Pith. sign in

REVIEW 4 major objections 7 minor 48 references

A Scalable Attention-Based Approach for Image-to-3D Texture Mapping

T0 review · 4 major / 7 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read A transformer predicts a full triplane texture field for any mesh from a single image in a single forward pass, eliminating UV maps and optimization.

desk verdict Solid LRM-style texture-field adaptation with a real speed win, but the fixed 55-view evaluation rig means arbitrary image-to-mesh alignment is never tested, so the headline claim is narrower than advertised. read the letter →

arxiv 2509.05131 v1 pith:URC7YRFA submitted 2025-09-05 cs.CV cs.LG

Arianna Rampini , Kanika Madan , Bruno Roy , AmirHossein Zamani , Derek Cheung This is my paper · ORCID
classification cs.CVcs.LG
keywords texturefieldtriplanetransformersingle-image3DreconstructionmeshtexturingdepthbackprojectionObjaversefeed-forward
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 texturing a known 3D mesh from a single reference image can be solved as a feed-forward prediction problem rather than a per-object optimization. The method encodes the image with a pretrained vision transformer, decodes a triplane feature field through a transformer with cross-attention, and supervises the result by backprojecting depth maps into 3D point clouds with ground-truth colors. The central result is that this UV-free, differentiable-rendering-free pipeline reconstructs textures that are more faithful to the input image and more perceptually similar to ground truth than leading diffusion-based optimization baselines, while running in about 0.2 seconds per mesh instead of tens of minutes. If correct, texture generation becomes fast enough for large-scale asset pipelines and robust to meshes that lack clean UV parametrization.

What carries the argument

The key mechanism is a triplane texture field decoded by a transformer: learned positional embeddings are passed through a transformer decoder that cross-attends to image features from a self-supervised vision encoder, then reshaped into three orthogonal feature planes (XY, XZ, YZ). Any 3D point on the mesh is projected onto the three planes, bilinearly sampled, concatenated, and decoded to RGB by a small MLP. This triplane field is supervised entirely through depth-map backprojection, which converts each pixel's depth and camera pose into a 3D query point with a ground-truth color, avoiding the need for UV maps or differentiable rasterization.

What would settle it

Run the trained model on real photographs of objects paired with meshes that are not rendered from those same views, or where depth comes from a monocular estimator instead of ground-truth precomputed depth maps; if PSNR/LPIPS on this test collapses compared to the paper's GSO numbers, the method's generality depends on the image being a known rendered view of the same mesh.

Watch

Extended reading notes

Core claim

The paper establishes that a continuous texture field over known geometry can be learned end-to-end from a single conditioning image without any UV mapping or differentiable rendering. The model, a transformer-based triplane decoder, maps image features to a 3D texture field T(p, I) that answers arbitrary 3D point queries with RGB colors. Supervision comes from depth-map backprojection: precomputed depth maps and camera poses convert each pixel into a 3D point carrying its ground-truth color, and the predicted field is penalized with a pixel-wise loss plus LPIPS. On the Google Scanned Objects benchmark the method reports CLIP-Score 90.09, LPIPS 0.075, and PSNR 27.65, substantially outperform

Load-bearing premise

The whole supervision signal assumes every conditioning image is a rendered view of the exact mesh being textured, with depth and camera pose already known; if image and mesh are not aligned this way, the learned correspondences break.

Editorial extensions

If this is right

  • Texture generation drops from minutes or hours of per-object optimization to roughly 0.2 seconds per mesh, making it practical to texture thousands of assets in a batch.
  • Because the method never requires a UV map, it can texture meshes produced by generative geometry models or real-world scans whose topology lacks a clean 2D parametrization.
  • The texture field is queried pointwise, so the same trained model works at any mesh resolution by sampling more points during inference.
  • The reported margins over diffusion-based baselines on CLIP, LPIPS, and PSNR imply that, for image-conditioned texture reconstruction, a direct feed-forward mapping can beat iterative optimization while being far faster.
  • The saturation point in model capacity (BASE over LARGE) suggests that the remaining bottleneck in texture quality is representation capacity or supervision, not just transformer size.

