Quality metric guided portrait line drawing generation from unpaired training data

Yi, Ran, Liu, Yong-Jin, Lai, Yu-Kun ORCID: https://orcid.org/0000-0002-2094-5680 and Rosin, Paul ORCID: https://orcid.org/0000-0002-4965-3884 2023. Quality metric guided portrait line drawing generation from unpaired training data. IEEE Transactions on Pattern Analysis and Machine Intelligence 45 (1) , pp. 905-918. 10.1109/TPAMI.2022.3147570

Preview

PDF - Accepted Post-Print Version Download (12MB) | Preview

Abstract

Face portrait line drawing is a unique style of art which is highly abstract and expressive. However, due to its high semantic constraints, many existing methods learn to generate portrait drawings using paired training data. In this paper, we propose a novel method to automatically transform face photos to portrait drawings using unpaired training data. Our method can (1) learn to generate high quality portrait drawings in multiple styles using a single network and (2) generate portrait drawings in a ‘`new style’' unseen in the training data. We observe that existing unpaired translation methods (such as CycleGAN) tend to embed invisible reconstruction information indiscriminately in the whole drawings due to significant information imbalance between the photo and portrait drawing domains, which leads to important facial features missing. To address this problem, we propose a novel asymmetric cycle mapping that enforces the reconstruction information to be visible and only embedded in selective facial regions. Along with localized discriminators for important facial regions, our method well preserves all important facial features. Generator dissection further explains that our model learns to incorporate face semantic information during drawing generation. Extensive experiments including a user study show that our model outperforms state-of-the-art methods.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Computer Science & Informatics
Publisher:	Institute of Electrical and Electronics Engineers
ISSN:	0162-8828
Date of First Compliant Deposit:	3 February 2022
Date of Acceptance:	26 January 2022
Last Modified:	17 Nov 2024 18:15
URI:	https://orca.cardiff.ac.uk/id/eprint/147154

Citation Data

Cited 7 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item