Fathallah Mohamed, Eletriby Sherif, Alsabaan Maazen, Ibrahem Mohamed I, Farok Gamal
Department of Computer Science, Faculty of Computers and Information, Kafr El-Sheikh University, Kafr El-Sheikh 33511, Egypt.
Department of Computer Science, Faculty of Computers and Information, Menoufia University, Menoufia 32511, Egypt.
Sensors (Basel). 2024 Sep 28;24(19):6280. doi: 10.3390/s24196280.
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.
本文提出了一种用于从单张二维图像进行三维人脸重建的新颖框架，并解决了现有方法中的关键局限性。我们的方法将改进的对抗神经网络与图神经网络相结合，以实现领先的性能。关键创新包括：（1）基于图卷积网络（GCN）的生成器架构，具有新颖的损失函数和恒等块，减轻模式崩溃和不稳定性；（2）整合面部标志和非参数高效网络解码器以增强特征捕获；（3）基于轻量级GCN的判别器，以提高准确性和稳定性。在300W-LP和AFLW2000-3D数据集上进行评估时，我们的方法优于现有方法，在300W-LP上，倒角距离减少了62.7%，推土机距离减少了57.1%。此外，我们的框架在头部定位、遮挡、噪声和光照条件变化时表现出卓越的鲁棒性，同时处理时间显著更快。
