Skel3D: Skeleton Guided Novel View Synthesis

Aron F\'othi; Bence Fazekas; Kristian Fenech; Natabara M\'at\'e Gy\"ongy\"ossy

REVIEW 4 major objections 6 minor 25 references

Skel3D: Skeleton Guided Novel View Synthesis

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Skel3D claims that conditioning a diffusion-based novel view synthesizer on the target object's 2D skeleton improves pose accuracy and multi-view consistency, outperforming the Free3D baseline on Objaverse objects.

desk verdict Reasonable skeleton-conditioning idea, but the experiments appear to test pose-conditioned frame synthesis with an oracle target skeleton, not monocular novel view synthesis. read the letter →

arxiv 2412.03407 v1 pith:WOIQYDIX submitted 2024-12-04 cs.CV

Aron Fóthi , Bence Fazekas , Natabara Máté Gyöngyössy , Kristian Fenech This is my paper · ORCID

classification cs.CV

keywords novelviewsynthesisdiffusionmodelskeletonconditioningObjaverseposeaccuracymulti-viewconsistencyadaptivenormalizationsingle-view3D

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

This paper claims that feeding a diffusion-based novel view synthesizer an explicit 2D skeleton of the target object, rather than relying only on camera pose, makes generated views more pose-accurate and consistent across viewpoints. It builds on Free3D and replaces its ray-conditioning normalization with a skeleton-conditioning normalization, training on a curated subset of Objaverse objects that have bone animations. On a 260-object held-out test set, the skeleton-guided model improves L1, SSIM, PSNR, LPIPS, and FID relative to the Free3D baseline, and the gains are statistically significant. The caveat is that the evaluation supplies the skeleton of the target view directly from the ground-truth 3D model, so the paper does not test the monocular setting in which the skeleton must first be estimated.

What carries the argument

The central mechanism is the Skeleton Conditioning Normalization (SCN) layer, an adaptive normalization inserted after every sub-module of the diffusion UNet. It applies group normalization to each activation, then modulates it with scale and shift parameters produced by an MLP from the encoded skeleton-image embedding, replacing Free3D's ray-conditioning normalization. This lets pre-trained image-generation features be steered by explicit structural cues without retraining the whole network.

What would settle it

Run the same 260-object evaluation with target-view skeletons produced by an automatic 2D-to-3D skeleton estimator instead of rendered ground-truth bones; if the metric improvements over Free3D shrink below statistical significance or reverse, the method's advantage depends on oracle skeleton access rather than on the conditioning mechanism itself.

Watch

Extended reading notes

Core claim

Skel3D establishes that explicit skeletal structure can serve as an effective conditioning signal for single-image novel view synthesis. The method intakes a source image and a target-view skeleton image, encodes both with the Stable Diffusion image encoder, and modulates every sub-module of the UNet through a Skeleton Conditioning Normalization (SCN) layer, replacing Free3D's Ray Conditioning Normalization. Training on roughly 12,000 animated Objaverse objects with rendered bone overlays, the model outperforms the Free3D baseline across all five evaluated metrics on a 260-object test set, with Mann-Whitney U tests showing p<.01 for the main comparisons. The paper further shows a monotone relationship between skeleton-to-object bounding-box IoU and the size of the improvement, indicating that the benefit is driven by how well the skeleton fits the object.

Load-bearing premise

The evaluation assumes the ground-truth skeleton of the target view is available at test time; in the claimed monocular setting that skeleton must be estimated from the source image, a step the paper does not test.

Editorial extensions

If this is right

Replacing ray conditioning with skeleton conditioning improves four pixel- and perception-level metrics on the Objaverse test set, with statistically significant differences at p<.01.
Keeping both skeleton and ray conditioning (Skel3D+RCN) gives the best average metrics, though the only significant gain over skeleton-only is in LPIPS at p≈0.044.
The method generalizes across diverse non-anthropomorphic object categories, not just human-like shapes.
Because skeletons are represented as images, the approach can accept skeletons from any current or future pose estimator, avoiding a fixed keypoint format.
The dependency of gains on skeleton quality shows that the method's advantage is bounded by how well the provided skeleton matches the object's true structure.

