TEDRA: Text-based Editing of Dynamic and Photoreal Actors

TL;DR

TEDRA enables realistic, dynamic 3D avatar editing from text prompts, maintaining high fidelity and motion consistency, representing a significant advance in user-friendly avatar customization.

Contribution

We introduce TEDRA, the first method for text-based editing of dynamic 3D avatars that preserves realism, motion, and view control, using a personalized diffusion model and novel training strategies.

Findings

Outperforms prior methods in visual quality and functionality.

Enables fine-grained clothing and style edits via text prompts.

Maintains high fidelity and motion coherence in edited avatars.

Abstract

Over the past years, significant progress has been made in creating photorealistic and drivable 3D avatars solely from videos of real humans. However, a core remaining challenge is the fine-grained and user-friendly editing of clothing styles by means of textual descriptions. To this end, we present TEDRA, the first method allowing text-based edits of an avatar, which maintains the avatar's high fidelity, space-time coherency, as well as dynamics, and enables skeletal pose and view control. We begin by training a model to create a controllable and high-fidelity digital replica of the real actor. Next, we personalize a pretrained generative diffusion model by fine-tuning it on various frames of the real character captured from different camera angles, ensuring the digital representation faithfully captures the dynamics and movements of the real person. This two-stage process lays the foundation for our approach to dynamic human avatar editing. Utilizing this personalized diffusion model, we modify the dynamic avatar based on a provided text prompt using our Personalized Normal Aligned Score Distillation Sampling (PNA-SDS) within a model-based guidance framework. Additionally, we propose a time step annealing strategy to ensure high-quality edits. Our results demonstrate a clear improvement over prior work in functionality and visual quality.