Pith. sign in

REVIEW 3 major objections 5 minor 44 references

HumanMaterial: Human Material Estimation from a Single Image via Progressive Training

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read HumanMaterial claims that a staged pipeline of three specialized prior models plus a joint finetuning model can estimate six physically based material maps from a single full-body photo, with a new dataset supplying the supervision.

desk verdict A useful dataset and a sensible progressive pipeline, but the finetuning model's train/test prior mismatch and reliance on self-built evaluation leave the SOTA claim weaker than advertised. read the letter →

arxiv 2507.18385 v1 pith:KA4MQBQ3 submitted 2025-07-24 cs.CV

Yu Jiang , Jiahao Xia , Jiongming Qin , Yusen Wang , Tuo Cao , Chunxia Xiao This is my paper · ORCID
classification cs.CV
keywords inverserenderingphysicallybasedmaterialestimationhumanrelightingsubsurfacescatteringprogressivetrainingPBRdatasetsingle-image
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

HumanMaterial claims that one full-body photograph carries enough information to infer six physically based material maps—surface normal, diffuse albedo, roughness, specular albedo, subsurface scattering, and displacement—provided the inference is split into stages instead of left to one end-to-end network. The paper introduces OpenHumanBRDF, a dataset built from scanned human models whose materials are assigned by category (hair, skin, fabric, leather) using statistical value ranges, rendered under many environment maps, and augmented with displacement and subsurface scattering for skin realism. Training uses three prior models, each responsible for one group of correlated maps, followed by a finetuning model that re-optimizes all maps together. A Controlled PBR Rendering loss fixes the non-target maps at physically plausible constants so that the map under training dominates the rendered-image error. The paper reports the best material and relighting PSNR among compared methods on its test set and shows relighting and material editing on real photos.

What carries the argument

The load-bearing mechanism is the Controlled PBR Rendering (CPR) loss. For each prior model, the paper renders both predicted and ground-truth materials under thirty-seven point-light illuminations, but with all maps except the one being optimized held fixed at chosen constants—low roughness and medium specular for the geometry model, high roughness and low specular for the albedo model, and GT geometry and albedo for the RSS model. This makes the target map the dominant cause of any rendered difference, so gradients flow to it rather than to confounding variables. The second mechanism is the progressive architecture: three prior models produce initial maps, a guidance encoder turns those priors into features, and a finetuning model with four decoders fuses those features with latent image features to emit the final render-ready maps.

What would settle it

Take real human subjects whose true roughness, specular albedo, and subsurface scattering are measured under calibrated, controlled illumination, run HumanMaterial on ordinary photos of the same subjects, and compare the estimated maps or the relit renderings against the measurements and against photos taken under the target lights.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central discovery is that training difficulty in multi-map inverse rendering is a balancing problem, not just a capacity problem. Because each material map influences the rendered image with different strength, an end-to-end model either overfits the dominant maps or underfits the rest. The paper claims that splitting the task into a Geometry Prior Model (normal plus displacement), an Albedo Prior Model (diffuse albedo), and an RSS Prior Model (roughness, specular albedo, subsurface scattering) gives each map dedicated supervision, and that the subsequent finetuning model restores the cross-map consistency that independence loses. The quantitative claim is state-of-the-art performance on OpenHumanBRDF and on real data, with the reported gains concentrated in roughness, subsurface scattering, and displacement—maps that prior single-image human pipelines did not produce.

Load-bearing premise

OpenHumanBRDF's ground-truth materials, which are hand-set per-category values rather than measurements of real skin, hair, fabric, and leather, must be close enough to real human appearance that training and testing on them transfers to real photographs; a secondary load-bearing premise is that training the finetuning model on ground-truth priors and testing it on predicted priors introduces no significant error propagation.

Editorial extensions

If this is right

  • If the claim holds, a single photo suffices to produce render-ready human materials for ray-traced relighting under novel, arbitrary environment illumination, without retraining a neural shader for each new light.
  • Class-level material editing becomes practical: the estimated roughness, specular, and subsurface maps carry category information, so a garment can be switched from fabric to leather and hair or cloth colors can be changed.
  • The dataset gives the field a benchmark for full-body PBR material estimation that includes displacement and subsurface scattering, which earlier human material datasets did not provide.
  • The progressive training recipe—specialized prior models plus a controlled rendering loss—can be transferred to other inverse rendering problems where multiple output maps compete for gradient signal.

