a specific leftward angle to a specific right￾ward angle relative to th…
Figure 9
Figure 9. Figure 9: Visualization of attention maps, warp images and generated views. We present the attention maps of the model alongside the warp images to validate the assumptions of our proposed warp-guided adaptive attention. The results demonstrate that the attention map adapts in a…
Figure 10
Figure 10. Figure 10: Illustration of consistency metrics. A figure that illustrates the primary consistency metrics used in this study: LPIPS-first, CLIPSIM-first, and LPIPS-next, CLIPSIM-next [7, 51]. The first metric measures LPIPS and CLIP similarity by comparing images from different …
Figure 11
Figure 11. Figure 11: Camera parameter visualization. These examples are taken from the camera accuracy evaluation process. The images below represent the results generated by each model: MegaScenes [39] and MegaScenes + WAVE, while the camera graph above visualizes the measured camera pos…
Figure 12
Figure 12. Figure 12: Comparison to diffusion methods on MegaScenes. We compare our framework with existing diffusion-based models, Ze￾roNVS [28] and MegaScenes [39]. Additional generation results are provided on the MegaScenes dataset. The images in the middle column represent the input i…
Figure 13
Figure 13. Figure 13: Comparison to diffusion methods on RE10K. We provide additional generation results on the RE10K [50] dataset to compare our method with baselines. The images in the middle column represent the input images [PITH_FULL_IMAGE:figures/full_fig_p020_13.png]
Figure 14
Figure 14. Figure 14: Qualitative results. Additionally, using samples from the RE10K dataset [50], we generate images with our method by taking the input view image in the middle column and continuous camera poses as inputs [PITH_FULL_IMAGE:figures/full_fig_p021_14.png]
Figure 15
Figure 15. Figure 15: Qualitative results. Additionally, we generate images using our method with samples from the MegaScenes [39] dataset, where the input view image in the middle column and continuous camera poses serve as inputs [PITH_FULL_IMAGE:figures/full_fig_p022_15.png]
Figure 16
Figure 16. Figure 16: Failure cases. Top: an example that shows that the model struggles to accurately generate objects with thin structures, such as lines. Bottom: an example that shows the model’s tendency to repeatedly generate identical structural patterns as the viewpoint difference i…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

51 extracted references · 34 canonical work pages

  1. [1]

    Large-scale data for multiple-view stereopsis

    Henrik Aanæs, Rasmus Ramsbøl Jensen, George V ogiatzis, Engin Tola, and Anders Bjorholm Dahl. Large-scale data for multiple-view stereopsis. International Journal of Computer Vision, pages 1–16, 2016. 5, 8, 15

    work page 2016

  2. [2]

    Mip-nerf 360: Unbounded anti-aliased neural radiance fields

    Jonathan T Barron, Ben Mildenhall, Dor Verbin, Pratul P Srinivasan, and Peter Hedman. Mip-nerf 360: Unbounded anti-aliased neural radiance fields. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5470–5479, 2022. 5, 8, 15

    work page 2022

  3. [3]

    Consistent diffusion models: Mitigating sampling drift by learning to be consistent.Advances in Neu- ral Information Processing Systems, 36, 2024

    Giannis Daras, Yuval Dagan, Alex Dimakis, and Constanti- nos Daskalakis. Consistent diffusion models: Mitigating sampling drift by learning to be consistent.Advances in Neu- ral Information Processing Systems, 36, 2024. 3

    2024

  4. [4]

    Objaverse-xl: A universe of 10m+ 3d objects

    Matt Deitke, Ruoshi Liu, Matthew Wallingford, Huong Ngo, Oscar Michel, Aditya Kusupati, Alan Fan, Christian Laforte, Vikram V oleti, Samir Yitzhak Gadre, et al. Objaverse-xl: A universe of 10m+ 3d objects. Advances in Neural Informa- tion Processing Systems, 36, 2024. 2

    work page 2024

  5. [5]

