Detangled: A Framework for Creating, Editing, and Inferencing Feature Rich Hair Strands
Quick Answer
The proposed framework enables the creation and editing of feature-rich hair strands through a five-dimensional parameter space, allowing precise texture generation independent of strand direction.
Quick Take
The proposed framework enables the creation and editing of feature-rich hair strands through a five-dimensional parameter space, allowing precise texture generation independent of strand direction. By utilizing centerline geometry, the method disentangles strand texture from overall style, facilitating new strand creation via generative AI techniques. This approach demonstrates versatility across various hair types.
Key Points
- Introduces a five-dimensional parameter space for hair strand textures.
- Disentangles strand texture from overall direction using centerline geometry.
- Generative AI enables creation of new strands based on texture inputs.
- Facilitates groom editing through texture transfer or user inputs.
- Demonstrates effectiveness across a variety of hair types.
Paper Resources
📖 Reader Mode
~2 min readAbstract:We present a framework for understanding and generating feature rich hair strands. Drawing upon both scientific and cultural expertise, we define strand texture as the various distinctive patterns (curling, switchbacks, twist, etc.) that are formed by forces internal to a hair strand. We begin by proposing a novel five-dimensional parameter space, intended to be a bijection with naturally occurring hair strand textures. This encoding is both qualitatively accessible, allowing users to readily locate their own hair in the parameter space, and quantitatively precise, allowing the generation of individual strands from texture inputs. Importantly, strand texture should be independent from the overall strand direction. In order to disentangle strand texture from the overall strand direction, we identify centerline geometry and use it to map strands into a canonical space (a strand texture space). We construct centerlines using a novel method that cleanly distills complex hair grooms, separating the strand texture from the overall style (parameterized by style guides). We enable the creation of new strands conforming to our parametric description of texture via a generative artificial intelligence approach supervised by a separate neural network trained to label candidate strands according to our five-parameter description. The ability to create new strands conforming to any desired texture enables groom editing using either texture transfer or user-provided inputs. We demonstrate results on a variety of hair types.
| Comments: | 18 pages, 18 figures |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| ACM classes: | I.3.7; I.3.5; I.3.8; I.6.3; J.3 |
| Cite as: | arXiv:2607.09811 [cs.CV] |
| (or arXiv:2607.09811v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.09811 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sarah Jobalia [view email]
[v1]
Fri, 10 Jul 2026 02:41:14 UTC (41,491 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.