Reading between the lines

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

  • The reported accuracy is measured against OpenHumanBRDF's own ground truth, so it certifies internal consistency more than physical accuracy; a real material-capture benchmark would be the natural next test.
  • Because the finetuning model is trained with ground-truth priors but tested with predicted priors, the published numbers likely understate the effect of prior errors; injecting deliberately corrupted priors during training would quantify this gap.
  • A testable extension is to replace the four hand-set material categories with per-texel measured BRDF data from real humans; the paper's own limitation section notes that the category assumption excludes composite materials such as dusty fabric.
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.

Referee Report

3 major / 5 minor

Summary. The paper introduces OpenHumanBRDF, a synthetic dataset of 147 scanned human models with six PBR material maps (normal, diffuse albedo, roughness, specular albedo, subsurface scattering, displacement), and HumanMaterial, a progressive training pipeline that first estimates material priors with three separate networks and then refines them with a finetuning model under a multi-illumination rendering loss. The authors report state-of-the-art PSNR on their dataset and show qualitative relighting and material editing results on real images.

Significance. If the results hold, the contribution is valuable: the dataset adds displacement and subsurface scattering maps that are absent from previous open human relighting datasets, and the progressive strategy addresses the difficulty of balancing many material maps in a single end-to-end model. The paper also ships (upon acceptance) a dataset that could support future work on full-body human material estimation. However, the central SOTA claim is not yet fully supported because the quantitative evaluation is limited to the authors' own synthetic dataset and because the finetuning model is trained with GT priors but evaluated with predicted priors, leaving the deployed pipeline's behavior unmeasured. The approach is credible and the presentation is generally clear; the missing experiment on prior-error propagation is the main obstacle to acceptance.

major comments (3)
  1. [IV-E, Table III] The finetuning model (FTM) is trained with ground-truth material priors as input but tested with model-predicted priors; this explicit train/test distribution mismatch is stated in Section IV-E. No experiment in Section V quantifies how the FTM's output degrades when predicted priors are substituted for GT priors. Consequently, the 'Ours' row in Table III is ambiguous: if it was produced with GT priors at test time, it does not reflect the deployed system, and if it was produced with predicted priors, the reported accuracy is unexplained given the training condition. The authors should report results under both input conditions and ideally retrain the FTM with predicted priors or with noise-injected priors to close this gap.
  2. [Table III] The quantitative support for the state-of-the-art claim rests on PSNR comparisons on the authors' own OpenHumanBRDF test set only, with single-trial measurements and no error bars. The margins over baselines are modest on key maps—normal 21.2 vs 20.5 dB, diffuse albedo 27.1 vs 26.2 dB, roughness 24.1 vs 22.9 dB—and all methods are trained and evaluated on the same synthetic distribution. Because the dataset is built from hand-picked per-category BRDF values (Table II), this evidence is primarily internal consistency rather than a demonstration of superiority on real imagery. A quantitative real-data benchmark or a cross-dataset evaluation would be necessary to support the abstract's claim.
  3. [Sec III, Table II] The ground-truth materials in OpenHumanBRDF are produced by assigning one fixed set of specular albedo, roughness, and subsurface scattering values to each of the four broad categories (hair, skin, fabric, leather) on RenderPeople scans. This is a deliberate simplification that makes dataset construction tractable, but it also means the learned material distributions are tied to the authors' chosen parameters. The qualitative results on real images (Figs. 9-11) are encouraging, but they do not quantify whether the model generalizes to real human materials that do not fall cleanly into one of the four categories or have composite properties (e.g., dusty fabrics). A quantitative evaluation on real captured materials, or a study varying the Table II values and measuring sensitivity, would strengthen the generalization claim.
minor comments (5)
  1. [IV-D, Eq. (2)] The notation Lr_r for the RSS prior model and Lm_r for the finetuning model is inconsistent with the text's claim that a CPR loss is designed for each model; clarify whether these terms denote the CPR loss in Eq. (5) or the ordinary relighting loss in Eq. (4).
  2. [IV-D] For the Geometry and Albedo prior models, the controlled material values cm' in Eq. (5) are only partially specified; please state the values used for diffuse albedo, subsurface scattering, and displacement when training the Geometry Prior Model.
  3. [III] The sentence 'The age range covered is the largest among all age groups, 19-50' is unclear; presumably the intended age range is 19 to 50 years, but the phrasing should be revised.
  4. [Table III / Fig. 4] The relationship between the RSS map (roughness, specular albedo, subsurface scattering concatenated) and the separate 'SSS' row in Table III should be clarified; it appears the table reports the three components individually, but this should be stated.
  5. [V-C] For SL, only qualitative comparison is possible because the authors do not release code; this is acknowledged, but the comparison would be stronger with a quantitative metric on real data if any public benchmark exists.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the training losses and dataset construction are self-contained, with external statistical grounding.