    Cat3d: Create anything in 3d with multi-view diffusion models

    Ruiqi Gao, Aleksander Holynski, Philipp Henzler, Arthur Brussee, Ricardo Martin-Brualla, Pratul Srinivasan, Jonathan T Barron, and Ben Poole. Cat3d: Create anything in 3d with multi-view diffusion models. arXiv preprint arXiv:2405.10314, 2024. 1

    arXiv 2024

  6. [6]

    Nerfdiff: Single-image view synthesis with nerf-guided dis- tillation from 3d-aware diffusion

    Jiatao Gu, Alex Trevithick, Kai-En Lin, Joshua M Susskind, Christian Theobalt, Lingjie Liu, and Ravi Ramamoorthi. Nerfdiff: Single-image view synthesis with nerf-guided dis- tillation from 3d-aware diffusion. In International Con- ference on Machine Learning, pages 11808–11826. PMLR,

    work page

  7. [7]

    Sparsectrl: Adding sparse controls to text-to-video diffusion models

    Yuwei Guo, Ceyuan Yang, Anyi Rao, Maneesh Agrawala, Dahua Lin, and Bo Dai. Sparsectrl: Adding sparse controls to text-to-video diffusion models. In European Conference on Computer Vision, pages 330–348. Springer, 2024. 2, 12, 13

    work page 2024

  8. [8]

    Prompt-to-prompt im- age editing with cross attention control

    Amir Hertz, Ron Mokady, Jay Tenenbaum, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Prompt-to-prompt im- age editing with cross attention control. arXiv preprint arXiv:2208.01626, 2022. 2, 3, 5

    arXiv 2022

Show all 51 references
  1. [9]

    Clipscore: A reference-free evaluation met- ric for image captioning

    Jack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras, and Yejin Choi. Clipscore: A reference-free evaluation met- ric for image captioning. arXiv preprint arXiv:2104.08718,

    arXiv

  2. [10]

    Videobooth: Diffusion-based video generation with image prompts

    Yuming Jiang, Tianxing Wu, Shuai Yang, Chenyang Si, Dahua Lin, Yu Qiao, Chen Change Loy, and Ziwei Liu. Videobooth: Diffusion-based video generation with image prompts. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 6689– 6700, 2024. 3

    2024

  3. [11]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Transactions on Graphics, 42 (4), 2023. 1, 2, 8, 16

    2023

  4. [12]

    Genwarp: Single image to novel views with semantic-preserving generative warping

    Seungryong Kim. Genwarp: Single image to novel views with semantic-preserving generative warping. In 38th Con- ference on Neural Information Processing Systems Work- shop, NeurIPS 2024. Neural information processing systems foundation, 2024. 2, 3, 6

    2024

  5. [13]

    Vivid-1-to-3: Novel view synthesis with video diffusion models

    Jeong-gi Kwak, Erqun Dong, Yuhe Jin, Hanseok Ko, Shweta Mahajan, and Kwang Moo Yi. Vivid-1-to-3: Novel view synthesis with video diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6775–6785, 2024. 2, 5

    2024

  6. [14]

    Infinitenature-zero: Learning perpetual view gen- eration of natural scenes from single images

    Zhengqi Li, Qianqian Wang, Noah Snavely, and Angjoo Kanazawa. Infinitenature-zero: Learning perpetual view gen- eration of natural scenes from single images. In European Conference on Computer Vision , pages 515–534. Springer,

  7. [15]

    Infinite na- ture: Perpetual view generation of natural scenes from a sin- gle image

    Andrew Liu, Richard Tucker, Varun Jampani, Ameesh Makadia, Noah Snavely, and Angjoo Kanazawa. Infinite na- ture: Perpetual view generation of natural scenes from a sin- gle image. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 14458–14467, 2021. 2, 3

    2021

  8. [16]