Reading between the lines

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

  • A natural extension, hinted at in the paper's future-work section, is to chain this fast feed-forward texture field as an initialization for a diffusion-based refinement stage, which could recover the high-frequency detail that the current model misses without sacrificing speed.
  • The depth-backprojection supervision could be made fully general by replacing precomputed ground-truth depth with monocular depth estimates during training or inference, which would let the method handle arbitrary real photographs rather than rendered views of the same mesh.
  • The same triplane field formulation could support multi-image conditioning by concatenating several cross-attention streams, potentially improving fidelity for complex or occluded objects beyond a single view.
  • Because the model learns a class-agnostic mapping from image to texture field, it may support texture transfer between different geometries that share semantic appearance, a capability the paper does not explore.
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 presents a feed-forward transformer-based method for texturing a given 3D mesh from a single image. A DINOv2 encoder extracts image features; a transformer decoder with learned positional embeddings and cross-attention produces a 32×32 triplane, upsampled to 64×64; an MLP decodes triplane features to RGB at arbitrary 3D points. Training supervises the texture field by backprojecting precomputed depth maps from 55 viewpoints and comparing predicted and ground-truth images with L2 and LPIPS losses. Inference is a single forward pass, reported at ~0.2 s per mesh with no UV mapping. Experiments compare to TEXTure, Paint3D, and EASI-Tex on GSO and Objaverse, including quantitative metrics, ablations on model size/conditioning/losses, and a user study.

Significance. If the claims hold, the method is a useful step toward practical, UV-free, fast texture generation from images, and the integration of triplanes with transformer conditioning is a reasonable architectural choice. Strengths include the 0.2 s feed-forward inference, cross-dataset evaluation on GSO, ablations, and a user study. However, the empirical support is weakened by an under-trained TEXTure baseline, missing feed-forward comparisons, and an unexamined image-mesh alignment assumption; these need to be addressed before the headline claims are fully supported.

major comments (4)
  1. [§4.1, Table 1] The text states that baselines are 'running each baseline to convergence', but the TEXTure baseline uses max train steps reduced from 10,000 to 1,000. This is 10% of the default and likely produces an undertrained model. The reported margins over TEXTure (CLIP 90.09 vs 80.24, LPIPS 0.075 vs 0.236, PSNR 27.65 vs 13.31) may therefore substantially overstate the improvement. Please provide a convergence-controlled comparison, or report TEXTure at its default step count and justify why 1,000 steps is sufficient.
  2. [§4.1, Table 1] Section 2.4 surveys feed-forward methods (LRM, Instant3D, GRM), but the experimental comparison includes only iterative optimization baselines (TEXTure, Paint3D, EASI-Tex). Given the paper's central claims are feed-forward speed and scalability, the absence of any feed-forward baseline leaves the 'state-of-the-art' claim unsupported. Please add at least one feed-forward method applicable to known-mesh texturing, or explicitly state why no such baseline is suitable.
  3. [§3.1, §3.4, §4.1] The model receives no camera pose or geometric conditioning (Section 3.1 explicitly drops geometric conditioning), yet training and evaluation data are rendered from a fixed 55-view rig. For GSO, the conditioning image is a rendered view of the same mesh, and no experiment uses a real photograph or an arbitrary viewpoint with unknown pose. The image-to-object correspondence must therefore be learned implicitly from the training-view distribution, and the paper does not test whether it transfers outside that distribution. This is load-bearing for the advertised 'single-image texture reconstruction' claim. Please run held-out-viewpoint experiments, or at minimum explicitly scope the claim to aligned rendered views.
  4. [§3.4] The training protocol is underspecified: the text says 'we process 4 random views from a set of 55 precomputed depth maps' and the loss sums over V views, but it never states which view is used as the conditioning image and whether that view is included in the four supervised views. The architecture (Section 3.3) and Figure 2 show a single conditioning image. If the conditioning view is always one of the supervised views, the model is trained to reconstruct an observed view plus nearby views, not to texture from a truly arbitrary single image. Please clarify the protocol and, ideally, exclude the conditioning view from the supervised views in evaluation.
minor comments (7)
  1. [§3.3] DINOv2 outputs patch tokens; the paper writes z∈R^768 as a single vector. Please clarify whether this is a pooled/CLS-like feature and how it is computed.
  2. [Algorithm 1] The inference loop says 'for each query point p∈M' but does not specify how query points are sampled or at what density. This affects runtime and texture quality; please specify the sampling strategy.
  3. [§4.1] For the GSO evaluation, please state explicitly whether the 10 novel views and the conditioning image are disjoint from the 55-view rig used in training/precomputation. This is important for interpreting the single-image generalization claim.
  4. [§4.2] The text says 'Our approach outperforms both baselines' but Table 1 lists three baselines (TEXTure, Paint3D, EASI-Tex). Please correct this inconsistency.
  5. [§4.4 and Figure 7] The failure cases are presented only qualitatively. A short quantitative analysis (e.g., resolution or frequency-band metrics) would make the limitation more precise. Also, the caption has a typo: 'Failure casesWhile'.
  6. [§4.4] The geometric conditioning baseline called 'Latent' [34] relies on a reference that is not described in the paper and is a self-citation. Please specify what the latent representation is and why it is a reasonable SDF feature source.
  7. [References] References [45] and [46] appear to be the same work (Zhang et al., 'The unreasonable effectiveness of deep features as a perceptual metric'). Please merge the duplicate.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity; the central claim is an empirical result with external ground-truth supervision, and the few self-citations are not load-bearing.

