.png){kind=icon}
Return
Optimization algorithm improved by multi-autoencoder and difference smoothing for digital human sign language actions
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2026.115830.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• The combination of stacked autoencoder and convolutional autoencoder is used to optimize the motion data, which improves the positional accuracy and smoothness of the network’s output motion data. • A time attention mechanism is added to make the model pay more attention to hand movement information in motion data and improve the quality of sign language action optimization. • By adopting a smoothing algorithm based on finite difference, the problem of reduced smoothness of motion data after convolutional autoencoder is resolved.
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
No organization information available
Cited Papers
ShareSaveShareSaveAn adaptive kernel correlation filter algorithm and synchronization error correction method for 3D motion reconstruction of flexible flapping wings using binocular vision.png)
MEASUREMENT
MEASUREMENTIF5.6
PREAI
ShareSaveShareSaveTranscranial focused ultrasound to V5 enhances human visual motion brain-computer interface by modulating feature-based attentionNATURE COMMUNICATIONS
OAAI
NATURE COMMUNICATIONSIF15.7
OAAIShareSave
View more