pith. sign in

arxiv: 2605.24915 · v1 · pith:NOYVOU2Knew · submitted 2026-05-24 · 💻 cs.GR · cs.CV

Snapshot Polarimetric Display Inverse Rendering

Seokjun Choi , Yunseong Moon , Kaizhang Kang , Hoon-Gyu Chung , Jin-Nyeong Kim , Giljoo Nam , Seung-Hwan Baek This is my paper

Pith reviewed 2026-06-29 23:58 UTC · model grok-4.3

classification 💻 cs.GR cs.CV
keywords inverse renderingpolarimetrysnapshot captureBRDF expansionmaterial estimationtransformer networkLCD displaypolarization camera
0
0 comments X

The pith

A single linearly polarized RGB binary pattern projected by an LCD and captured by a polarization camera with a quarter-wave plate supplies spectro-polarimetric data that a feed-forward transformer converts to per-pixel normals, albedo, rou

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

The paper establishes that inverse rendering can be performed from a single snapshot by combining an LCD projector that displays a linearly polarized RGB binary pattern with an RGB polarization camera augmented by a quarter-wave plate. These measurements provide enough spectro-polarimetric information per frame for a feed-forward transformer to estimate the four material properties at every pixel. To train the transformer despite limited real polarimetric BRDF measurements, the authors expand the available data with a generative manifold. A sympathetic reader would care because the method targets lightweight desktop workflows where multiple shots or heavy temporal modulation are impractical, and the reported real-world evaluations show it outperforms prior single-shot approaches across diverse scenes.

Core claim

By projecting a linearly polarized RGB binary pattern from an LCD and acquiring measurements with an RGB polarization camera plus quarter-wave plate, the system obtains single-shot spectro-polarimetric observations that a feed-forward transformer directly maps to per-pixel normal, albedo, roughness, and metallicity estimates; the transformer is trained on polarimetric BRDFs expanded from a small measured set via a generative manifold, and real desktop experiments confirm accurate recovery that exceeds existing methods.

What carries the argument

The feed-forward transformer that maps single-shot spectro-polarimetric measurements to per-pixel material properties, trained on data expanded by the generative manifold from measured polarimetric BRDFs.

Load-bearing premise

The generative manifold can accurately expand a limited set of measured polarimetric BRDFs to create sufficient and realistic training data for the feed-forward transformer to generalize to real-world scenes.

What would settle it

If the recovered properties on a test scene containing a material whose polarimetric response lies outside the generative manifold's expansion deviate substantially from ground-truth measurements obtained with a multi-shot reference method, the central claim would be falsified.

Figures

Figures reproduced from arXiv: 2605.24915 by Giljoo Nam, Hoon-Gyu Chung, Jin-Nyeong Kim, Kaizhang Kang, Seokjun Choi, Seung-Hwan Baek, Yunseong Moon.