full rationale

The paper's claims are empirical performance claims, not derivations. The pipeline (DINOv2 encoding, cross-attention transformer, triplane decoder, depth-map backprojection querying, pixel and LPIPS losses) is fully specified and supervised by external ground-truth RGB values from precomputed render depth maps; no quantity is defined in terms of the target output. Section 3.1 explicitly acknowledges the correspondence problem, and Section 3.4 trains with known camera intrinsics and extrinsics; this limits generalization to unposed images but is an evaluated empirical setting, not a circular reduction. The only self-citations are [34] (WALa) in the geometric-conditioning ablation and [42] in related work; neither supplies the loss, benchmark, or main architectural components, and the paper actually reports that the [34]-derived conditioning does not improve results, so the citations are not load-bearing. Baselines are external official implementations and evaluation uses held-out novel views. No quoted equation reduces a prediction to its input by construction, so the correct finding is no significant circularity.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

Everything in the ledger is standard for an ML systems paper: there are no invented physical entities. The hand-set hyperparameters are minor because they are not offered as derivations. The load-bearing assumptions are the same-object image-mesh pairing, accurate depth and camera supervision, and the sufficiency of DINO features, none of which are stress-tested in the paper.

free parameters (4)
  • Loss weights lambda_pixel and lambda_perc = 1.0 each
    Set by hand in Section 3.4 with the comment that this 'provided a good balance'; no ablation of the trade-off is reported.
  • Number of conditioning views per training step = 4 of 55 precomputed views
    Chosen in Section 3.4 to ensure broad appearance coverage; no ablation of this choice.
  • Triplane resolution and feature dimension = 64x64 planes, f_dim=48
    Architecture choices from Section 3.3; the model-size ablation varies transformer width and depth but not this resolution.
  • Backprojection background masking criterion
    Depth-map background pixels are said to be masked out in Section 3.4, but the threshold or mask criterion is not stated, so the exact point sampling distribution is a free choice.
assumptions (3)
  • domain assumption The conditioning image and the target mesh depict the same physical object with consistent texture.
    The problem formulation and evaluation assume the input image is a render of the mesh whose texture is to be recovered; Section 3.2 defines the task as consistency between image and geometry, and all Objaverse and GSO evaluations use renders of the same object. No test covers the realistic case where image and mesh come from different sources.
  • domain assumption Precomputed depth maps and camera poses are accurate enough for backprojection to establish correct image-to-surface correspondences.
    The entire supervision signal in Section 3.4 is computed by inverting depth with known intrinsics and extrinsics; any depth or pose error directly mislabels queried colors, and the paper provides no sensitivity analysis.
  • domain assumption DINOv2 image features are sufficient to condition texture appearance without explicit geometry.
    The authors state in Section 3.1 and ablate in Section 4.4 that explicit geometric conditioning does not help; the final model relies entirely on DINOv2 features, so this sufficiency is load-bearing.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Scalable Attention-Based Approach for Image-to-3D Texture Mapping." pith.science (2026). https://pith.science/paper/URC7YRFA

