computer-code-5
Temporal Human Motion Composition
3260 papers • 126 benchmarks • 313 datasets
Generate temporal sequences of human motions guided by different signals, e.g., textual descriptions, showing smooth, plausible, and realistic transitions.
(Image credit: Papersgraph)
Benchmarks
These leaderboards are used to track progress in temporal-human-motion-composition-11
Trend
Dataset
Best Model
Actions
No benchmarks available.
Libraries
iUse these libraries to find temporal-human-motion-composition-11 models and implementations
Datasets
No datasets available.
Subtasks
No subtasks available.
Most implemented papers
Seamless Human Motion Composition with Blended Positional Encodings
Sergio Escalera, Cristina Palmero, Germán Barquero•Thu Feb 22 2024
This work introduces FlowMDM, the first diffusion-based model that generates seamless Human Motion Compositions (HMC) without any postprocessing or redundant denoising steps, and introduces the Blended Positional Encodings, a technique that leverages both absolute and relative positional encodings in the denoising chain.
60