pith. sign in

arxiv: 2606.08041 · v1 · pith:3FA6JGPOnew · submitted 2026-06-06 · 💻 cs.GR · cs.CV

Wispy to Voluminous: Prior-free Multi-view Capture of Strand-level Facial Hair

Jaeseong Lee , Giljoo Nam , Adrian Jarabo , Carlos Aliaga This is my paper

Pith reviewed 2026-06-27 19:04 UTC · model grok-4.3

classification 💻 cs.GR cs.CV
keywords facial hair reconstruction3D Gaussian splattingstrand-level modelingmulti-view capturedigital human avatarshair renderingcomputer graphicscurve extraction
0
0 comments X

The pith

A four-stage pipeline converts 3D Gaussian face models into explicit, editable facial hair strands from multi-view images alone.

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

The paper aims to show that facial hair can be recovered as individual strands rather than baked volumetric density by starting from optimized 3D Gaussians and applying successive geometric and photometric steps. This matters for digital avatars because current methods either merge hair into skin geometry or target only dense scalp hair, leaving sparse beards, mustaches, lashes and brows uneditable. The approach uses tracked head geometry to limit ray marching, a crossing-robust tracer to build curves, a surface-grounding prior to fix root-tip direction, and final opacity-driven refinement. If successful the output strands carry the correct local orientation and sparsity of real facial hair and can be fed directly into animation, simulation and grooming pipelines.

Core claim

The central claim is that an unstructured 3D Gaussian representation of a face can be turned into high-fidelity curve-based facial hair strands through four sequential operations: Gaussian optimization with early ray termination, crossing-robust strand tracing, physically motivated surface grounding that resolves root-tip ambiguity, and opacity-driven photometric refinement, and that the resulting strands preserve the orientation and sparsity patterns of real facial hair while remaining immediately usable for production tasks.

What carries the argument

The four-stage conversion pipeline that turns constrained 3D Gaussians into continuous curve strands via early-termination optimization, robust tracing, physical surface grounding, and photometric density control.

If this is right

  • Recovered strands preserve the local orientation and sparsity patterns of real facial hair.
  • Output assets can be used directly for facial animation and physical simulation.
  • The strands support geometric grooming, transfer between faces, appearance editing, and physics-based rendering.
  • The same pipeline works for beards, mustaches, lashes and brows from ordinary multi-view image sets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The method could be chained with existing Gaussian avatar pipelines to produce complete editable heads rather than separate face and hair assets.
  • Because the output is explicit curves, it opens the possibility of real-time strand-level simulation on mobile devices once the initial capture is done.
  • The same tracing and grounding logic might apply to other sparse linear structures such as eyelashes on non-human characters or thin vegetation in outdoor scenes.

Load-bearing premise

The assumption that the combination of early ray termination, crossing-robust tracing, physical grounding prior, and opacity refinement is enough to resolve root-tip and geometric ambiguities without any extra manual intervention or stronger priors.

What would settle it

A controlled capture of known facial hair geometry where the output strands show incorrect root-to-tip directions or merge across visible crossings that the method claims to handle.

read the original abstract

Facial hair is a defining trait of personal identity, yet remains a critical bottleneck for digital avatars. Recent volumetric methods achieve photorealism but bake hair into the underlying face geometry, preventing editability and failing to resolve sparse, strand-like structures. Meanwhile, scalp-hair reconstruction methods target dense hair volumes and do not transfer to the sparse, spatially-varying nature of facial hair. We present a pipeline that automatically reconstructs facial hair -- beard, mustache, lashes, and brows -- from multi-view images, converting an unstructured 3D Gaussian representation into an explicit curve-based strand representation. We resolve geometric ambiguities in four stages: (i) optimizing 3D Gaussians constrained by tracked head geometry to enforce early ray termination and suppress sub-surface noise; (ii) tracing continuous strands robust to frequent crossings and extreme curvature; (iii) grounding strands to the surface and resolving root-tip ambiguity via a physically-motivated prior; and (iv) refining the reconstruction through opacity-driven density control under photometric optimization. To our knowledge, this is the first method to reconstruct high-fidelity facial hair strands from a 3D Gaussian representation. The recovered strands faithfully preserve the orientation and sparsity patterns characteristic of facial hair, and yield assets immediately suitable for downstream production tasks, including facial animation and physical simulation, geometric grooming and transfer, appearance editing, and physics-based rendering.