@misc{pith2026250905131,
  author       = {Pith},
  title        = {Pith review of: A Scalable Attention-Based Approach for Image-to-3D Texture Mapping},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/URC7YRFA}},
  note         = {Machine review of arXiv:2509.05131}
}
read the original abstract

High-quality textures are critical for realistic 3D content creation, yet existing generative methods are slow, rely on UV maps, and often fail to remain faithful to a reference image. To address these challenges, we propose a transformer-based framework that predicts a 3D texture field directly from a single image and a mesh, eliminating the need for UV mapping and differentiable rendering, and enabling faster texture generation. Our method integrates a triplane representation with depth-based backprojection losses, enabling efficient training and faster inference. Once trained, it generates high-fidelity textures in a single forward pass, requiring only 0.2s per shape. Extensive qualitative, quantitative, and user preference evaluations demonstrate that our method outperforms state-of-the-art baselines on single-image texture reconstruction in terms of both fidelity to the input image and perceptual quality, highlighting its practicality for scalable, high-quality, and controllable 3D content creation.

Figures

Figures reproduced from arXiv: 2509.05131 by the authors.

Figure 1
Figure 1. Given a single input image (left), our method predicts [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. An overview of the training stage of our method. Given a single input image and a 3D mesh, we extract visual features from [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Results on Objaverse (validation set). Our feed￾forward model generalizes across diverse categories and geome￾tries, reconstructing high-fidelity textures from a single image (left). Qualitative comparison [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Comparison on GSO. Given the same conditioning image and mesh, our method (bottom row) produces textures with higher fidelity and fewer artifacts than diffusion-based baselines. 4.3. Qualitative User Study We conducted a user study to evaluate and compare our method ag…
Figure 5
Figure 5. Figure 5: Comparison with TEXTure [31] on single-image tex [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 7
Figure 7. Figure 7: Failure cases While our method produces coherent tex￾tures in most cases, it struggles with high-frequency details. Typ￾ical failure modes include handling complex patterns (top row), reconstructing legible text (middle row), and recovering unseen regions such as the b…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 36 canonical work pages

  1. [1]

    Mesh2tex: Generating mesh textures from image queries, 2023

    Alexey Bokhovkin, Shubham Tulsiani, and Angela Dai. Mesh2tex: Generating mesh textures from image queries, 2023

    work page 2023

  2. [2]

    Texfusion: Synthesizing 3d textures with text-guided image diffusion models, 2023

    Tianshi Cao, Karsten Kreis, Sanja Fidler, Nicholas Sharp, and Kangxue Yin. Texfusion: Synthesizing 3d textures with text-guided image diffusion models, 2023

    work page 2023

  3. [3]

    Efficient geometry-aware 3d gen- erative adversarial networks.” arxiv, 2021

    Eric R Chan, Connor Z Lin, Matthew A Chan, Koki Nagano, Boxiao Pan, Shalini De Mello, Orazio Gallo, LJ Guibas, J Tremblay, S Khamis, et al. Efficient geometry-aware 3d gen- erative adversarial networks.” arxiv, 2021

    work page 2021

  4. [4]

    Text2tex: Text-driven tex- ture synthesis via diffusion models, 2023

    Dave Zhenyu Chen, Yawar Siddiqui, Hsin-Ying Lee, Sergey Tulyakov, and Matthias Nießner. Text2tex: Text-driven tex- ture synthesis via diffusion models, 2023

    work page 2023

  5. [5]

    Fan- tasia3d: Disentangling geometry and appearance for high- quality text-to-3d content creation, 2023

    Rui Chen, Yongwei Chen, Ningxin Jiao, and Kui Jia. Fan- tasia3d: Disentangling geometry and appearance for high- quality text-to-3d content creation, 2023

    work page 2023

  6. [6]

    Tango: Text-driven photorealistic and robust 3d styliza- tion via lighting decomposition, 2022

