Efficient Region-Aware Neural Radiance Fields for High-Fidelity Talking Portrait Synthesis

TL;DR

This paper introduces ER-NeRF, a region-aware neural radiance field architecture that enables fast, high-fidelity talking portrait synthesis with real-time rendering and efficient model size, leveraging spatial region contributions and explicit audio-region connections.

Contribution

The paper proposes a novel ER-NeRF architecture with a Tri-Plane Hash Representation, a Region Attention Module, and Adaptive Pose Encoding for improved talking portrait synthesis.

Findings

Achieves state-of-the-art high-fidelity talking portrait videos.

Enables real-time rendering with small model size.

Demonstrates superior performance in accuracy and efficiency.

Abstract

This paper presents ER-NeRF, a novel conditional Neural Radiance Fields (NeRF) based architecture for talking portrait synthesis that can concurrently achieve fast convergence, real-time rendering, and state-of-the-art performance with small model size. Our idea is to explicitly exploit the unequal contribution of spatial regions to guide talking portrait modeling. Specifically, to improve the accuracy of dynamic head reconstruction, a compact and expressive NeRF-based Tri-Plane Hash Representation is introduced by pruning empty spatial regions with three planar hash encoders. For speech audio, we propose a Region Attention Module to generate region-aware condition feature via an attention mechanism. Different from existing methods that utilize an MLP-based encoder to learn the cross-modal relation implicitly, the attention mechanism builds an explicit connection between audio features and spatial regions to capture the priors of local motions. Moreover, a direct and fast Adaptive Pose Encoding is introduced to optimize the head-torso separation problem by mapping the complex transformation of the head pose into spatial coordinates. Extensive experiments demonstrate that our method renders better high-fidelity and audio-lips synchronized talking portrait videos, with realistic details and high efficiency compared to previous methods.