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 / 0 minor

Summary. The manuscript presents a four-stage pipeline that converts multi-view images into explicit strand-level facial hair (beard, mustache, lashes, brows) via 3D Gaussian optimization. Stage (i) optimizes Gaussians with tracked head geometry for early ray termination; (ii) performs crossing-robust strand tracing; (iii) grounds strands to the surface and resolves root-tip ambiguity with a physically-motivated prior; (iv) refines via opacity-driven photometric optimization. The central claim is that this yields the first high-fidelity strand reconstruction from a 3D Gaussian representation, preserving orientation and sparsity patterns and producing production-ready assets.

Significance. If the pipeline delivers faithful strand recovery without manual intervention, the work would advance digital avatar pipelines by bridging volumetric capture and editable curve-based hair assets, enabling downstream tasks such as animation, physical simulation, grooming transfer, and physics-based rendering that current baked volumetric methods cannot support.

major comments (2)
  1. [Abstract / Title] Abstract (stage (iii) description) and title: The title and abstract position the method as 'prior-free,' yet stage (iii) explicitly invokes a 'physically-motivated prior' to ground strands and resolve root-tip ambiguity. If this prior encodes non-trivial constraints (e.g., surface normals, growth direction, or curvature bounds), it contradicts the prior-free claim and the assertion that the sequence of Gaussian optimization, tracing, and refinement suffices without additional strong priors. This is load-bearing for the central positioning and must be clarified with a precise definition of 'prior-free' and the exact form of the prior.
  2. [Abstract] Abstract: No quantitative validation, error analysis, ablation studies, or baseline comparisons are reported. The claim of faithful preservation of orientation and sparsity therefore rests on unshown implementation details and qualitative results, weakening assessment of whether the four-stage sequence actually resolves the stated geometric ambiguities in sparse facial hair.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which identify key issues in the positioning and validation of our method. We respond to each major comment below.

read point-by-point responses

  1. Referee: [Abstract / Title] Abstract (stage (iii) description) and title: The title and abstract position the method as 'prior-free,' yet stage (iii) explicitly invokes a 'physically-motivated prior' to ground strands and resolve root-tip ambiguity. If this prior encodes non-trivial constraints (e.g., surface normals, growth direction, or curvature bounds), it contradicts the prior-free claim and the assertion that the sequence of Gaussian optimization, tracing, and refinement suffices without additional strong priors. This is load-bearing for the central positioning and must be clarified with a precise definition of 'prior-free' and the exact form of the prior.

    Authors: We agree this is an inconsistency. The term 'prior-free' was meant to indicate the absence of learned or template-based priors common in other hair methods, but stage (iii) does apply a physically-motivated prior with geometric constraints. We will revise the title to remove 'prior-free', clarify the claim in the abstract and introduction, and provide the exact mathematical definition of the prior used. revision: yes

  2. Referee: [Abstract] Abstract: No quantitative validation, error analysis, ablation studies, or baseline comparisons are reported. The claim of faithful preservation of orientation and sparsity therefore rests on unshown implementation details and qualitative results, weakening assessment of whether the four-stage sequence actually resolves the stated geometric ambiguities in sparse facial hair.

    Authors: The manuscript presents the first method for this task and relies primarily on qualitative results across diverse subjects. We acknowledge the lack of quantitative metrics, ablations, and comparisons. In revision we will add error analysis (including against synthetic ground truth), stage-wise ablations, and any feasible baseline comparisons, and will incorporate key quantitative highlights into the abstract. revision: yes

Circularity Check

0 steps flagged

No significant circularity; procedural pipeline with no derivation chain reducing to inputs

full rationale

The paper presents a four-stage procedural pipeline for converting 3D Gaussians to strand curves, with stages involving optimization, tracing, grounding via a physically-motivated prior, and photometric refinement. No equations, fitted parameters renamed as predictions, or self-citation load-bearing steps are described in the provided text. The 'prior-free' title claim conflicts with stage (iii) but does not create a circular derivation; the method is self-contained as an empirical pipeline without reducing any claimed result to its own inputs by construction. This matches the default expectation of no circularity for non-derivational method papers.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; the 'physically-motivated prior' for root-tip resolution is referenced but not formalized or evidenced as independent.

pith-pipeline@v0.9.1-grok · 5783 in / 1135 out tokens · 20027 ms · 2026-06-27T19:04:11.420371+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

80 extracted references · 2 canonical work pages

  1. [1]

    ACM Transactions on Graphics (Proceedings SIGGRAPH 2015) , Volume =

    Liwen Hu and Chongyang Ma and Linjie Luo and Hao Li , Title =. ACM Transactions on Graphics (Proceedings SIGGRAPH 2015) , Volume =. 2015 , Month =

    2015

  2. [2]

    European Conference on Computer Vision , pages=

    Human hair reconstruction with strand-aligned 3d gaussians , author=. European Conference on Computer Vision , pages=. 2024 , organization=