    Yongwei Chen, Rui Chen, Jiabao Lei, Yabin Zhang, and Kui Jia. Tango: Text-driven photorealistic and robust 3d styliza- tion via lighting decomposition, 2022

    work page 2022

  7. [7]

    Tuvf: Learning generalizable texture uv radiance fields, 2023

    An-Chieh Cheng, Xueting Li, Sifei Liu, and Xiaolong Wang. Tuvf: Learning generalizable texture uv radiance fields, 2023

    work page 2023

  8. [8]

    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. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 13142–13153, 2023

    2023

Show all 48 references
  1. [9]

    Diffusion models beat gans on image synthesis.Advances in neural informa- tion processing systems, 34:8780–8794, 2021

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis.Advances in neural informa- tion processing systems, 34:8780–8794, 2021

    2021

  2. [10]

    Google scanned objects: A high- quality dataset of 3d scanned household items

    Laura Downs, Anthony Francis, Nate Koenig, Brandon Kin- man, Ryan Hickman, Krista Reymann, Thomas B McHugh, and Vincent Vanhoucke. Google scanned objects: A high- quality dataset of 3d scanned household items. In2022 In- ternational Conference on Robotics and Automation (ICRA)...

    2022

  3. [11]

    Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio

    Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks, 2014

    2014

  4. [12]

    Avatarclip: Zero-shot text- driven generation and animation of 3d avatars, 2022

    Fangzhou Hong, Mingyuan Zhang, Liang Pan, Zhongang Cai, Lei Yang, and Ziwei Liu. Avatarclip: Zero-shot text- driven generation and animation of 3d avatars, 2022

    2022

  5. [13]

    Lrm: Large reconstruction model for single image to 3d, 2024

    Yicong Hong, Kai Zhang, Jiuxiang Gu, Sai Bi, Yang Zhou, Difan Liu, Feng Liu, Kalyan Sunkavalli, Trung Bui, and Hao Tan. Lrm: Large reconstruction model for single image to 3d, 2024

    2024

  6. [14]

    Tex- gen: Text-guided 3d texture generation with multi-view sam- pling and resampling

    Dong Huo, Zixin Guo, Xinxin Zuo, Zhihao Shi, Juwei Lu, Peng Dai, Songcen Xu, Li Cheng, and Yee-Hong Yang. Tex- gen: Text-guided 3d texture generation with multi-view sam- pling and resampling. InEuropean Conference on Computer Vision, pages 352–368. Springer, 2024

    2024

  7. [15]

    A style-based generator architecture for generative adversarial networks, 2019

    Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks, 2019

    2019

  8. [16]

    Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model, 2023

    Jiahao Li, Hao Tan, Kai Zhang, Zexiang Xu, Fujun Luan, Yinghao Xu, Yicong Hong, Kalyan Sunkavalli, Greg Shakhnarovich, and Sai Bi. Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model, 2023

    2023

  9. [17]

    Magic3d: High-resolution text-to-3d content creation, 2023

    Chen-Hsuan Lin, Jun Gao, Luming Tang, Towaki Takikawa, Xiaohui Zeng, Xun Huang, Karsten Kreis, Sanja Fidler, Ming-Yu Liu, and Tsung-Yi Lin. Magic3d: High-resolution text-to-3d content creation, 2023

    2023

  10. [18]

    Texoct: Generating textures of 3d models with octree-based diffusion

    Jialun Liu, Chenming Wu, Xinqi Liu, Xing Liu, Jinbo Wu, Haotian Peng, Chen Zhao, Haocheng Feng, Jingtuo Liu, and Errui Ding. Texoct: Generating textures of 3d models with octree-based diffusion. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognitio...

