GLVD: Guided Learned Vertex Descent
This addresses the problem of efficient and expressive 3D face reconstruction for computer vision applications, representing an incremental improvement over existing methods.
The paper tackles 3D face reconstruction from few-shot images by introducing GLVD, a hybrid method that integrates per-vertex neural field optimization with global structural guidance from predicted 3D keypoints. It achieves state-of-the-art performance in single-view settings and remains highly competitive in multi-view scenarios while substantially reducing inference time.
Existing 3D face modeling methods usually depend on 3D Morphable Models, which inherently constrain the representation capacity to fixed shape priors. Optimization-based approaches offer high-quality reconstructions but tend to be computationally expensive. In this work, we introduce GLVD, a hybrid method for 3D face reconstruction from few-shot images that extends Learned Vertex Descent (LVD) by integrating per-vertex neural field optimization with global structural guidance from dynamically predicted 3D keypoints. By incorporating relative spatial encoding, GLVD iteratively refines mesh vertices without requiring dense 3D supervision. This enables expressive and adaptable geometry reconstruction while maintaining computational efficiency. GLVD achieves state-of-the-art performance in single-view settings and remains highly competitive in multi-view scenarios, all while substantially reducing inference time.