full rationale

I walked the paper's derivation chain and found no circular step that reduces a claimed prediction to its own inputs by construction. OpenHumanBRDF is constructed by assigning material parameter values derived from external statistical data (RTR [6]) to RenderPeople scans and baking the corresponding maps; the prior models are trained with pixel-wise L1 loss and rendering losses whose targets are those GT maps, so the supervision is independent of the fitted model parameters. The Controlled PBR Rendering loss (Eq. 5) compares renderings of predicted and GT materials while fixing non-optimized materials to either manually chosen reasonable values or GT values; this is a standard rendering-aware training loss, not a definitional equivalence between the prediction and its input. The Finetuning Model's train/test prior mismatch in Sec. IV-E (GT priors at training, predicted priors at testing) is a genuine exposure-bias and internal-validity concern, and the paper provides no error-propagation experiment, but this is an unvalidated distribution-shift assumption rather than a circularity: the final outputs are not algebraically forced to equal the input priors, and the reported ablation shows the finetuning model changes results relative to using priors alone. The paper contains no load-bearing self-citations: the cited statistical ranges, Blender, RenderPeople, and baseline methods are all external sources. The fact that all quantitative evaluation is on the authors' self-built synthetic dataset raises external-validity questions, but under the stated criteria that is not a circularity argument. I therefore report no significant circularity.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claims depend on the hand-constructed OpenHumanBRDF dataset and on several modeling choices that are not derived from first principles or externally validated: the four-category material model, the hand-set BRDF values in Table II, and the use of Disney BSDF. Additionally, the finetuning stage assumes that GT-prior training transfers to predicted priors at test time (Sec. IV-E). These are assumptions the reader is asked to accept, not results established by independent measurements.

free parameters (5)
  • Per-category specular albedo = hair 0.239, skin 0.184, fabric 0.263, leather 0.224
    Hand-set from RTR statistical ranges (Table II) to define all ground-truth materials in OpenHumanBRDF; realism of the dataset and every trained model depends on these values.
  • Per-category roughness = hair 0.500, skin 0.400, fabric 0.850, leather 0.250
    Hand-set based on the stated ordering of material roughness (Table II); no sensitivity analysis is provided.
  • Per-category subsurface scattering = skin 0.08, others 0.00
    Manually chosen; this drives the skin realism claim and the SSS estimation training target.
  • CPR controlled-material values = roughness 0.2 / specular 0.5 for geometry; roughness 0.8 / specular 0.03 for albedo
    Chosen by hand in the design of the CPR loss (Sec. IV-D) to make the target material dominate the rendering; central to the claimed benefit of the loss.
  • Rendering illumination count M and random light intensity range = M=37; intensity 3.0-8.0
    Hyperparameters for the multi-illumination rendering loss (Sec. IV-D); no study of sensitivity to these values is provided.
assumptions (4)
  • domain assumption Full-body appearance can be represented by four material categories: hair, skin, fabric, leather.
    Stated in Sec. III as "we make a reasonable assumption"; all material ground truth in OpenHumanBRDF is generated under it, and Sec. V.F admits difficulty with composite materials such as dusty fabric.
  • domain assumption RenderPeople scanned meshes with manually assigned BRDF parameters are realistic enough to train a model that generalizes to real photos.
    Sec. III describes hand adjustment of roughness, specular, and SSS per category and claims realism through visual comparison only; no measured BRDF validation is provided.
  • domain assumption The statistical BRDF ranges reported in RTR reference [6] transfer to human hair, skin, fabric, and leather.
    Sec. III sets per-category specular albedo from these ranges; this is the quantitative basis for all ground-truth material maps.
  • domain assumption The Disney BSDF with subsurface scattering captures the appearance needed for human material inference.
    Sec. IV-C chooses Disney BSDF and the rendering equation; there is no comparison against measured BTDFs or alternative shading models.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HumanMaterial: Human Material Estimation from a Single Image via Progressive Training." pith.science (2026). https://pith.science/paper/KA4MQBQ3