    2024

  11. [19]

    X- mesh: Towards fast and accurate text-driven 3d stylization via dynamic textual guidance, 2023

    Yiwei Ma, Xiaioqing Zhang, Xiaoshuai Sun, Jiayi Ji, Haowei Wang, Guannan Jiang, Weilin Zhuang, and Rongrong Ji. X- mesh: Towards fast and accurate text-driven 3d stylization via dynamic textual guidance, 2023

    2023

  12. [20]

    Text2mesh: Text-driven neural stylization for meshes, 2021

    Oscar Michel, Roi Bar-On, Richard Liu, Sagie Benaim, and Rana Hanocka. Text2mesh: Text-driven neural stylization for meshes, 2021

    2021

  13. [21]

    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, 2020

    2020

  14. [22]

    Clip-mesh: Generating textured meshes from text using pretrained image-text models

    Nasir Mohammad Khalid, Tianhao Xie, Eugene Belilovsky, and Tiberiu Popa. Clip-mesh: Generating textured meshes from text using pretrained image-text models. InSIGGRAPH Asia 2022 Conference Papers, page 1–8. ACM, 2022

    2022

  15. [23]

    Instant neural graphics primitives with a multires- olution hash encoding.ACM Transactions on Graphics, 41 (4):1–15, 2022

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

    2022

  16. [24]

    Texture fields: Learning tex- ture representations in function space, 2019

    Michael Oechsle, Lars Mescheder, Michael Niemeyer, Thilo Strauss, and Andreas Geiger. Texture fields: Learning tex- ture representations in function space, 2019

    2019

  17. [25]

    Dinov2: Learning robust visual features with- out supervision, 2024

    Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Mah- moud Assran, Nicolas Ballas, Wojciech Galuba, Russell Howes, Po-Yao Huang, Shang-Wen Li, Ishan Misra, Michae...

    2024

  18. [26]

    Easi-tex: Edge-aware mesh texturing from sin- gle image.ACM Transactions on Graphics (TOG), 43(4): 1–11, 2024

    Sai Raj Kishore Perla, Yizhi Wang, Ali Mahdavi-Amiri, and Hao Zhang. Easi-tex: Edge-aware mesh texturing from sin- gle image.ACM Transactions on Graphics (TOG), 43(4): 1–11, 2024

    2024

  19. [27]

    Barron, and Ben Milden- hall

    Ben Poole, Ajay Jain, Jonathan T. Barron, and Ben Milden- hall. Dreamfusion: Text-to-3d using 2d diffusion, 2022

    2022

  20. [28]

    Pointnet: Deep learning on point sets for 3d classification and segmentation

    Charles R Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. Pointnet: Deep learning on point sets for 3d classification and segmentation. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 652–660, 2017. 10

    2017

  21. [29]

    Learning transferable visual models from natural language supervision, 2021

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

    2021

  22. [30]

    Learning transferable visual models from natural language supervi- sion

    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 supervi- sion. InInternational conference on machine learning, p...

    2021

  23. [31]

    Texture: Text-guided texturing of 3d shapes, 2023

    Elad Richardson, Gal Metzer, Yuval Alaluf, Raja Giryes, and Daniel Cohen-Or. Texture: Text-guided texturing of 3d shapes, 2023

    2023

  24. [32]

    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. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    2022

  25. [33]

    Photorealistic text-to-image diffusion models with deep language understanding.Advances in neural information processing systems, 35:36479–36494, 2022

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. Photorealistic text-to-image diffusion models with deep language understanding.Advances in neural information ...

    2022

  26. [34]

    Wavelet latent diffusion (wala): Billion-parameter 3d generative model with compact wavelet encodings.arXiv preprint arXiv:2411.08017, 2024

    Aditya Sanghi, Aliasghar Khani, Pradyumna Reddy, Arianna Rampini, Derek Cheung, Kamal Rahimi Malekshan, Kanika Madan, and Hooman Shayani. Wavelet latent diffusion (wala): Billion-parameter 3d generative model with compact wavelet encodings.arXiv preprint arXiv:2411.08017, 2024

    2024 arXiv

  27. [35]

    Texturify: Generating textures on 3d shape surfaces, 2022

    Yawar Siddiqui, Justus Thies, Fangchang Ma, Qi Shan, Matthias Nießner, and Angela Dai. Texturify: Generating textures on 3d shape surfaces, 2022

    2022

  28. [36]

    Very deep convo- lutional networks for large-scale image recognition.arXiv preprint arXiv:1409.1556, 2014

    Karen Simonyan and Andrew Zisserman. Very deep convo- lutional networks for large-scale image recognition.arXiv preprint arXiv:1409.1556, 2014

    2014 arXiv

  29. [37]