Figure 1
Figure 1. Figure 1: Display–camera imaging system. (a) Polarizers are mounted on both the illumination and the sensor, enabling polarimetric imaging. (b) The illumination pattern encodes distinct shading cues across the RGB channels. (c) A captured image decomposed into unpolarized, linearly polarized (LP), circularly polarized (CP) components. to additionally access circular polarization cues. Under the chosen axis alignment… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of feed-forward inverse rendering. Our framework takes a single spectro￾polarimetrically encoded RAW image as input and decomposes it into nine measurements spanning RGB lighting directions and polarization states. An encoder–decoder transformer then estimates PBR parameter maps in a feed-forward manner. All quantities above are defined at the pixel level; we compute them independently for each pi… view at source ↗
Figure 3
Figure 3. Figure 3: Overview of the expanded pBRDF dataset generation pipeline. (a) Principal components and corresponding weights are extracted from measured pBRDF dataset via PCA, separately for intensity and polarimetric components. (b) A weight generator is trained to predict PCA weights conditioned on input PBR parameters. (c) After training, the generator samples weights from randomly sampled PBR parameters, and synthet… view at source ↗
Figure 4
Figure 4. Figure 4: Expanded pBRDF instances. Rendered examples of synthesized pBRDFs showing diverse reflectance. (a) RGB appearance under unpolarized illumination. (b) Polarimetric rendering visualized as AoLP (Angle of Linear Polarization), DoP (Degree of Polarization), and CoP (Chirality of Polarization). (c) PBR parameter maps fitted to each synthesized pBRDF and used as ground-truth supervision during training. Tab. 1: … view at source ↗
Figure 6
Figure 6. Figure 6: Diffuse–specular disambiguation. Single-image configurations (b), (c), (d) cannot observe lighting-dependent specular variations and therefore tend to entangle them with diffuse re￾flectance. This ambiguity propagates to the over￾all parameter estimation, producing rendered images that deviate noticeably from the target. In contrast, our method embeds lighting variation into the spec￾tral channels of a sin… view at source ↗
Figure 7
Figure 7. Figure 7: Metallic disambiguation. Methods that do not exploit CP cues struggle to distinguish the metallic parts of the object. Input Est. roughness Input Est. albedo Expanded pBRDF Analytic pBRDF Expanded pBRDF Analytic pBRDF [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Ablation on the expanded pBRDF dataset. Unlike our measured-pBRDF-based expan￾sion, training on analytic pBRDFs fail to faithfully disentangle real–world light–material interactions. 12 [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: Relighting under environment maps. We render the recovered PBR maps of real captured objects under several environment maps that differ substantially from the display illumination used dur￾ing capture. Our method preserves consistent mate￾rial identity and produces coherently oriented high￾lights, yielding relighting results that blend naturally with the composited backgrounds without method￾specific post… view at source ↗
Figure 11
Figure 11. Figure 11 [PITH_FULL_IMAGE:figures/full_fig_p013_11.png] view at source ↗
read the original abstract

Inverse rendering remains a core challenge in graphics and vision, especially in the snapshot configurations required for lightweight desktop workflows, where the per-frame information budget is highly constrained. Previous inverse rendering work explores various available dimensions for enriching the per-shot information, including temporal modulation, spectral encoding, and polarization. In this work, we introduce polarimetric display inverse rendering, using an LCD to project a linearly polarized RGB binary pattern and an RGB polarization camera augmented with a quarter-wave plate to acquire spectro-polarimetric measurements in a single shot. A feed-forward transformer maps these measurements to per-pixel normal, albedo, roughness, and metallicity. To overcome training data scarcity, we expand a limited set of measured polarimetric bidirectional reflectance distribution functions via a generative manifold. Evaluations on a real desktop setup demonstrate accurate inverse rendering across diverse scenes, outperforming existing approaches.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper introduces polarimetric display inverse rendering for single-shot capture: an LCD projects linearly polarized RGB binary patterns while an RGB polarization camera augmented with a quarter-wave plate acquires spectro-polarimetric measurements. A feed-forward transformer maps these measurements to per-pixel normal, albedo, roughness, and metallicity. Training data scarcity is addressed by expanding a limited set of measured polarimetric BRDFs via a generative manifold. Real desktop evaluations are claimed to demonstrate accurate inverse rendering across diverse scenes, outperforming existing approaches.

Significance. If the central claim holds, the combination of polarization encoding with generative manifold data expansion could enable practical lightweight single-shot inverse rendering for desktop graphics and vision workflows, reducing reliance on multi-shot or lower-accuracy methods.