Reading between the lines

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

In a genuine single-image deployment, the target-view skeleton would come from an estimator rather than the ground-truth model; the paper's Figure 4 suggests that imperfect skeletons, measured by low IoU, erase or reverse the gains, so the practical value hinges on unseen estimation error.
The skeleton-image representation is a natural interface for category-agnostic 3D skeleton lifters such as 3D-LFM, which the paper points to; a testable next step is pairing that estimator with Skel3D and measuring end-to-end performance versus Free3D.
The same conditioning could be extended to video: Objaverse animations provide temporal skeleton sequences, so the method may serve as a base for temporally consistent synthesis from a single frame.
Since skeleton quality directly modulates improvement, a promising research direction is learning a skeleton-refinement module inside the diffusion loop, which the paper does not explore.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The skeleton-conditioning idea is simple and worth a few minutes: replace the ray conditioning in a Free3D-style diffusion model with a skeleton conditioning layer, feeding in a rendered skeleton image of the target view. The experiments, however, do not support the paper's central claim about monocular novel view synthesis. The strongest reading of the text is that the 'views' are actually animation frames rendered with a fixed camera — no camera trajectory is ever described — and the target skeleton is ground truth obtained from the 3D model. Under that reading, the task is pose-conditioned image synthesis, not NVS, and the improvements in Table 1 measure how well the model copies a pose from an oracle rather than any real viewpoint generalization.

Give credit where it's due. The SCN layer is cleanly specified (modulated group norm with MLP from skeleton embeddings), and the choice to represent skeletons as images is a smart way around fixed keypoint counts (it's why ControlNet-style guides can work for arbitrary objects). The curated 260-object test set is deliberately disjoint from Free3D's training data, and they run Mann-Whitney U tests instead of just reporting means. Figure 4's skeleton-quality correlation and Figure 5's failure cases are genuinely informative. These are careful experimentalists.

The soft spots, in proportion. (1) The NVS framing is load-bearing and not demonstrated. Section 3.2 says they render every fourth frame of the first 24 animation frames, with no camera motion described. Worse, the model replaces all RCN layers with SCN layers, so the network has no ray or camera input at all. If the camera did not move, the 'learned views' are just different animation frames at the same camera pose. (2) The target skeleton at test time is ground truth rendered from the 3D model, exactly the explicit 3D structure the abstract claims not to use. The limitations paragraph in Section 6 does flag this, but there is no estimation experiment, despite citing 3D-LFM as a path. (3) Minor: the only baseline is original Free3D; the fine-tuned Free3D is mentioned as worse but not shown; Zero-1-to-3 is discussed but not evaluated. Effect sizes are small (about +1.1 dB PSNR), although the nonparametric tests argue they are real.

The stress-test note holds up on reading. The data-pipeline language is consistent with animation-frame rendering, and the RCN-to-SCN replacement is consistent with the model having no camera information.

Who this is for: someone working on skeleton-guided pose-conditioned generation might borrow the SCN design and the skeleton-as-image principle. As an NVS contribution, it needs reframing and re-evaluation with estimated skeletons plus a camera-motion baseline before it can claim any SOTA status. I would not cite it in the next 12 months.

Recommendation for peer review: send it to reviewers, but with an explicit request that they pin down the data pipeline. If the camera did not move, the paper needs a new title and abstract, and a practical-skeleton experiment must be added. The idea is real; the framing is not.

Referee Report

4 major / 6 minor

Summary. The paper proposes Skel3D, a diffusion-based approach to single-view novel view synthesis conditioned on object skeletons. Building on the Free3D architecture, it replaces ray conditioning normalization (RCN) with a skeleton conditioning normalization (SCN) layer and trains on a curated subset of Objaverse animated objects with bone structures. The authors report quantitative improvements over Free3D on 260 held-out objects across L1, SSIM, PSNR, LPIPS, and FID, and claim that skeleton guidance improves pose accuracy and multi-view consistency for monocular open-set NVS without explicit 3D representations.