    2024

  3. [3]

    arXiv preprint arXiv:2402.10483 , year=

    Gaussianhair: Hair modeling and rendering with light-aware gaussians , author=. arXiv preprint arXiv:2402.10483 , year=

    arXiv

  4. [4]

    ACM Transactions on Graphics (TOG) , volume=

    Groomcap: High-fidelity prior-free hair capture , author=. ACM Transactions on Graphics (TOG) , volume=. 2024 , publisher=

    2024

  5. [5]

    ACM Transactions on Graphics (TOG) , volume=

    Groomgen: A high-quality generative hair model using hierarchical latent representations , author=. ACM Transactions on Graphics (TOG) , volume=. 2023 , publisher=

    2023

  6. [6]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Strand-accurate multi-view hair capture , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  7. [7]

    arXiv preprint arXiv:2509.07774 , year=

    HairGS: Hair Strand Reconstruction based on 3D Gaussian Splatting , author=. arXiv preprint arXiv:2509.07774 , year=

    arXiv

  8. [8]

    ACM Transactions on Graphics (TOG) , volume=

    Ct2hair: High-fidelity 3d hair modeling using computed tomography , author=. ACM Transactions on Graphics (TOG) , volume=. 2023 , publisher=

    2023

  9. [9]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    GroomLight: Hybrid Inverse Rendering for Relightable Human Hair Appearance Modeling , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

  10. [10]

    2021 , publisher=

    Human hair inverse rendering using multi-view photometric data , author=. 2021 , publisher=

    2021

  11. [11]

    Computer Graphics Forum , volume=

    Controllable Biophysical Human Faces , author=. Computer Graphics Forum , volume=. 2025 , organization=

    2025

  12. [12]

    ACM Transactions on Graphics (TOG) , volume=

    Structure-aware hair capture , author=. ACM Transactions on Graphics (TOG) , volume=. 2013 , publisher=

    2013

  13. [13]

    ACM Transactions on Graphics (ToG) , volume=

    Coupled 3D reconstruction of sparse facial hair and skin , author=. ACM Transactions on Graphics (ToG) , volume=. 2012 , publisher=

    2012

  14. [14]

    Proceedings of Graphics Interface (GI) , pages=

    Image-based hair capture by inverse lighting , author=. Proceedings of Graphics Interface (GI) , pages=

  15. [15]

    ACM transactions on graphics (TOG) , volume=

    Capture of hair geometry from multiple images , author=. ACM transactions on graphics (TOG) , volume=. 2004 , publisher=

    2004

  16. [16]

    , author=

    Hair photobooth: geometric and photometric acquisition of real hairstyles. , author=. ACM Trans. Graph. , volume=

  17. [17]

    Computer Graphics Forum , volume=

    3D Reconstruction and Semantic Modeling of Eyelashes , author=. Computer Graphics Forum , volume=. 2024 , organization=

    2024

  18. [18]

    2007 15th European signal processing conference , pages=

    The effective rank: A measure of effective dimensionality , author=. 2007 15th European signal processing conference , pages=. 2007 , organization=

    2007

  19. [19]

    Advances in Neural Information Processing Systems , volume=

    Effective rank analysis and regularization for enhanced 3d gaussian splatting , author=. Advances in Neural Information Processing Systems , volume=

  20. [20]

    IEEE Transactions on Visualization and Computer Graphics , volume=

    Optical Models for Direct Volume Rendering , author=. IEEE Transactions on Visualization and Computer Graphics , volume=. 1995 , publisher=

    1995

  21. [21]

    ACM Transactions on Graphics (TOG) , volume=

    Facial hair tracking for high fidelity performance capture , author=. ACM Transactions on Graphics (TOG) , volume=. 2022 , publisher=

    2022

  22. [22]

    The Visual Computer , volume=

    Strand-accurate multi-view facial hair reconstruction and tracking , author=. The Visual Computer , volume=. 2024 , publisher=

    2024

  23. [23]

    ACM SIGGRAPH 2012 Talks , pages=

    High fidelity facial hair capture , author=. ACM SIGGRAPH 2012 Talks , pages=

    2012

  24. [24]

    ACM Transactions on Graphics (TOG) , volume=

    EMS: 3D Eyebrow Modeling from Single-view Images , author=. ACM Transactions on Graphics (TOG) , volume=. 2023 , publisher=

    2023

  25. [25]

    International journal of cosmetic science , volume=

    The male beard hair and facial skin--challenges for shaving , author=. International journal of cosmetic science , volume=. 2016 , publisher=

    2016

  26. [26]