@misc{pith2026250718385,
  author       = {Pith},
  title        = {Pith review of: HumanMaterial: Human Material Estimation from a Single Image via Progressive Training},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KA4MQBQ3}},
  note         = {Machine review of arXiv:2507.18385}
}
read the original abstract

Full-body Human inverse rendering based on physically-based rendering aims to acquire high-quality materials, which helps achieve photo-realistic rendering under arbitrary illuminations. This task requires estimating multiple material maps and usually relies on the constraint of rendering result. The absence of constraints on the material maps makes inverse rendering an ill-posed task. Previous works alleviated this problem by building material dataset for training, but their simplified material data and rendering equation lead to rendering results with limited realism, especially that of skin. To further alleviate this problem, we construct a higher-quality dataset (OpenHumanBRDF) based on scanned real data and statistical material data. In addition to the normal, diffuse albedo, roughness, specular albedo, we produce displacement and subsurface scattering to enhance the realism of rendering results, especially for the skin. With the increase in prediction tasks for more materials, using an end-to-end model as in the previous work struggles to balance the importance among various material maps, and leads to model underfitting. Therefore, we design a model (HumanMaterial) with progressive training strategy to make full use of the supervision information of the material maps and improve the performance of material estimation. HumanMaterial first obtain the initial material results via three prior models, and then refine the results by a finetuning model. Prior models estimate different material maps, and each map has different significance for rendering results. Thus, we design a Controlled PBR Rendering (CPR) loss, which enhances the importance of the materials to be optimized during the training of prior models. Extensive experiments on OpenHumanBRDF dataset and real data demonstrate that our method achieves state-of-the-art performance.

Figures

Figures reproduced from arXiv: 2507.18385 by the authors.

Figure 1
Figure 1. HumanMaterial. Our method aims to estimate PBR materials from a [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Data realism enhancement. (a) represents the original data we [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Three samples of OpenHumanBRDF. “Appearance” means the input image of model. In “Materials”, the normal, diffuse albedo, SSR (subsurface [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: Method overview. Our method, HumanMaterial, estimates the PBR materials from a single human image, and achieves photorealistic relighting result. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: The structure of the prior model. The network of the three prior [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Three kinds of light illuminations for PBR rendering. “Fixed Single” [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Input, estimated subsurface scattering, and displacement results. The [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: Performance comparison with previous works (RADN [13], and HATSNet [14]) for PBR materials estimation on OpenHumanBRDF dataset. The [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Performance comparison with previous works (RADN [13], HATSNet [14], and SL [5]) for PBR materials estimation on real data. The “Input”, [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 10
Figure 10. Figure 10: Relighting results of the estimated materials. The estimated material maps are from Fig. 8 (left samples are on the OpenHumanBRDF dataset) and [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Performance comparison with previous works (FBHR [2], TR [1]) for relighting on OpenHumanBRDF dataset and real data. The left three are the [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 12
Figure 12. Figure 12: Ablation study for realism enhancement on OpenHumanBRDF [PITH_FULL_IMAGE:figures/full_fig_p011_12.png]
Figure 13
Figure 13. Figure 13: Ablation study on OpenHumanBRDF dataset. “N”, “D”, “R”, “S”, [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]
Figure 15
Figure 15. Figure 15: Limitation samples. “N” means normal. “D” means diffuse albedo. [PITH_FULL_IMAGE:figures/full_fig_p013_15.png]
Figure 14
Figure 14. Figure 14: Material Editing. Please see the dynamic editing process in the [PITH_FULL_IMAGE:figures/full_fig_p013_14.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 42 canonical work pages

  1. [1]

    Total relighting: Learning to relight portraits for background replacement,

    R. Pandey, S. O. Escolano, C. Legendre, C. Haene, S. Bouaziz, C. Rhe- mann, P. Debevec, and S. Fanello, “Total relighting: Learning to relight portraits for background replacement,” vol. 40, no. 4, August 2021

    work page 2021

  2. [2]

    Single-image full-body human relighting,

    M. Lagunas, X. Sun, J. Yang, R. Villegas, J. Zhang, Z. Shu, B. Ma- sia, and D. Gutierrez, “Single-image full-body human relighting,” in Eurographics Symposium on Rendering (EGSR) . The Eurographics Association, 2021

    work page 2021

  3. [3]

