BeClaude
Research2026-05-12

KAN Text to Vision? The Exploration of Kolmogorov-Arnold Networks for Multi-Scale Sequence-Based Pose Animation from Sign Language Notation

Source: Arxiv CS.AI

arXiv:2605.09572v1 Announce Type: cross Abstract: Sign language production from symbolic notation offers a scalable route to accessible sign animation. We present KANMultiSign, a multi-scale sequence generator that translates HamNoSys notation into two-dimensional human pose sequences. Our...

arxivpapersvision