    Intex: Interactive text-to-texture syn- thesis via unified depth-aware inpainting.arXiv preprint arXiv:2403.11878, 2024

    Jiaxiang Tang, Ruijie Lu, Xiaokang Chen, Xiang Wen, Gang Zeng, and Ziwei Liu. Intex: Interactive text-to-texture syn- thesis via unified depth-aware inpainting.arXiv preprint arXiv:2403.11878, 2024

    2024 arXiv

  30. [38]

    Prolificdreamer: High-fidelity and diverse text-to-3d generation with variational score distilla- tion, 2023

    Zhengyi Wang, Cheng Lu, Yikai Wang, Fan Bao, Chongxuan Li, Hang Su, and Jun Zhu. Prolificdreamer: High-fidelity and diverse text-to-3d generation with variational score distilla- tion, 2023

    2023

  31. [39]

    Grm: Large gaussian reconstruction model for ef- ficient 3d reconstruction and generation, 2024

    Yinghao Xu, Zifan Shi, Wang Yifan, Hansheng Chen, Ceyuan Yang, Sida Peng, Yujun Shen, and Gordon Wet- zstein. Grm: Large gaussian reconstruction model for ef- ficient 3d reconstruction and generation, 2024

    2024

  32. [40]

    Texturedreamer: Image-guided texture synthesis through geometry-aware diffusion, 2024

    Yu-Ying Yeh, Jia-Bin Huang, Changil Kim, Lei Xiao, Thu Nguyen-Phuoc, Numair Khan, Cheng Zhang, Manmohan Chandraker, Carl S Marshall, Zhao Dong, and Zhengqin Li. Texturedreamer: Image-guided texture synthesis through geometry-aware diffusion, 2024

    2024

  33. [41]

    Texture generation on 3d meshes with point-uv diffusion, 2023

    Xin Yu, Peng Dai, Wenbo Li, Lan Ma, Zhengzhe Liu, and Xiaojuan Qi. Texture generation on 3d meshes with point-uv diffusion, 2023

    2023

  34. [42]

    Geometry-Aware Preference Learning for 3D Texture Generation.arXiv preprint arXiv:2506.18331, 2025

    AmirHossein Zamani, Tianhao Xie, Amir G Aghdam, Tiberiu Popa, and Eugene Belilovsky. Geometry-Aware Preference Learning for 3D Texture Generation.arXiv preprint arXiv:2506.18331, 2025

    2025

  35. [43]

    Paint3d: Paint anything 3d with lighting-less texture diffusion models, 2023

    Xianfang Zeng, Xin Chen, Zhongqi Qi, Wen Liu, Zibo Zhao, Zhibin Wang, Bin Fu, Yong Liu, and Gang Yu. Paint3d: Paint anything 3d with lighting-less texture diffusion models, 2023

    2023

  36. [44]

    Texpainter: Generative mesh texturing with multi-view consistency, 2024

    Hongkun Zhang, Zherong Pan, Congyi Zhang, Lifeng Zhu, and Xifeng Gao. Texpainter: Generative mesh texturing with multi-view consistency, 2024

    2024

  37. [45]

    Efros, Eli Shecht- man, and Oliver Wang

    Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric, 2018

    2018

  38. [46]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. InProceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 586–595, 2018

    2018

  39. [47]

    Dreammat: High-quality pbr material generation with geometry- and light-aware dif- fusion models, 2024

    Yuqing Zhang, Yuan Liu, Zhiyu Xie, Lei Yang, Zhongyuan Liu, Mengzhou Yang, Runze Zhang, Qilong Kou, Cheng Lin, Wenping Wang, and Xiaogang Jin. Dreammat: High-quality pbr material generation with geometry- and light-aware dif- fusion models, 2024

    2024

  40. [48]

    Triplane meets gaussian splatting: Fast and generalizable single-view 3d reconstruction with transformers, 2023

    Zi-Xin Zou, Zhipeng Yu, Yuan-Chen Guo, Yangguang Li, Ding Liang, Yan-Pei Cao, and Song-Hai Zhang. Triplane meets gaussian splatting: Fast and generalizable single-view 3d reconstruction with transformers, 2023. 11

    2023

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.