HairManip: High quality hair manipulation via hair element disentangling

Zhao, Huihuang, Zhang, Lin, Rosin, Paul L. ORCID: https://orcid.org/0000-0002-4965-3884, Lai, Yu-Kun ORCID: https://orcid.org/0000-0002-2094-5680 and Wang, Yaonan 2024. HairManip: High quality hair manipulation via hair element disentangling. Pattern Recognition 147 , 110132. 10.1016/j.patcog.2023.110132

Preview

PDF - Accepted Post-Print Version Download (4MB) | Preview

Abstract

Hair editing is challenging due to the complexity and variety of hair materials and shapes. Existing methods employ reference images or user-painted masks to edit hair and have achieved promising results. However, discrepancies in color and shape between the source and target hair can occasionally result in unrealistic results. Therefore, we propose a new hair editing method named HairManip, which decouples the hair information from the input source image into shape and color components. We then train hairstyle and hair color editing sub-networks to handle this complex information independently. To further enhance editing efficiency and accuracy, we introduce a latent code preprocessing module that effectively extracts meaningful features from hair regions, thereby improving the model’s editing capabilities. The experimental results demonstrate that our method achieves significant results in editing accuracy and authenticity, thanks to the carefully designed network structure and loss functions. Code can be found at https://github.com/Zlin0530/HairManip.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Computer Science & Informatics
Publisher:	Elsevier
ISSN:	0031-3203
Date of First Compliant Deposit:	19 November 2023
Date of Acceptance:	17 November 2023
Last Modified:	17 Nov 2024 02:45
URI:	https://orca.cardiff.ac.uk/id/eprint/163984

Actions (repository staff only)

Edit Item