    Revista Brasileira de Cirurgia Pl

    Beard transplantation , author=. Revista Brasileira de Cirurgia Pl. 2017 , publisher=

    2017

  27. [27]

    Communications of the ACM , volume=

    Nerf: Representing scenes as neural radiance fields for view synthesis , author=. Communications of the ACM , volume=. 2021 , publisher=

    2021

  28. [28]

    , author=

    3d gaussian splatting for real-time radiance field rendering. , author=. ACM Trans. Graph. , volume=

  29. [29]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Monohair: High-fidelity hair modeling from a monocular video , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  30. [30]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Neuralhdhair: Automatic high-fidelity hair modeling from a single image using implicit neural representations , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  31. [31]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Difflocks: Generating 3d hair from a single image using diffusion models , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

  32. [32]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Hairstep: Transfer synthetic to real using strand and depth maps for single-view 3d hair modeling , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  33. [33]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Neural haircut: Prior-guided strand-based hair reconstruction , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  34. [34]

    European Conference on Computer Vision , pages=

    Neural strands: Learning hair geometry and appearance from multi-view images , author=. European Conference on Computer Vision , pages=. 2022 , organization=

    2022

  35. [35]

    arXiv preprint arXiv:1906.07751 , year=

    Neural volumes: Learning dynamic renderable volumes from images , author=. arXiv preprint arXiv:1906.07751 , year=

    arXiv 1906

  36. [36]

    ACM Transactions on Graphics (TOG) , volume=

    Eyelashnet: A dataset and a baseline method for eyelash matting , author=. ACM Transactions on Graphics (TOG) , volume=. 2021 , publisher=

    2021

  37. [37]

    Computer Graphics Forum , volume=

    Effective eyebrow matting with domain adaptation , author=. Computer Graphics Forum , volume=. 2022 , organization=

    2022

  38. [38]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Analyzing and improving the image quality of stylegan , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  39. [39]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Gaussianavatars: Photorealistic head avatars with rigged 3d gaussians , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  40. [40]

    1999 , isbn =

    Blanz, Volker and Vetter, Thomas , title =. 1999 , isbn =. doi:10.1145/311535.311556 , booktitle =

    work page doi:10.1145/311535.311556 1999

  41. [41]

    ACM Transactions on Graphics (ToG) , volume=

    Mixture of volumetric primitives for efficient neural rendering , author=. ACM Transactions on Graphics (ToG) , volume=. 2021 , publisher=

    2021

  42. [42]

    13th International Conference on Learning Representations, ICLR 2025 , pages=

    SURFHEAD: AFFINE RIG BLENDING FOR GEOMETRICALLY ACCURATE 2D GAUSSIAN SURFEL HEAD AVATARS , author=. 13th International Conference on Learning Representations, ICLR 2025 , pages=. 2025 , organization=

    2025

  43. [43]

    SIGGRAPH Asia 2024 Conference Papers , pages=

    Npga: Neural parametric gaussian avatars , author=. SIGGRAPH Asia 2024 Conference Papers , pages=

    2024

  44. [44]

    Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers , pages=

    TeGA: Texture Space Gaussian Avatars for High-Resolution Dynamic Head Modeling , author=. Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers , pages=

  45. [45]

    Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers , pages=

    Scaffoldavatar: High-fidelity gaussian avatars with patch expressions , author=. Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers , pages=

  46. [46]

    Thirteenth International Conference on 3D Vision , year=

    TexAvatars: Hybrid Texel-3D Representations for Stable Rigging of Photorealistic Gaussian Head Avatars , author=. Thirteenth International Conference on 3D Vision , year=

  47. [47]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Im avatar: Implicit morphable head avatars from videos , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  48. [48]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Pointavatar: Deformable point-based head avatars from videos , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  49. [49]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Learning neural parametric head models , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  50. [50]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Dynamic neural radiance fields for monocular 4d facial avatar reconstruction , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  51. [51]

    arXiv preprint arXiv:2310.17519 , year=

    Flare: Fast learning of animatable and relightable mesh avatars , author=. arXiv preprint arXiv:2310.17519 , year=

    arXiv

  52. [52]

    ACM Transactions on Graphics (ToG) , volume=

    Learning an animatable detailed 3D face model from in-the-wild images , author=. ACM Transactions on Graphics (ToG) , volume=. 2021 , publisher=

    2021

  53. [53]

    SIGGRAPH Asia 2024 Conference Papers , pages=

    Spark: Self-supervised personalized real-time monocular face capture , author=. SIGGRAPH Asia 2024 Conference Papers , pages=