Significance. If the claimed results held in a true monocular setting, injecting a category-agnostic 2D skeleton image as conditioning into a diffusion NVS model would be a practically useful idea, particularly because it avoids fixed keypoint definitions and could in principle work with any 2D pose estimator. The architectural modification is clearly described, and the use of a non-parametric significance test is a positive step. However, the significance is conditional on an evaluation protocol that the paper does not provide: the target-view skeletons are oracle renderings from the 3D model, and the test frames appear to be animation frames rather than camera viewpoints. As it stands, the contribution is closer to oracle-skeleton-conditioned frame synthesis than to monocular novel view synthesis, so the main claimed advance is not demonstrated.

major comments (4)

[Section 3.2, Eq. (4), Section 6] The test-time conditioning signal S in Eq. (4) is the ground-truth skeleton of the target view, rendered directly from the Objaverse 3D model as described in Section 3.2. In the claimed monocular setting this skeleton is not available; it would have to be estimated from the source image. The paper does not test this, and Section 6 explicitly defers "integrating 2D-3D skeleton lifting models" to future work. Consequently, Table 1 compares Skel3D with oracle pose information against Free3D without it, so the reported gains are an upper bound and do not support the abstract's monocular NVS claim.
[Section 3.2, Section 3.3, Table 1] The data pipeline renders "every fourth frame of the first 24 animation frames" and no camera motion or viewpoint sampling is described anywhere. Moreover, Skel3D replaces all Ray Conditioning Normalization layers with Skeleton Conditioning Normalization (Section 3.3), so the model no longer receives camera rays and is conditioned only on the source image and the target skeleton (Eq. 4). Under the described pipeline, the target skeleton encodes an animation frame, not a camera viewpoint. The paper never operationalizes what is meant by a "view" in terms of camera pose, so Table 1 appears to measure pose-conditioned frame synthesis rather than novel view synthesis.
[Section 5.1, Table 1] The comparison set consists of a single baseline, Free3D, while the abstract claims to outperform "existing state-of-the-art NVS techniques"; no results for Zero-1-to-3 or other recent NVS methods are reported. In addition, the paper states that fine-tuning the original Free3D architecture on the curated training set decreased performance, but the numbers for that fine-tuned baseline are not reported. This missing baseline matters because Skel3D is fine-tuned on the curated Objaverse subset while the reported Free3D result is the original pretrained model, so the reader cannot separate the effect of skeleton conditioning from the effect of additional fine-tuning on the training distribution.
[Section 5, Section 6, Figures 4 and 5] The paper repeatedly claims improvements in "pose accuracy" and "multi-view consistency," but Section 5 lists only image-similarity metrics (L1, SSIM, PSNR, LPIPS, FID); no pose error metric and no consistency metric between generated views is reported. Furthermore, the authors' own analysis shows that when the skeleton fits the object poorly, Skel3D performs worse than Free3D (Figure 5) and that average improvement scales with skeleton-object IoU (Figure 4). Since all test skeletons are ground-truth renderings, the reported average improvements are conditional on perfect skeleton quality, and no experiment with an estimated skeleton is presented.

minor comments (6)

[Abstract, Section 3.2] The abstract says the method works "without relying on explicit 3D representations," but Section 3.2 uses the 3D bone structures from Objaverse, projects them into 2D skeleton images, and uses those as conditioning; this wording should be clarified.
[Figure 1] Figure 1's caption says "predicted skeleton," but no skeleton prediction module is described in the paper; the skeleton is obtained from the 3D object or would need an external estimator. The caption should say "provided skeleton" or an estimation pipeline should be described.
[Section 5.1, Table 1] The FID-Score is reported without a standard deviation and without details on how the FID is computed over the test set (number of generated samples, reference distribution, and whether the same random seeds are used); this should be specified.
[Section 5.1, Table 2] The Mann-Whitney U test appears to treat each image as an independent sample, but images from the same object are likely correlated; a paired or per-object analysis would be more appropriate.
[Section 6] The paper mentions alternative architectures and their results in the supplementary material, but the arXiv version does not include a supplementary document; this material should be provided or the reference removed.
[Section 3.5 and Section 4] Training details such as learning rate, optimizer, noise schedule, and exact number of optimization steps are not reported; these details are needed for reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the skeleton is an input to the model, not an output of the derivation; the oracle-skeleton evaluation is a validity concern, not a logical circle.