    All-frequency Full-body Human Image Relighting

    D. Tajima, Y . Kanamori, and Y . Endo, “All-frequency full-body human image relighting,” arXiv preprint arXiv:2411.00356 , 2024

    work page Pith review arXiv 2024

  4. [4]

    Scaling in-the-wild training for diffusion-based illumination harmonization and editing by imposing consistent light transport,

    L. Zhang, A. Rao, and M. Agrawala, “Scaling in-the-wild training for diffusion-based illumination harmonization and editing by imposing consistent light transport,” in The Thirteenth International Conference on Learning Representations , 2025

    work page 2025

  5. [5]

    Switchlight: Co-design of physics-driven architecture and pre-training framework for human portrait relighting,

    H. Kim, M. Jang, W. Yoon, J. Lee, D. Na, S. Woo, and B. Ai, “Switchlight: Co-design of physics-driven architecture and pre-training framework for human portrait relighting,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. 25 096–25 106, Jan 2024

    work page 2024

  6. [6]

    Real-time rendering,

    T. Akenine-Mo and N. Hoffman, “Real-time rendering,” 2018

    work page 2018

  7. [7]

    [Online]

    Blender Online Community, Blender - a 3D modelling and rendering package, Blender Foundation, Stichting Blender Foundation, Amsterdam, 2018. [Online]. Available: http://www.blender.org

    2018

  8. [8]

    Photorealistic monocular 3d reconstruction of humans wearing clothing,

    T. Alldieck, M. Zanfir, and C. Sminchisescu, “Photorealistic monocular 3d reconstruction of humans wearing clothing,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022

    work page 2022

Show all 44 references
  1. [9]

    Unsupervised learning for intrinsic image decomposition from a single image,

    Y . Liu, Y . Li, S. You, and F. Lu, “Unsupervised learning for intrinsic image decomposition from a single image,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2020

    2020

  2. [10]

    Single image portrait relighting,

    T. Sun, J. T. Barron, Y .-T. Tsai, Z. Xu, X. Yu, G. Fyffe, C. Rhemann, J. Busch, P. Debevec, and R. Ramamoorthi, “Single image portrait relighting,” ACM Transactions on Graphics (TOG) , p. 1–12, Aug 2019

    2019

  3. [11]

    Learning to reconstruct shape and spatially-varying reflectance from a single image,

    Z. Li, Z. Xu, R. Ramamoorthi, K. Sunkavalli, and M. Chandraker, “Learning to reconstruct shape and spatially-varying reflectance from a single image,” ACM Transactions on Graphics (TOG) , vol. 37, no. 6, dec 2018

    2018

  4. [12]

    Nero: Neural geometry and brdf reconstruction of reflective objects from multiview images,

    Y . Liu, P. Wang, C. Lin, X. Long, J. Wang, L. Liu, T. Komura, and W. Wang, “Nero: Neural geometry and brdf reconstruction of reflective objects from multiview images,” ACM Transactions on Graphics (TOG) , vol. 42, no. 4, jul 2023

    2023

  5. [13]

    Single-image svbrdf capture with a rendering-aware deep network,

    V . Deschaintre, M. Aittala, F. Durand, G. Drettakis, and A. Bousseau, “Single-image svbrdf capture with a rendering-aware deep network,” ACM Transactions on Graphics (TOG) , vol. 37, no. 4, jul 2018

    2018

  6. [14]

    Highlight-aware two-stream network for single-image svbrdf acquisi- tion,

    J. Guo, S. Lai, C. Tao, Y . Cai, L. Wang, Y . Guo, and L.-Q. Yan, “Highlight-aware two-stream network for single-image svbrdf acquisi- tion,” ACM Transactions on Graphics (TOG) , vol. 40, no. 4, jul 2021

    2021

  7. [15]

    PhySG: Inverse rendering with spherical gaussians for physics-based material editing and relighting,

    K. Zhang, F. Luan, Q. Wang, K. Bala, and N. Snavely, “PhySG: Inverse rendering with spherical gaussians for physics-based material editing and relighting,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . New York, NY , USA: Associa...

