1 paper
Siyuan Fan, Bo Du, Xiantao Cai +2
In this work, we propose TextIM, a novel framework for synthesizing TEXT-driven human Interactive Motions, with a focus on the precise alignment of part-level semantics. Existing m…