major comments (2)
  1. [Abstract] Abstract: the claim of 'accurate inverse rendering across diverse scenes, outperforming existing approaches' is presented without any validation metrics, error analysis, comparison tables, or quantitative results, which is load-bearing for assessing whether the feed-forward transformer generalizes from manifold-generated data to real captures.
  2. [Evaluation section] Evaluation section (implied by abstract claims): no held-out real BRDF reconstruction error, distribution distances (e.g., between manifold samples and real polarimetric statistics), or ablation removing the generative manifold is reported to test whether the manifold expansion avoids domain gap in cross-talk between linear polarization, quarter-wave plate effects, and RGB channels; this assumption is the least secure link for the real-world accuracy claim.
minor comments (1)
  1. [Abstract] Abstract: the term 'spectro-polarimetric measurements' is used without clarifying the exact spectral sampling or polarization state encoding, which could be clarified for readers.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We agree that strengthening the quantitative support in the abstract and evaluation sections will improve the manuscript, and we will revise accordingly to address the concerns about validation metrics and ablations.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim of 'accurate inverse rendering across diverse scenes, outperforming existing approaches' is presented without any validation metrics, error analysis, comparison tables, or quantitative results, which is load-bearing for assessing whether the feed-forward transformer generalizes from manifold-generated data to real captures.

    Authors: We acknowledge that the abstract as written summarizes results without embedding specific metrics. In the revised version we will add concise quantitative results (e.g., mean angular error, albedo RMSE, and comparison deltas versus baselines) directly into the abstract to make the generalization claim verifiable from the abstract alone. revision: yes

  2. Referee: [Evaluation section] Evaluation section (implied by abstract claims): no held-out real BRDF reconstruction error, distribution distances (e.g., between manifold samples and real polarimetric statistics), or ablation removing the generative manifold is reported to test whether the manifold expansion avoids domain gap in cross-talk between linear polarization, quarter-wave plate effects, and RGB channels; this assumption is the least secure link for the real-world accuracy claim.

    Authors: We agree these specific analyses are missing and constitute the weakest link in the current evidence. The revised evaluation section will include (1) held-out real BRDF reconstruction errors, (2) distribution-distance statistics between manifold-augmented and measured polarimetric data, and (3) an ablation that removes the generative manifold, with explicit discussion of polarization cross-talk and quarter-wave-plate effects. revision: yes

Circularity Check

0 steps flagged

No circularity in derivation chain

full rationale

The paper presents a pipeline that acquires real spectro-polarimetric measurements via LCD projection and polarization camera, expands a limited set of measured polarimetric BRDFs using a generative manifold to create training data, and trains a feed-forward transformer to regress per-pixel material parameters. No equations, derivations, or self-citations are described that reduce any claimed prediction or output to an input by construction. The evaluations on real desktop captures provide an external benchmark independent of the training data generation process. The generative manifold step is an empirical data-augmentation technique whose validity is testable against held-out real measurements rather than being tautological.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

Based solely on the abstract, the central claim depends on the effectiveness of the generative manifold for data expansion and the transformer's generalization ability; no free parameters are explicitly named, but the approach assumes standard polarimetric imaging models hold.

axioms (2)
  • domain assumption Polarimetric BRDFs measured from limited real samples can be expanded via generative manifold to produce realistic training data
    Invoked to overcome training data scarcity for the transformer
  • domain assumption Single-shot spectro-polarimetric measurements contain sufficient information to recover per-pixel normal, albedo, roughness, and metallicity
    Underlying the feed-forward mapping claim
invented entities (1)
  • generative manifold for polarimetric BRDFs no independent evidence
    purpose: Expand limited measured data to train the transformer
    Introduced in the abstract to address data scarcity; no independent evidence provided

pith-pipeline@v0.9.1-grok · 5686 in / 1589 out tokens · 34838 ms · 2026-06-29T23:58:33.124579+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

62 extracted references · 4 canonical work pages · 1 internal anchor

  1. [1]

    Practical svbrdf capture in the frequency domain

    Miika Aittala, Tim Weyrich, and Jaakko Lehtinen. Practical svbrdf capture in the frequency domain. ACM Trans. Graph., 32 0 (4): 0 110--1, 2013

    2013

  2. [2]