    Zero-1-to- 3: Zero-shot one image to 3d object

    Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tok- makov, Sergey Zakharov, and Carl V ondrick. Zero-1-to- 3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF international conference on computer vision, pages 9298–9309, 2023. 2

    2023

  9. [17]

    Wonder3d: Sin- gle image to 3d using cross-domain diffusion

    Xiaoxiao Long, Yuan-Chen Guo, Cheng Lin, Yuan Liu, Zhiyang Dou, Lingjie Liu, Yuexin Ma, Song-Hai Zhang, Marc Habermann, Christian Theobalt, et al. Wonder3d: Sin- gle image to 3d using cross-domain diffusion. In Proceed- ings of the IEEE/CVF conference on computer vision and pa...

    2024

  10. [18]

    Improving text- to-image consistency via automatic prompt optimization

    Oscar Ma ˜nas, Pietro Astolfi, Melissa Hall, Candace Ross, Jack Urbanek, Adina Williams, Aishwarya Agrawal, Adri- ana Romero-Soriano, and Michal Drozdzal. Improving text- to-image consistency via automatic prompt optimization. arXiv preprint arXiv:2403.17804, 2024. 3

    2024 arXiv

  11. [19]

    Post- rendering 3d warping

    William R Mark, Leonard McMillan, and Gary Bishop. Post- rendering 3d warping. In Proceedings of the 1997 sympo- sium on Interactive 3D graphics, pages 7–ff, 1997. 2, 3

    1997

  12. [20]

    Srinivasan, Matthew Tancik, Jonathan T

    Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view syn- thesis. In ECCV, 2020. 1

    2020

  13. [21]

    Multidiff: Consistent novel view synthesis from a single image

    Norman M ¨uller, Katja Schwarz, Barbara R ¨ossle, Lorenzo Porzi, Samuel Rota Bul `o, Matthias Nießner, and Peter Kontschieder. Multidiff: Consistent novel view synthesis from a single image. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition...

    2024

  14. [22]

    Instant neural graphics primitives with a mul- tiresolution hash encoding

    Thomas M ¨uller, Alex Evans, Christoph Schied, and Alexan- der Keller. Instant neural graphics primitives with a mul- tiresolution hash encoding. ACM transactions on graphics (TOG), 41(4):1–15, 2022. 15

    2022

  15. [23]

    Com- mon objects in 3d: Large-scale learning and evaluation of real-life 3d category reconstruction

    Jeremy Reizenstein, Roman Shapovalov, Philipp Henzler, Luca Sbordone, Patrick Labatut, and David Novotny. Com- mon objects in 3d: Large-scale learning and evaluation of real-life 3d category reconstruction. In Proceedings of the IEEE/CVF international conference on computer vi...

    2021

  16. [24]

    Consisti2v: Enhancing visual consistency for image-to-video generation

    Weiming Ren, Huan Yang, Ge Zhang, Cong Wei, Xinrun Du, Wenhao Huang, and Wenhu Chen. Consisti2v: Enhancing visual consistency for image-to-video generation. Transac- tions on Machine Learning Research. 2, 3, 5, 14, 17

  17. [25]

    Generalized in- tersection over union: A metric and a loss for bounding box regression

    Hamid Rezatofighi, Nathan Tsoi, JunYoung Gwak, Amir Sadeghian, Ian Reid, and Silvio Savarese. Generalized in- tersection over union: A metric and a loss for bounding box regression. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 658–666,

  18. [26]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 3

    2022

  19. [27]

    Diffusion models with learned adap- tive noise

    Subham Sekhar Sahoo, Aaron Gokaslan, Chris De Sa, and V olodymyr Kuleshov. Diffusion models with learned adap- tive noise. arXiv preprint arXiv:2312.13236, 2023. 2, 5

    2023 arXiv

  20. [28]

    Zeronvs: Zero-shot 360- degree view synthesis from a single image

    Kyle Sargent, Zizhang Li, Tanmay Shah, Charles Herrmann, Hong-Xing Yu, Yunzhi Zhang, Eric Ryan Chan, Dmitry La- gun, Li Fei-Fei, Deqing Sun, et al. Zeronvs: Zero-shot 360- degree view synthesis from a single image. In Proceedings of the IEEE/CVF Conference on Computer Vision a...

