Inloggen
inzicht - MiKASA Transformer
MiKASA: Multi-Key-Anchor & Scene-Aware Transformer for 3D Visual Grounding
MiKASA Transformer enhances 3D visual grounding with scene-aware object encoding and multi-key-anchor technique.
MiKASA: Multi-Key-Anchor & Scene-Aware Transformer for 3D Visual Grounding
MiKASA Transformer combines scene-aware object encoding and multi-key-anchor technique to enhance 3D visual grounding accuracy and explainability.
Producten
© 2024 by Linnk AI