    2024

  54. [54]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Fitme: Deep photorealistic 3d morphable model avatars , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  55. [55]

    2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) , pages=

    Fitdiff: Robust monocular 3d facial shape and reflectance estimation using diffusion models , author=. 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) , pages=. 2025 , organization=

    2025

  56. [56]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Relightify: Relightable 3d faces from a single image via diffusion models , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  57. [57]

    in-the-wild

    Avatarme: Realistically renderable 3d facial reconstruction" in-the-wild" , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  58. [58]

    arXiv preprint arXiv:2404.04104 , year=

    SMIRK: 3D Facial Expressions through Analysis-by-Neural-Synthesis , author=. arXiv preprint arXiv:2404.04104 , year=

    arXiv

  59. [59]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Monocular Facial Appearance Capture in the Wild , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  60. [60]

    SIGGRAPH Asia 2024 Conference Papers , pages=

    Cafca: High-quality novel view synthesis of expressive faces from casual few-shot captures , author=. SIGGRAPH Asia 2024 Conference Papers , pages=

    2024

  61. [61]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Preface: A data-driven volumetric prior for few-shot ultra high-resolution face synthesis , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  62. [62]

    Li, Tianye and Bolkart, Timo and Black, Michael. J. and Li, Hao and Romero, Javier , journal =. Learning a model of facial shape and expression from. 2017 , pages =

    2017

  63. [63]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Neural head avatars from monocular rgb videos , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  64. [64]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Pixel codec avatars , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  65. [65]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Relightable gaussian codec avatars , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  66. [66]

    arXiv preprint arXiv:2512.15711 , year=

    Gaussian Pixel Codec Avatars: A Hybrid Representation for Efficient Rendering , author=. arXiv preprint arXiv:2512.15711 , year=

    arXiv

  67. [67]

    2012 IEEE Conference on Computer Vision and Pattern Recognition , pages=

    Multi-view hair capture using orientation fields , author=. 2012 IEEE Conference on Computer Vision and Pattern Recognition , pages=. 2012 , organization=

    2012

  68. [68]

    Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

    Wide-baseline hair capture using strand-based refinement , author=. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

  69. [69]

    ACM Transactions on Graphics (TOG) , volume=

    Lighting hair from the inside: A thermal approach to hair reconstruction , author=. ACM Transactions on Graphics (TOG) , volume=. 2012 , publisher=

    2012

  70. [70]

    hair: Reconstructing scalp-connected hair strands without pre-training via differentiable rendering of line segments , author=

    Dr. hair: Reconstructing scalp-connected hair strands without pre-training via differentiable rendering of line segments , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  71. [71]

    and Grabli, Stephane and Chiang, Matt Jen-Yuan and Hery, Christophe and Roble, Doug and Ramamoorthi, Ravi and Li, Tzu-Mao and Maury, Olivier , title =

    Chang, Wesley and Russell, Andrew L. and Grabli, Stephane and Chiang, Matt Jen-Yuan and Hery, Christophe and Roble, Doug and Ramamoorthi, Ravi and Li, Tzu-Mao and Maury, Olivier , title =. 2025 , issue_date =. doi:10.1145/3731168 , journal =

    work page doi:10.1145/3731168 2025

  72. [72]

    Proceedings of the IEEE/CVF international conference on computer vision , pages=

    Soft rasterizer: A differentiable renderer for image-based 3d reasoning , author=. Proceedings of the IEEE/CVF international conference on computer vision , pages=

  73. [73]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Im2Haircut: Single-view Strand-based Hair Reconstruction for Human Avatars , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  74. [74]

    , author=

    Patchmatch stereo-stereo matching with slanted support windows. , author=. Bmvc , volume=

  75. [75]

    Image and Vision Computing , pages=

    Matte anything: Interactive natural image matting with segment anything model , author=. Image and Vision Computing , pages=. 2024 , publisher=

    2024

  76. [76]

    Advances in neural information processing systems , volume=

    Volume rendering of neural implicit surfaces , author=. Advances in neural information processing systems , volume=

  77. [77]

    arXiv preprint arXiv:2106.10689 , year=

    Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction , author=. arXiv preprint arXiv:2106.10689 , year=

    Pith/arXiv arXiv

  78. [78]

    ACM SIGGRAPH 2015 Talks , pages=

    A practical and controllable hair and fur model for production path tracing , author=. ACM SIGGRAPH 2015 Talks , pages=

    2015

  79. [79]

    Mitsuba 3 renderer , author =

  80. [80]

    ACM Transactions on Graphics , year =

    Modular Primitives for High-Performance Differentiable Rendering , author =. ACM Transactions on Graphics , year =