English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The rising demand for creating lifelike avatars in the digital realm has ledto an increased need for generating high-quality human videos guided by textualdescriptions and poses. We propose Dancing Avatar, designed to fabricate humanmotion videos driven by poses and textual cues. Our approach employs apretrained T2I diffusion model to generate each video frame in anautoregressive fashion. The crux of innovation lies in our adept utilization ofthe T2I diffusion model for producing video frames successively whilepreserving contextual relevance. We surmount the hurdles posed by maintaininghuman character and clothing consistency across varying poses, along withupholding the background's continuity amidst diverse human movements. To ensureconsistent human appearances across the entire video, we devise an intra-framealignment module. This module assimilates text-guided synthesized humancharacter knowledge into the pretrained T2I diffusion model, synergizinginsights from ChatGPT. For preserving background continuity, we put forth abackground alignment pipeline, amalgamating insights from segment anything andimage inpainting techniques. Furthermore, we propose an inter-frame alignmentmodule that draws inspiration from an auto-regressive pipeline to augmenttemporal consistency between adjacent frames, where the preceding frame guidesthe synthesis process of the current frame. Comparisons with state-of-the-artmethods demonstrate that Dancing Avatar exhibits the capacity to generate humanvideos with markedly superior quality, both in terms of human and backgroundfidelity, as well as temporal coherence compared to existing state-of-the-artapproaches.

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