    2024

  21. [29]

    Structure-from-motion revisited

    Johannes Lutz Sch ¨onberger and Jan-Michael Frahm. Structure-from-motion revisited. In Conference on Com- puter Vision and Pattern Recognition (CVPR), 2016. 15, 16

    2016

  22. [30]

    Pixelwise view selection for un- structured multi-view stereo

    Johannes Lutz Sch ¨onberger, Enliang Zheng, Marc Pollefeys, and Jan-Michael Frahm. Pixelwise view selection for un- structured multi-view stereo. In European Conference on Computer Vision (ECCV), 2016. 15, 16

    2016

  23. [31]

    Let 2d diffusion model know 3d- consistency for robust text-to-3d generation

    Junyoung Seo, Wooseok Jang, Min-Seop Kwak, Hyeonsu Kim, Jaehoon Ko, Junho Kim, Jin-Hwa Kim, Jiyoung Lee, and Seungryong Kim. Let 2d diffusion model know 3d- consistency for robust text-to-3d generation. arXiv preprint arXiv:2303.07937, 2023. 8

    2023 arXiv

  24. [32]

    Zero123++: a single image to consistent multi-view dif- fusion base model, 2023

    Ruoxi Shi, Hansheng Chen, Zhuoyang Zhang, Minghua Liu, Chao Xu, Xinyue Wei, Linghao Chen, Chong Zeng, and Hao Su. Zero123++: a single image to consistent multi-view dif- fusion base model, 2023. 2

    2023

  25. [33]

    Mvdream: Multi-view diffusion for 3d gen- eration, 2024

    Yichun Shi, Peng Wang, Jianglong Ye, Mai Long, Kejie Li, and Xiao Yang. Mvdream: Multi-view diffusion for 3d gen- eration, 2024. 8

    2024

  26. [34]

    Realmdreamer: Text-driven 3d scene generation with inpainting and depth diffusion

    Jaidev Shriram, Alex Trevithick, Lingjie Liu, and Ravi Ra- mamoorthi. Realmdreamer: Text-driven 3d scene generation with inpainting and depth diffusion. 2025. 2, 3

    2025

  27. [35]

    Very deep convolutional networks for large-scale image recognition

    Karen Simonyan. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014. 12

    2014 arXiv

  28. [36]

    Denois- ing diffusion implicit models, 2022

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denois- ing diffusion implicit models, 2022. 12

    2022

  29. [37]

    Dreamgaussian: Generative gaussian splatting for effi- cient 3d content creation

    Jiaxiang Tang, Jiawei Ren, Hang Zhou, Ziwei Liu, and Gang Zeng. Dreamgaussian: Generative gaussian splatting for effi- cient 3d content creation. arXiv preprint arXiv:2309.16653,

    arXiv

  30. [38]

    Training-free consis- tent text-to-image generation

    Yoad Tewel, Omri Kaduri, Rinon Gal, Yoni Kasten, Lior Wolf, Gal Chechik, and Yuval Atzmon. Training-free consis- tent text-to-image generation. ACM Transactions on Graph- ics (TOG), 43(4):1–18, 2024. 2, 3, 4, 14, 16

    2024

  31. [39]

    Megascenes: Scene-level view synthesis at scale

    Joseph Tung, Gene Chou, Ruojin Cai, Guandao Yang, Kai Zhang, Gordon Wetzstein, Bharath Hariharan, and Noah Snavely. Megascenes: Scene-level view synthesis at scale. In European Conference on Computer Vision, pages 197–214. Springer, 2025. 1, 2, 3, 5, 7, 8, 14, 15, 16, 18, 19, 22