full rationale

The paper's derivation chain is self-contained and non-circular. The method conditions a diffusion model on source images and target-view skeleton images, with the training loss in Eq. (4) being a standard denoising objective over provided target views and skeleton embeddings. Nothing is fitted to the evaluation metric, and no predicted quantity is defined in terms of its own inputs. The skeleton is an independent conditioning signal, not a learned output or a fitted proxy for the target view. The use of ground-truth target skeletons rendered from the Objaverse 3D model (Section 3.2) certainly makes the comparison to Free3D favorable and weakens the claim of a true monocular setting, but this is an evaluation-design issue, not circular reasoning: Free3D receives no such oracle conditioning, and the skeleton remains an external input rather than a quantity derived from the model's own predictions. There are no load-bearing self-citations: the paper builds on prior independent work (Free3D, Zero-1-to-3, Stable Diffusion, ControlNet), and none of the authors are cited as the sole justification for a premise. The paper also explicitly acknowledges limitations in Section 6, including the small scale of public skeleton datasets and the need for in-the-wild validation, as well as the dependency on skeleton quality shown in Figures 4 and 5. These limitations are correctly framed as risks to external validity rather than as steps that reduce the method to its inputs. Overall, no circular step can be exhibited, so the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The method introduces no new free parameters with reported values. The model's trainable weights are learned, not hand-set. The key assumptions are that skeleton images are an effective conditioning signal and that target-view skeletons are available at inference. The latter is the main ad hoc assumption.

assumptions (4)

domain assumption The pre-trained Stable Diffusion image variations model provides a sufficiently strong image prior for the NVS fine-tune.
Section 3.3 states the model uses the same image-to-image checkpoint; if this prior is weak for the object categories, the results would differ.
domain assumption Objaverse skeleton animations are an appropriate and sufficient supervision source for learning skeleton-guided NVS.
Section 3.2 describes using the Diffusion4D-curated 12K subset; the representativeness of this subset is not analyzed.
ad hoc to paper A rendered 2D skeleton image, encoded with the same image encoder, carries the pose and structure information needed to guide view synthesis.
Section 3.4 uses E(s_i) as conditioning; this choice is motivated only by empirical results, not by analysis.
ad hoc to paper The ground-truth skeleton of the target view is available at test time.
Section 5.1 evaluation feeds encoded target skeletons to the model; this is the oracle assumption that makes the comparison favorable.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Skel3D: Skeleton Guided Novel View Synthesis." pith.science (2026). https://pith.science/paper/WOIQYDIX

@misc{pith2026241203407,
  author       = {Pith},
  title        = {Pith review of: Skel3D: Skeleton Guided Novel View Synthesis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WOIQYDIX}},
  note         = {Machine review of arXiv:2412.03407}
}

read the original abstract

In this paper, we present an approach for monocular open-set novel view synthesis (NVS) that leverages object skeletons to guide the underlying diffusion model. Building upon a baseline that utilizes a pre-trained 2D image generator, our method takes advantage of the Objaverse dataset, which includes animated objects with bone structures. By introducing a skeleton guide layer following the existing ray conditioning normalization (RCN) layer, our approach enhances pose accuracy and multi-view consistency. The skeleton guide layer provides detailed structural information for the generative model, improving the quality of synthesized views. Experimental results demonstrate that our skeleton-guided method significantly enhances consistency and accuracy across diverse object categories within the Objaverse dataset. Our method outperforms existing state-of-the-art NVS techniques both quantitatively and qualitatively, without relying on explicit 3D representations.

Figures

Figures reproduced from arXiv: 2412.03407 by the authors.

**Figure 2.** Architecture of our skeleton-guided model for NVS. Given a single input image, we introduce [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** The first column shows the source image for NVS, followed by the target view in the second column. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: illustrates the correlation between skeleton quality and model improvement. The x-axis represents the Intersection over Union (IoU) of the bounding boxes of the object and its skeleton, serving as a measure of how well the skeleton fits the object. The y-axis shows …

**Figure 5.** Figure 5: When the guidance skeleton is insufficient, our model’s performance drops compared to the baseline [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 17 canonical work pages

[1]

High- resolution image synthesis with latent diffusion mod- els

Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨ orn Ommer. High- resolution image synthesis with latent diffusion mod- els. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 2, 3, 5

work page 2022
[2]

Nerf: Representing scenes as neural ra- diance fields for view synthesis

Ben Mildenhall, Pratul P Srinivasan, Matthew Tan- cik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural ra- diance fields for view synthesis. Communications of the ACM, 65(1):99–106, 2021. 1, 3

work page 2021
[3]