    Polarimetric spatio-temporal light transport probing

    Seung-Hwan Baek and Felix Heide. Polarimetric spatio-temporal light transport probing. ACM Transactions on Graphics (TOG), 40 0 (6): 0 1--18, 2021

    2021

  3. [3]

    Simultaneous acquisition of polarimetric svbrdf and normals

    Seung-Hwan Baek, Daniel S Jeon, Xin Tong, and Min H Kim. Simultaneous acquisition of polarimetric svbrdf and normals. ACM Trans. Graph., 37 0 (6): 0 268, 2018

    2018

  4. [4]

    Image-based acquisition and modeling of polarimetric reflectance

    Seung-Hwan Baek, Tizian Zeltner, Hyunjin Ku, Inseung Hwang, Xin Tong, Wenzel Jakob, and Min H Kim. Image-based acquisition and modeling of polarimetric reflectance. ACM Trans. Graph., 39 0 (4): 0 139, 2020

    2020

  5. [5]

    Physically-based shading at disney

    Brent Burley. Physically-based shading at disney. In Acm siggraph, volume 2012, pages 1--7. vol. 2012, 2012

    2012

  6. [6]

    Pisr: Polarimetric neural implicit surface reconstruction for textureless and specular objects

    Guangcheng Chen, Yicheng He, Li He, and Hong Zhang. Pisr: Polarimetric neural implicit surface reconstruction for textureless and specular objects. In European Conference on Computer Vision, pages 205--222. Springer, 2024

    2024

  7. [7]

    Intrinsicanything: Learning diffusion priors for inverse rendering under unknown illumination

    Xi Chen, Sida Peng, Dongchen Yang, Yuan Liu, Bowen Pan, Chengfei Lv, and Xiaowei Zhou. Intrinsicanything: Learning diffusion priors for inverse rendering under unknown illumination. In European Conference on Computer Vision, pages 450--467. Springer, 2025 a

    2025

  8. [8]

    Uni-renderer: Unifying rendering and inverse rendering via dual stream diffusion

    Zhifei Chen, Tianshuo Xu, Wenhang Ge, Leyi Wu, Dongyu Yan, Jing He, Luozhou Wang, Lu Zeng, Shunsi Zhang, and Ying-Cong Chen. Uni-renderer: Unifying rendering and inverse rendering via dual stream diffusion. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 26504--26513, 2025 b

    2025

  9. [9]

    Differentiable display photometric stereo

    Seokjun Choi, Seungwoo Yoon, Giljoo Nam, Seungyong Lee, and Seung-Hwan Baek. Differentiable display photometric stereo. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11831--11840, 2024

    2024

  10. [10]

    A real-world display inverse rendering dataset

    Seokjun Choi, Hoon-Gyu Chung, Yujin Jeon, Giljoo Nam, and Seung-Hwan Baek. A real-world display inverse rendering dataset. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 25272--25283, 2025

    2025

  11. [11]

    Differentiable point-based inverse rendering

    Hoon-Gyu Chung, Seokjun Choi, and Seung-Hwan Baek. Differentiable point-based inverse rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

    2024

  12. [12]

    Differentiable inverse rendering with interpretable basis brdfs

    Hoon-Gyu Chung, Seokjun Choi, and Seung-Hwan Baek. Differentiable inverse rendering with interpretable basis brdfs. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 475--484, 2025

    2025

  13. [13]

    Conditions for the physical realisability of matrix operators in polarimetry

    Shane R Cloude. Conditions for the physical realisability of matrix operators in polarimetry. In Polarization Considerations for Optical Systems II, volume 1166, pages 177--187. SPIE, 1990

    1990

  14. [14]

    Field guide to polarization, volume 15

    Edward Collett. Field guide to polarization, volume 15. SPIE press Bellingham, Washington, 2005

    2005

  15. [15]

    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. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 13142--13153, 2023

    2023

  16. [16]

    Deep polarization imaging for 3d shape and svbrdf acquisition

    Valentin Deschaintre, Yiming Lin, and Abhijeet Ghosh. Deep polarization imaging for 3d shape and svbrdf acquisition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 15567--15576, 2021

