Dancing Avatar: Pose and Text-Guided Human Motion Videos Synthesis with Image Diffusion Model

TL;DR

Dancing Avatar introduces a novel method for generating high-quality, pose and text-guided human motion videos using a pretrained T2I diffusion model, with modules ensuring character consistency, background continuity, and temporal coherence.

Contribution

The paper presents a new framework that leverages a pretrained T2I diffusion model with alignment modules for consistent, high-quality human motion video synthesis guided by text and poses.

Findings

Superior video quality compared to state-of-the-art methods

Enhanced temporal coherence and background fidelity

Effective preservation of human character and clothing across poses

Abstract

The rising demand for creating lifelike avatars in the digital realm has led to an increased need for generating high-quality human videos guided by textual descriptions and poses. We propose Dancing Avatar, designed to fabricate human motion videos driven by poses and textual cues. Our approach employs a pretrained T2I diffusion model to generate each video frame in an autoregressive fashion. The crux of innovation lies in our adept utilization of the T2I diffusion model for producing video frames successively while preserving contextual relevance. We surmount the hurdles posed by maintaining human character and clothing consistency across varying poses, along with upholding the background's continuity amidst diverse human movements. To ensure consistent human appearances across the entire video, we devise an intra-frame alignment module. This module assimilates text-guided synthesized human character knowledge into the pretrained T2I diffusion model, synergizing insights from ChatGPT. For preserving background continuity, we put forth a background alignment pipeline, amalgamating insights from segment anything and image inpainting techniques. Furthermore, we propose an inter-frame alignment module that draws inspiration from an auto-regressive pipeline to augment temporal consistency between adjacent frames, where the preceding frame guides the synthesis process of the current frame. Comparisons with state-of-the-art methods demonstrate that Dancing Avatar exhibits the capacity to generate human videos with markedly superior quality, both in terms of human and background fidelity, as well as temporal coherence compared to existing state-of-the-art approaches.