| Autor: |
Fathallah M; Department of Computer Science, Faculty of Computers and Information, Kafr El-Sheikh University, Kafr El-Sheikh 33511, Egypt., Eletriby S; Department of Computer Science, Faculty of Computers and Information, Menoufia University, Menoufia 32511, Egypt., Alsabaan M; Department of Computer Engineering, College of Computer and Information Sciences, King Saud University, P.O. Box 51178, Riyadh 11543, Saudi Arabia., Ibrahem MI; School of Computer and Cyber Sciences, Augusta University, Augusta, GA 30912, USA., Farok G; Department of Computer Science, Faculty of Computers and Information, Menoufia University, Menoufia 32511, Egypt. |
| Abstrakt: |
This paper presents a novel framework for 3D face reconstruction from single 2D images and addresses critical limitations in existing methods. Our approach integrates modified adversarial neural networks with graph neural networks to achieve state-of-the-art performance. Key innovations include (1) a generator architecture based on Graph Convolutional Networks (GCNs) with a novel loss function and identity blocks, mitigating mode collapse and instability; (2) the integration of facial landmarks and a non-parametric efficient-net decoder for enhanced feature capture; and (3) a lightweight GCN-based discriminator for improved accuracy and stability. Evaluated on the 300W-LP and AFLW2000-3D datasets, our method outperforms existing approaches, reducing Chamfer Distance by 62.7% and Earth Mover's Distance by 57.1% on 300W-LP. Moreover, our framework demonstrates superior robustness to variations in head positioning, occlusion, noise, and lighting conditions while achieving significantly faster processing times. |
