LokiTalk: Learning Fine-Grained and Generalizable Correspondences to Enhance NeRF-based Talking Head Synthesis

TL;DR

LokiTalk enhances NeRF-based talking head synthesis by establishing fine-grained, generalizable correspondences, reducing artifacts, and improving training efficiency through region-specific deformation fields and ID-aware knowledge transfer.

Contribution

The paper introduces LokiTalk, a novel framework with region-specific deformation fields and ID-aware knowledge transfer for better accuracy and efficiency in NeRF-based talking head synthesis.

Findings

Achieves higher fidelity in synthesized talking heads.

Reduces training time and computational costs.

Improves dynamic accuracy and artifact reduction.

Abstract

Despite significant progress in talking head synthesis since the introduction of Neural Radiance Fields (NeRF), visual artifacts and high training costs persist as major obstacles to large-scale commercial adoption. We propose that identifying and establishing fine-grained and generalizable correspondences between driving signals and generated results can simultaneously resolve both problems. Here we present LokiTalk, a novel framework designed to enhance NeRF-based talking heads with lifelike facial dynamics and improved training efficiency. To achieve fine-grained correspondences, we introduce Region-Specific Deformation Fields, which decompose the overall portrait motion into lip movements, eye blinking, head pose, and torso movements. By hierarchically modeling the driving signals and their associated regions through two cascaded deformation fields, we significantly improve dynamic accuracy and minimize synthetic artifacts. Furthermore, we propose ID-Aware Knowledge Transfer, a plug-and-play module that learns generalizable dynamic and static correspondences from multi-identity videos, while simultaneously extracting ID-specific dynamic and static features to refine the depiction of individual characters. Comprehensive evaluations demonstrate that LokiTalk delivers superior high-fidelity results and training efficiency compared to previous methods. The code will be released upon acceptance.