    2021

  8. [16]

    Neilf: Neural incident light field for physically-based material estimation,

    Y . Yao, J. Zhang, J. Liu, Y . Qu, T. Fang, D. McKinnon, Y . Tsin, and L. Quan, “Neilf: Neural incident light field for physically-based material estimation,” in European Conference on Computer Vision (ECCV), 2022

    2022

  9. [17]

    Efficient reflectance capture with a deep gated mixture-of-experts,

    X. Ma, Y . Yu, H. Wu, and K. Zhou, “Efficient reflectance capture with a deep gated mixture-of-experts,” IEEE Transactions on Visualization and Computer Graphics (TVCG) , vol. 30, no. 7, pp. 4246–4256, 2024

    2024

  10. [18]

    Frost-brdf: A fast and robust optimal sampling technique for brdf acquisition,

    E. Miandji, T. Tongbuasirilai, S. Hajisharif, B. Kavoosighafi, and J. Unger, “Frost-brdf: A fast and robust optimal sampling technique for brdf acquisition,” IEEE Transactions on Visualization and Computer Graphics (TVCG) , vol. 30, no. 7, pp. 4390–4402, 2024

    2024

  11. [19]

    Dpcs: Path tracing-based differ- entiable projector-camera systems,

    J. Li, Q. Deng, H. Ling, and B. Huang, “Dpcs: Path tracing-based differ- entiable projector-camera systems,” IEEE Transactions on Visualization and Computer Graphics (TVCG) , vol. 31, no. 5, pp. 3666–3676, 2025

    2025

  12. [20]

    Neural reconstruction of relightable human model from monocular video,

    W. Sun, Y . Che, H. Huang, and Y . Guo, “Neural reconstruction of relightable human model from monocular video,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 397–407

    2023

  13. [21]

    Nerfactor: Neural factorization of shape and reflectance under an unknown illumination,

    X. Zhang, P. P. Srinivasan, B. Deng, P. Debevec, W. T. Freeman, and J. T. Barron, “Nerfactor: Neural factorization of shape and reflectance under an unknown illumination,” ACM Transactions on Graphics (TOG), vol. 40, no. 6, pp. 1–18, 2021

    2021

  14. [22]

    Modeling surface appearance from a single photograph using self-augmented convolutional neural networks,

    X. Li, Y . Dong, P. Peers, and X. Tong, “Modeling surface appearance from a single photograph using self-augmented convolutional neural networks,” ACM Transactions on Graphics (TOG) , vol. 36, no. 4, jul 2017

    2017

  15. [23]

    Look-ahead training with learned reflectance loss for single-image svbrdf estimation,

    X. Zhou and N. K. Kalantari, “Look-ahead training with learned reflectance loss for single-image svbrdf estimation,” ACM Transactions on Graphics (TOG) , vol. 41, no. 6, nov 2022

    2022

  16. [24]

    Deep inverse rendering for high-resolution svbrdf estimation from an arbitrary number of images,

    D. Gao, X. Li, Y . Dong, P. Peers, K. Xu, and X. Tong, “Deep inverse rendering for high-resolution svbrdf estimation from an arbitrary number of images,” ACM Transactions on Graphics (TOG) , vol. 38, no. 4, jul 2019

    2019

  17. [25]

    Materialgan: Reflectance capture using a generative svbrdf model,

    Y . Guo, C. Smith, M. Ha ˇsan, K. Sunkavalli, and S. Zhao, “Materialgan: Reflectance capture using a generative svbrdf model,”ACM Transactions on Graphics (TOG) , vol. 39, no. 6, nov 2020

    2020

  18. [26]

    Modeling indirect illumination for inverse rendering,

    Y . Zhang, J. Sun, X. He, H. Fu, R. Jia, and X. Zhou, “Modeling indirect illumination for inverse rendering,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . New York, NY , USA: Association for Computing Machinery, 2022, pp. 18 643–18 652

    2022

  19. [27]

    Relightify: Relightable 3d faces from a single image via diffusion models,

    F. Paraperas Papantoniou, A. Lattas, S. Moschoglou, and S. Zafeiriou, “Relightify: Relightable 3d faces from a single image via diffusion models,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , 2023

    2023

  20. [28]

    Tensoir: Tensorial inverse rendering,

    H. Jin, I. Liu, P. Xu, X. Zhang, S. Han, S. Bi, X. Zhou, Z. Xu, and H. Su, “Tensoir: Tensorial inverse rendering,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 2023

    2023

  21. [29]

    Tensorf: Tensorial radiance fields,

    A. Chen, Z. Xu, A. Geiger, J. Yu, and H. Su, “Tensorf: Tensorial radiance fields,” in European Conference on Computer Vision (ECCV) , 2022