Deep unsuper- vised learning using nonequilibrium thermodynam- ics

Jascha Sohl-Dickstein, Eric Weiss, Niru Mah- eswaranathan, and Surya Ganguli. Deep unsuper- vised learning using nonequilibrium thermodynam- ics. In International conference on machine learning, pages 2256–2265. PMLR, 2015. 1

work page 2015
[4]

Denois- ing diffusion probabilistic models

Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denois- ing diffusion probabilistic models. Advances in neu- ral information processing systems, 33:6840–6851,

work page
[5]

Free3d: Con- sistent novel view synthesis without 3d representa- tion

Chuanxia Zheng and Andrea Vedaldi. Free3d: Con- sistent novel view synthesis without 3d representa- tion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9720–9731, 2024. 1, 3, 4, 5, 7, 9

work page 2024
[6]

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

Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tokmakov, Sergey Zakharov, and Carl Vondrick. Zero-1-to-3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF international confer- ence on computer vision, pages 9298–9309, 2023. 1, 3, 5, 7

work page 2023
[7]

Hugs: Human gaussian splats

Muhammed Kocabas, Jen-Hao Rick Chang, James Gabriel, Oncel Tuzel, and Anurag Ranjan. Hugs: Human gaussian splats. In Proceedings of the IEEE/CVF conference on computer vision and pat- tern recognition, pages 505–515, 2024. 2

work page 2024
[8]

Stable diffusion image variations model

Lambdalabs. Stable diffusion image variations model. https://huggingface.co/lambdalabs/ sd-image-variations-diffusers , 2022. Accessed: 2024-08-15. 3, 5

work page 2022

Show all 25 references

[9]

Consis- tent view synthesis with pose-guided diffusion mod- els

Hung-Yu Tseng, Qinbo Li, Changil Kim, Suhib Al- sisan, Jia-Bin Huang, and Johannes Kopf. Consis- tent view synthesis with pose-guided diffusion mod- els. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 16773–16783, 2023. 3

2023
[10]

Diffusion4d: Fast spatial-temporal consistent 4d generation via video diffusion models

Hanwen Liang, Yuyang Yin, Dejia Xu, Hanxue Liang, Zhangyang Wang, Konstantinos N Platani- otis, Yao Zhao, and Yunchao Wei. Diffusion4d: Fast spatial-temporal consistent 4d generation via video diffusion models. arXiv preprint arXiv:2405.16645,

arXiv
[11]

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

Vikram Voleti, Chun-Han Yao, Mark Boss, Adam Letts, David Pankratz, Dmitry Tochilkin, Christian Laforte, Robin Rombach, and Varun Jampani. Sv3d: Novel multi-view synthesis and 3d generation from a single image using latent video diffusion. arXiv preprint arXiv:2403.12008, 2024. 3

2024 arXiv
[12]

Nonrigid structure from motion in trajec- tory space

Ijaz Akhter, Yaser Sheikh, Sohaib Khan, and Takeo Kanade. Nonrigid structure from motion in trajec- tory space. Advances in neural information process- ing systems, 21, 2008. 3

2008
[13]

Objaverse: A universe of annotated 3d objects

Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, and Ali Farhadi. Objaverse: A universe of annotated 3d objects. arXiv preprint arXiv:2212.08051, 2022. 4, 5

2022 arXiv
[14]

Realtime multi-person 2d pose estimation using part affinity fields

Zhe Cao, Tomas Simon, Shih-En Wei, and Yaser Sheikh. Realtime multi-person 2d pose estimation using part affinity fields. In Proceedings of the IEEE 11 conference on computer vision and pattern recogni- tion, pages 7291–7299, 2017. 4

2017
[15]

Adding conditional control to text-to-image diffusion models, 2023

Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models, 2023. 4

2023
[16]

Animate-x: Univer- sal character image animation with enhanced motion representation

Shuai Tan, Biao Gong, Xiang Wang, Shiwei Zhang, Dandan Zheng, Ruobin Zheng, Kecheng Zheng, Jingdong Chen, and Ming Yang. Animate-x: Univer- sal character image animation with enhanced motion representation. arXiv preprint arXiv:2410.10306,

arXiv
[17]