    2021

  17. [17]

    Single-shot reflectance measurement from polarized color gradient illumination

    Graham Fyffe and Paul Debevec. Single-shot reflectance measurement from polarized color gradient illumination. In 2015 IEEE International Conference on Computational Photography (ICCP), pages 1--10. IEEE, 2015

    2015

  18. [18]

    Estimating specular roughness and anisotropy from second order spherical gradient illumination

    Abhijeet Ghosh, Tongbo Chen, Pieter Peers, Cyrus A Wilson, and Paul Debevec. Estimating specular roughness and anisotropy from second order spherical gradient illumination. In Computer Graphics Forum, volume 28, pages 1161--1170. Wiley Online Library, 2009

    2009

  19. [19]

    Circularly polarized spherical illumination reflectometry

    Abhijeet Ghosh, Tongbo Chen, Pieter Peers, Cyrus A Wilson, and Paul Debevec. Circularly polarized spherical illumination reflectometry. In ACM SIGGRAPH Asia 2010 Papers, pages 1--12. ACM, 2010

    2010

  20. [20]

    A simple necessary and sufficient condition on physically realizable mueller matrices

    Clark R Givens and Alexander B Kostinski. A simple necessary and sufficient condition on physically realizable mueller matrices. Journal of Modern Optics, 40 0 (3): 0 471--481, 1993

    1993

  21. [21]

    Neural lightrig: Unlocking accurate object normal and material estimation with multi-light diffusion

    Zexin He, Tengfei Wang, Xin Huang, Xingang Pan, and Ziwei Liu. Neural lightrig: Unlocking accurate object normal and material estimation with multi-light diffusion. arXiv preprint arXiv:2412.09593, 2024

    work page arXiv 2024

  22. [22]

    guest-host interactions in nematic liquid crystals

    George H Heilmeier and LA Zanoni. guest-host interactions in nematic liquid crystals. a new electro-optic effect. Applied Physics Letters, 13 0 (3): 0 91--92, 1968

    1968

  23. [23]

    Spiders: Structured polarization for invisible depth and reflectance sensing

    Tomoki Ichikawa, Shohei Nobuhara, and Ko Nishino. Spiders: Structured polarization for invisible depth and reflectance sensing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

    2024

  24. [24]

    Scalable, detailed and mask-free universal photometric stereo

    Satoshi Ikehata. Scalable, detailed and mask-free universal photometric stereo. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13198--13207, 2023

    2023

  25. [25]

    Physics-free spectrally multiplexed photometric stereo under unknown spectral composition

    Satoshi Ikehata and Yuta Asano. Physics-free spectrally multiplexed photometric stereo under unknown spectral composition. arXiv preprint arXiv:2410.20716, 2024

    work page arXiv 2024

  26. [26]

    Mitsuba 3 renderer, 2022

    Wenzel Jakob, Sébastien Speierer, Nicolas Roussel, Merlin Nimier-David, Delio Vicini, Tizian Zeltner, Baptiste Nicolet, Miguel Crespo, Vincent Leroy, and Ziyi Zhang. Mitsuba 3 renderer, 2022. https://mitsuba-renderer.org

    2022

  27. [27]

    Tensoir: Tensorial inverse rendering

    Haian Jin, Isabella Liu, Peijia Xu, Xiaoshuai Zhang, Songfang Han, Sai Bi, Xiaowei Zhou, Zexiang Xu, and Hao Su. Tensoir: Tensorial inverse rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 165--174, 2023

    2023

  28. [28]

    Diffuse-specular separation using binary spherical gradient illumination

    Christos Kampouris, Stefanos Zafeiriou, and Abhijeet Ghosh. Diffuse-specular separation using binary spherical gradient illumination. EGSR (EI&I), 1 0 (10), 2018

    2018

  29. [29]