    2025

  32. [40]

    Attention is all you need

    A Vaswani. Attention is all you need. Advances in Neural Information Processing Systems, 2017. 3

    2017

  33. [41]

    Sv3d: Novel multi-view syn- thesis and 3d generation from a single image using latent video diffusion

    Vikram V oleti, Chun-Han Yao, Mark Boss, Adam Letts, David Pankratz, Dmitry Tochilkin, Christian Laforte, Robin Rombach, and Varun Jampani. Sv3d: Novel multi-view syn- thesis and 3d generation from a single image using latent video diffusion. In European Conference on Computer...

    2024

  34. [42]

    Vistadream: Sampling multi- view consistent images for single-view scene reconstruction

    Haiping Wang, Yuan Liu, Ziwei Liu, Wenping Wang, Zhen Dong, and Bisheng Yang. Vistadream: Sampling multi- view consistent images for single-view scene reconstruction. arXiv preprint arXiv:2410.16892, 2024. 1, 2, 3, 5, 6, 16

    2024 arXiv

  35. [43]

    A latent space of stochastic diffusion models for zero-shot image editing and guidance

    Chen Henry Wu and Fernando De la Torre. A latent space of stochastic diffusion models for zero-shot image editing and guidance. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 7378–7387, 2023. 3

    2023

  36. [44]

    Freeinit: Bridging initialization gap in video dif- fusion models

    Tianxing Wu, Chenyang Si, Yuming Jiang, Ziqi Huang, and Ziwei Liu. Freeinit: Bridging initialization gap in video dif- fusion models. In European Conference on Computer Vision, pages 378–394. Springer, 2025. 2, 3, 5, 14

    2025

  37. [45]

    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. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10371–10381, 2024. 14, 17

    2024

  38. [46]

    Wonderjourney: Going from anywhere to everywhere

    Hong-Xing Yu, Haoyi Duan, Junhwa Hur, Kyle Sargent, Michael Rubinstein, William T Freeman, Forrester Cole, De- qing Sun, Noah Snavely, Jiajun Wu, et al. Wonderjourney: Going from anywhere to everywhere. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern R...

    2024

  39. [47]

    Viewcrafter: Taming video diffusion models for high-fidelity novel view synthesis.arXiv preprint arXiv:2409.02048, 2024

    Wangbo Yu, Jinbo Xing, Li Yuan, Wenbo Hu, Xiaoyu Li, Zhipeng Huang, Xiangjun Gao, Tien-Tsin Wong, Ying Shan, and Yonghong Tian. Viewcrafter: Taming video diffusion models for high-fidelity novel view synthesis.arXiv preprint arXiv:2409.02048, 2024. 6

    2024 arXiv

  40. [48]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In CVPR, 2018. 12

    2018

  41. [49]

    Free3d: Consistent novel view synthesis without 3d representation

    Chuanxia Zheng and Andrea Vedaldi. Free3d: Consistent novel view synthesis without 3d representation. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9720–9731, 2024. 3, 4, 16

    2024

  42. [50]

    Stereo magnification: learning view synthesis using multiplane images

    Tinghui Zhou, Richard Tucker, John Flynn, Graham Fyffe, and Noah Snavely. Stereo magnification: learning view synthesis using multiplane images. ACM Transactions on Graphics (TOG), 37(4):1–12, 2018. 5, 8, 15, 16, 18, 20, 21

    2018

  43. [51]

    Storydiffusion: Consistent self- attention for long-range image and video generation

    Yupeng Zhou, Daquan Zhou, Ming-Ming Cheng, Jiashi Feng, and Qibin Hou. Storydiffusion: Consistent self- attention for long-range image and video generation. arXiv preprint arXiv:2405.01434, 2024. 2, 3, 12, 13, 16 A. Code and Website A project website has been created to introd...

    2024 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.