Effective whole-body pose estimation with two- stages distillation

Zhendong Yang, Ailing Zeng, Chun Yuan, and Yu Li. Effective whole-body pose estimation with two- stages distillation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 4210–4220, 2023. 4

2023
[18]

3d- lfm: Lifting foundation model

Mosam Dabhi, L´ aszl´ o A Jeni, and Simon Lucey. 3d- lfm: Lifting foundation model. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10466–10475, 2024. 4, 10

2024
[19]

Pose for everything: Towards category- agnostic pose estimation

Lumin Xu, Sheng Jin, Wang Zeng, Wentao Liu, Chen Qian, Wanli Ouyang, Ping Luo, and Xiao- gang Wang. Pose for everything: Towards category- agnostic pose estimation. In European conference on computer vision, pages 398–416. Springer, 2022. 4

2022
[20]

Pose anything: A graph-based approach for category-agnostic pose es- timation

Or Hirschorn and Shai Avidan. Pose anything: A graph-based approach for category-agnostic pose es- timation. arXiv preprint arXiv:2311.17891, 2023. 4

2023 arXiv
[21]

Blender - a 3d modelling and rendering software, 2024

Blender Online Community. Blender - a 3d modelling and rendering software, 2024. Accessed: 2024-08-15. 5

2024
[22]

Learning transferable visual models from natural language supervision

Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervision. In International conference on machine learning, pa...

2021
[23]

Arbitrary style transfer in real-time with adaptive instance normal- ization

Xun Huang and Serge Belongie. Arbitrary style transfer in real-time with adaptive instance normal- ization. In Proceedings of the IEEE international conference on computer vision, pages 1501–1510,
[24]

A learned representation for artistic style

Vincent Dumoulin, Jonathon Shlens, and Manju- nath Kudlur. A learned representation for artistic style. arXiv preprint arXiv:1610.07629, 2016. 7

2016 arXiv
[25]

Group normalization

Yuxin Wu and Kaiming He. Group normalization. In Proceedings of the European conference on computer vision (ECCV), pages 3–19, 2018. 7 12

2018

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

High- resolution image synthesis with latent diffusion mod- els

Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨ orn Ommer. High- resolution image synthesis with latent diffusion mod- els. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 2, 3, 5

work page 2022

[2] [2]

Nerf: Representing scenes as neural ra- diance fields for view synthesis

Ben Mildenhall, Pratul P Srinivasan, Matthew Tan- cik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural ra- diance fields for view synthesis. Communications of the ACM, 65(1):99–106, 2021. 1, 3

work page 2021

[3] [3]

Deep unsuper- vised learning using nonequilibrium thermodynam- ics

Jascha Sohl-Dickstein, Eric Weiss, Niru Mah- eswaranathan, and Surya Ganguli. Deep unsuper- vised learning using nonequilibrium thermodynam- ics. In International conference on machine learning, pages 2256–2265. PMLR, 2015. 1

work page 2015

[4] [4]

Denois- ing diffusion probabilistic models

Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denois- ing diffusion probabilistic models. Advances in neu- ral information processing systems, 33:6840–6851,

work page

[5] [5]

Free3d: Con- sistent novel view synthesis without 3d representa- tion

Chuanxia Zheng and Andrea Vedaldi. Free3d: Con- sistent novel view synthesis without 3d representa- tion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9720–9731, 2024. 1, 3, 4, 5, 7, 9

work page 2024

[6] [6]

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

Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tokmakov, Sergey Zakharov, and Carl Vondrick. Zero-1-to-3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF international confer- ence on computer vision, pages 9298–9309, 2023. 1, 3, 5, 7

work page 2023

[7] [7]

Hugs: Human gaussian splats

Muhammed Kocabas, Jen-Hao Rick Chang, James Gabriel, Oncel Tuzel, and Anurag Ranjan. Hugs: Human gaussian splats. In Proceedings of the IEEE/CVF conference on computer vision and pat- tern recognition, pages 505–515, 2024. 2

work page 2024

[8] [8]

Stable diffusion image variations model

Lambdalabs. Stable diffusion image variations model. https://huggingface.co/lambdalabs/ sd-image-variations-diffusers , 2022. Accessed: 2024-08-15. 3, 5

work page 2022