Skip to main navigation Skip to search Skip to main content

Intrinsic Morphological Relationship Guided 3D Craniofacial Reconstruction Using Siamese Cycle Attention GAN

  • Junli Zhao
  • , Chengyuan Wang
  • , Yu Hui Wen
  • , Fuqing Duan
  • , Ran Yi
  • , Yong Jin Liu
  • , Qingdong Long
  • , Zhenkuan Pan
  • , Xianfeng Gu
  • Qingdao University
  • Beijing Jiaotong University
  • Beijing Normal University
  • Shanghai Jiao Tong University
  • Tsinghua University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Craniofacial reconstruction is essential in forensic science and has widespread applications. It is challenging due to the detailed facial geometry, complex skull topology, and nonlinear skull-face relationship. We propose a novel approach for 3D craniofacial reconstruction using a Siamese cycle attention mechanism within Generative Adversarial Networks (GAN). Benefiting from the cycle attention mechanism, our method focuses on high-frequency features and morphological connections between the skull and face. Additionally, a Siamese network preserves its identity consistently. Extensive experiments demonstrate superior accuracy and high-quality details of our approach.

Original languageEnglish
Title of host publicationProceedings - SIGGRAPH Asia 2024 Technical Communications, SA 2024
EditorsStephen N. Spencer
PublisherAssociation for Computing Machinery, Inc
ISBN (Electronic)9798400711404
DOIs
StatePublished - Dec 3 2024
Event2024 SIGGRAPH Asia 2024 Technical Communications, SA 2024 - Tokyo, Japan
Duration: Dec 3 2024Dec 6 2024

Publication series

NameProceedings - SIGGRAPH Asia 2024 Technical Communications, SA 2024

Conference

Conference2024 SIGGRAPH Asia 2024 Technical Communications, SA 2024
Country/TerritoryJapan
CityTokyo
Period12/3/2412/6/24

Keywords

  • Craniofacial reconstruction
  • Cycle Attention GAN
  • Intrinsic morphological relationship
  • Siamese network

Fingerprint

Dive into the research topics of 'Intrinsic Morphological Relationship Guided 3D Craniofacial Reconstruction Using Siamese Cycle Attention GAN'. Together they form a unique fingerprint.

Cite this