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/

Zendy Plus

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

Zendy Tools

Zendy Open

Recently, significant progress has been made in text-based motion generation,enabling the generation of diverse and high-quality human motions that conformto textual descriptions. However, generating motions beyond the distribution oforiginal datasets remains challenging, i.e., zero-shot generation. By adoptinga divide-and-conquer strategy, we propose a new framework named Fine-GrainedHuman Motion Diffusion Model (FG-MDM) for zero-shot human motion generation.Specifically, we first parse previous vague textual annotations intofine-grained descriptions of different body parts by leveraging a largelanguage model. We then use these fine-grained descriptions to guide atransformer-based diffusion model, which further adopts a design of parttokens. FG-MDM can generate human motions beyond the scope of original datasetsowing to descriptions that are closer to motion essence. Our experimentalresults demonstrate the superiority of FG-MDM over previous methods inzero-shot settings. We will release our fine-grained textual annotations forHumanML3D and KIT.

FG-MDM: Towards Zero-Shot Human Motion Generation via ChatGPT-Refined Descriptions