arXiv - CSCL: "Text-to-Video: a Two-stage Framework for Zero-sho…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

Text-to-Video: a Two-stage Framework for Zero-shot Identity-agnostic Talking-head Generation. (arXiv:2308.06457v1 [cs.CV])

http://arxiv.org/abs/2308.06457 #arXiv #NLProc

Aug 15, 2023, 03:17 · · arxiv-cscl · · ·

Sign in to participate in the conversation