    Learning efficient illumination multiplexing for joint capture of reflectance and shape

    Kaizhang Kang, Cihui Xie, Chengan He, Mingqi Yi, Minyi Gu, Zimin Chen, Kun Zhou, and Hongzhi Wu. Learning efficient illumination multiplexing for joint capture of reflectance and shape. ACM Trans. Graph., 38 0 (6): 0 165--1, 2019

    2019

  30. [30]

    Neural reflectance capture in the view-illumination domain

    Kaizhang Kang, Minyi Gu, Cihui Xie, Xuanda Yang, Hongzhi Wu, and Kun Zhou. Neural reflectance capture in the view-illumination domain. IEEE Transactions on Visualization and Computer Graphics, 29 0 (2): 0 1450--1462, 2021

    2021

  31. [31]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk \"u hler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph., 42 0 (4): 0 139--1, 2023

    2023

  32. [32]

    Practical and scalable desktop-based high-quality facial capture

    Alexandros Lattas, Yiming Lin, Jayanth Kannan, Ekin Ozturk, Luca Filipi, Giuseppe Claudio Guarnera, Gaurav Chawla, and Abhijeet Ghosh. Practical and scalable desktop-based high-quality facial capture. In European Conference on Computer Vision, pages 522--537. Springer, 2022

    2022

  33. [33]

    Neisf: Neural incident stokes field for geometry and material estimation

    Chenhao Li, Taishi Ono, Takeshi Uemori, Hajime Mihara, Alexander Gatto, Hajime Nagahara, and Yusuke Moriuchi. Neisf: Neural incident stokes field for geometry and material estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21434--21445, 2024

    2024

  34. [34]

    Neisf++: Neural incident stokes field for polarized inverse rendering of conductors and dielectrics

    Chenhao Li, Taishi Ono, Takeshi Uemori, Sho Nitta, Hajime Mihara, Alexander Gatto, Hajime Nagahara, and Yusuke Moriuchi. Neisf++: Neural incident stokes field for polarized inverse rendering of conductors and dielectrics. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 26493--26503, 2025 a

    2025

  35. [35]

    Light of normals: Unified feature representation for universal photometric stereo

    Hong Li, Houyuan Chen, Chongjie Ye, Zhaoxi Chen, Bohan Li, Shaocong Xu, Xianda Guo, Xuhui Liu, Yikai Wang, Baochang Zhang, et al. Light of normals: Unified feature representation for universal photometric stereo. 2026

    2026

  36. [36]

    Neural reflectance for shape recovery with shadow handling

    Junxuan Li and Hongdong Li. Neural reflectance for shape recovery with shadow handling. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16221--16230, 2022

    2022

  37. [37]

    Lirm: Large inverse rendering model for progressive reconstruction of shape, materials and view-dependent radiance fields

    Zhengqin Li, Dilin Wang, Ka Chen, Zhaoyang Lv, Thu Nguyen-Phuoc, Milim Lee, Jia-Bin Huang, Lei Xiao, Yufeng Zhu, Carl S Marshall, et al. Lirm: Large inverse rendering model for progressive reconstruction of shape, materials and view-dependent radiance fields. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 505--517, 2025 b

    2025

  38. [38]

    Diffusion renderer: Neural inverse and forward rendering with video diffusion models

    Ruofan Liang, Zan Gojcic, Huan Ling, Jacob Munkberg, Jon Hasselgren, Chih-Hao Lin, Jun Gao, Alexander Keller, Nandita Vijaykumar, Sanja Fidler, et al. Diffusion renderer: Neural inverse and forward rendering with video diffusion models. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 26069--26080, 2025

    2025

  39. [39]

    Shape and material capture at home

    Daniel Lichy, Jiaye Wu, Soumyadip Sengupta, and David W Jacobs. Shape and material capture at home. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6123--6133, 2021

    2021

  40. [40]