    2022

  22. [30]

    Nerd: Neural reflectance decomposition from image collections,

    M. Boss, R. Braun, V . Jampani, J. T. Barron, C. Liu, and H. P. Lensch, “Nerd: Neural reflectance decomposition from image collections,” in IEEE International Conference on Computer Vision (ICCV) , 2021

    2021

  23. [31]

    Extracting Triangular 3D Models, Materials, and Lighting From Images,

    J. Munkberg, J. Hasselgren, T. Shen, J. Gao, W. Chen, A. Evans, T. M ¨uller, and S. Fidler, “Extracting Triangular 3D Models, Materials, and Lighting From Images,” in Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR) , June 2022, pp. 8280–8290

    2022

  24. [32]

    Factored-neus: Reconstructing surfaces, illumination, and materials of possibly glossy objects,

    Y . Fan, I. Skorokhodov, O. V oynov, S. Ignatyev, E. Burnaev, P. Wonka, and Y . Wang, “Factored-neus: Reconstructing surfaces, illumination, and materials of possibly glossy objects,” arXiv preprint arXiv:2305.17929 , 2023

    2023 arXiv

  25. [33]

    Avatarme++: Facial shape and brdf inference with photo- realistic rendering-aware gans,

    A. Lattas, S. Moschoglou, S. Ploumpis, B. Gecer, A. Ghosh, and S. Zafeiriou, “Avatarme++: Facial shape and brdf inference with photo- realistic rendering-aware gans,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 44, no. 12, pp. 9269–9284, 2021

    2021

  26. [34]

    Avatarme: Realistically renderable 3d facial reconstruction

    A. Lattas, S. Moschoglou, B. Gecer, S. Ploumpis, V . Triantafyllou, A. Ghosh, and S. Zafeiriou, “Avatarme: Realistically renderable 3d facial reconstruction” in-the-wild”,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 2020, pp. 760–769

    2020

  27. [35]

    Portrait lighting transfer using a mass transport approach,

    Z. Shu, S. Hadap, E. Shechtman, K. Sunkavalli, S. Paris, and D. Samaras, “Portrait lighting transfer using a mass transport approach,” ACM Transactions on Graphics (TOG) , p. 1, Jul 2017

    2017

  28. [36]

    Relit-neulf: Efficient novel view synthesis with neural 4d light field,

    Z. Li, L. Song, C. Liu, J. Yuan, and Y . Xu, “Relit-neulf: Efficient novel view synthesis with neural 4d light field,” in Proceedings of the 31st ACM International Conference on Multimedia , 2023

    2023

  29. [37]

    High- res facial appearance capture from polarized smartphone images,

    D. Azinovi ´c, O. Maury, C. Hery, M. Nießner, and J. Thies, “High- res facial appearance capture from polarized smartphone images,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , June 2023

    2023

  30. [38]

    Relighting4d: Neural relightable human from videos,

    Z. Chen and Z. Liu, “Relighting4d: Neural relightable human from videos,” in European Conference on Computer Vision . Springer, 2022, pp. 606–623

    2022

  31. [39]

    Relightable and animatable neural avatar from sparse-view video,

    Z. Xu, S. Peng, C. Geng, L. Mou, Z. Yan, J. Sun, H. Bao, and X. Zhou, “Relightable and animatable neural avatar from sparse-view video,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 2024, pp. 990–1000

    2024

  32. [40]

    HDR environment images,

    PolyHaven, “HDR environment images,” https://polyhaven.com/hdris

  33. [41]

    Renderpeople,

    RenderPeople Team, “Renderpeople,” https://renderpeople.com, 2021

    2021

  34. [42]

    Gatis, 2020

    D. Gatis, 2020. [Online]. Available: https://github.com/danielgatis/rembg

    2020

  35. [43]

    Physically-based shading at disney,

    B. Burley and W. D. A. Studios, “Physically-based shading at disney,” in Acm Siggraph , vol. 2012. vol. 2012, 2012, pp. 1–7

    2012

  36. [44]

    Tensorflow: a system for large-scale machine learning,

    M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, M. Kudlur, J. Levenberg, R. Monga, S. Moore, D. G. Murray, B. Steiner, P. Tucker, V . Vasudevan, P. Warden, M. Wicke, Y . Yu, and X. Zheng, “Tensorflow: a system for large-sca...

    2016

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.