    Single-shot facial capture using polarized rgb sinusoidal illumination

    Arvin Lin and Abhijeet Ghosh. Single-shot facial capture using polarized rgb sinusoidal illumination. In Proceedings of the SIGGRAPH Asia 2025 Conference Papers, pages 1--11, 2025

    2025

  41. [41]

    o fer, Thomas Leimk \

    Linjie Lyu, Ayush Tewari, Marc Habermann, Shunsuke Saito, Michael Zollh \"o fer, Thomas Leimk \"u hler, and Christian Theobalt. Diffusion posterior illumination for ambiguity-aware inverse rendering. ACM Transactions on Graphics (TOG), 42 0 (6): 0 1--14, 2023

    2023

  42. [42]

    Sfpuel: Shape from polarization under unknown environment light

    Youwei Lyu, Heng Guo, Kailong Zhang, Si Li, and Boxin Shi. Sfpuel: Shape from polarization under unknown environment light. Advances in Neural Information Processing Systems, 37: 0 97184--97202, 2024

    2024

  43. [43]

    Rapid acquisition of specular and diffuse normal maps from polarized spherical gradient illumination

    Wan-Chun Ma, Tim Hawkins, Pieter Peers, Charles-Felix Chabert, Malte Weiss, Paul E Debevec, et al. Rapid acquisition of specular and diffuse normal maps from polarized spherical gradient illumination. Rendering Techniques, 9 0 (10): 0 2, 2007

    2007

  44. [44]

    Free-form scanning of non-planar appearance with neural trace photography

    Xiaohe Ma, Kaizhang Kang, Ruisheng Zhu, Hongzhi Wu, and Kun Zhou. Free-form scanning of non-planar appearance with neural trace photography. ACM Transactions on Graphics (TOG), 40 0 (4): 0 1--13, 2021

    2021

  45. [45]

    Nerf: Representing scenes as neural radiance fields for view synthesis

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view synthesis. Communications of the ACM, 65 0 (1): 0 99--106, 2021

    2021

  46. [46]

    Hyperspectral polarimetric brdfs of real-world materials

    Yunseong Moon, Ryota Maeda, Suhyun Shin, Inseung Hwang, Youngchan Kim, Min H Kim, and Seung-Hwan Baek. Hyperspectral polarimetric brdfs of real-world materials. In Proceedings of the SIGGRAPH Asia 2025 Conference Papers, pages 1--11, 2025

    2025

  47. [47]

    On optimal, minimal brdf sampling for reflectance acquisition

    Jannik Boll Nielsen, Henrik Wann Jensen, and Ravi Ramamoorthi. On optimal, minimal brdf sampling for reflectance acquisition. ACM Transactions on Graphics (TOG), 34 0 (6): 0 1--11, 2015

    2015

  48. [48]

    DINOv2: Learning Robust Visual Features without Supervision

    Maxime Oquab, Timoth \'e e Darcet, Th \'e o Moutakanni, Huy Vo, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  49. [49]

    Vision transformers for dense prediction

    Ren \'e Ranftl, Alexey Bochkovskiy, and Vladlen Koltun. Vision transformers for dense prediction. In Proceedings of the IEEE/CVF international conference on computer vision, pages 12179--12188, 2021

    2021

  50. [50]

    A versatile parameterization for measured material manifolds

    Cyril Soler, Kartic Subr, and Derek Nowrouzezahrai. A versatile parameterization for measured material manifolds. In Computer graphics forum, volume 37, pages 135--144. Wiley Online Library, 2018

    2018

  51. [51]

    Connecting measured brdfs to analytic brdfs by data-driven diffuse-specular separation

    Tiancheng Sun, Henrik Wann Jensen, and Ravi Ramamoorthi. Connecting measured brdfs to analytic brdfs by data-driven diffuse-specular separation. ACM Transactions on Graphics (TOG), 37 0 (6): 0 1--15, 2018

    2018

  52. [52]

    Light stage super-resolution: continuous high-frequency relighting

    Tiancheng Sun, Zexiang Xu, Xiuming Zhang, Sean Fanello, Christoph Rhemann, Paul Debevec, Yun-Ta Tsai, Jonathan T Barron, and Ravi Ramamoorthi. Light stage super-resolution: continuous high-frequency relighting. ACM Transactions on Graphics (TOG), 39 0 (6): 0 1--12, 2020

    2020

  53. [53]

    Microfacet models for refraction through rough surfaces

    Bruce Walter, Stephen R Marschner, Hongsong Li, and Kenneth E Torrance. Microfacet models for refraction through rough surfaces. Rendering techniques, 2007: 0 18th, 2007

    2007

  54. [54]

    Performance relighting and reflectance transformation with time-multiplexed illumination

    Andreas Wenger, Andrew Gardner, Chris Tchou, Jonas Unger, Tim Hawkins, and Paul Debevec. Performance relighting and reflectance transformation with time-multiplexed illumination. ACM Transactions on Graphics (TOG), 24 0 (3): 0 756--764, 2005

    2005

  55. [55]

    Ps-nerf: Neural inverse rendering for multi-view photometric stereo

    Wenqi Yang, Guanying Chen, Chaofeng Chen, Zhenfang Chen, and Kwan-Yee K Wong. Ps-nerf: Neural inverse rendering for multi-view photometric stereo. In European Conference on Computer Vision, pages 266--284. Springer, 2022

    2022

  56. [56]

    InEuropean Conference on Computer Vision (ECCV)

    Zheng Zeng, Valentin Deschaintre, Iliyan Georgiev, Yannick Hold-Geoffroy, Yiwei Hu, Fujun Luan, Ling-Qi Yan, and Milo s Ha s an. Rgb↔x: Image decomposition and synthesis using material- and lighting-aware diffusion models. In ACM SIGGRAPH 2024 Conference Papers, SIGGRAPH '24, New York, NY, USA, 2024. Association for Computing Machinery. ISBN 9798400705250...

    work page doi:10.1145/3641519.3657445 2024

  57. [57]

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

    Kai Zhang, Fujun Luan, Qianqian Wang, Kavita Bala, and Noah 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, pages 5453--5462, 2021 a

    2021

  58. [58]

    Iron: Inverse rendering by optimizing neural sdfs and materials from photometric images

    Kai Zhang, Fujun Luan, Zhengqi Li, and Noah Snavely. Iron: Inverse rendering by optimizing neural sdfs and materials from photometric images. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5565--5574, 2022 a

    2022

  59. [59]

    Deep svbrdf estimation from single image under learned planar lighting

    Lianghao Zhang, Fangzhou Gao, Li Wang, Minjing Yu, Jiamin Cheng, and Jiawan Zhang. Deep svbrdf estimation from single image under learned planar lighting. In ACM SIGGRAPH 2023 conference proceedings, pages 1--11, 2023

    2023

  60. [60]

    Sparse svbrdf acquisition via importance-aware illumination multiplexing

    Lianghao Zhang, Zixuan Wang, Li Wang, Fangzhou Gao, Ruya Sun, and Jiawan Zhang. Sparse svbrdf acquisition via importance-aware illumination multiplexing. ACM Transactions on Graphics (TOG), 44 0 (6): 0 1--13, 2025

    2025

  61. [61]

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

    Xiuming Zhang, Pratul P Srinivasan, Boyang Deng, Paul Debevec, William T Freeman, and Jonathan T Barron. Nerfactor: Neural factorization of shape and reflectance under an unknown illumination. ACM Transactions on Graphics (ToG), 40 0 (6): 0 1--18, 2021 b

    2021

  62. [62]

    Modeling indirect illumination for inverse rendering

    Yuanqing Zhang, Jiaming Sun, Xingyi He, Huan Fu, Rongfei Jia, and Xiaowei Zhou. Modeling indirect illumination for inverse rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18643--18652, 2022 b

    2022