UniVVT reports state-of-the-art video and image virtual try-on by conditioning a diffusion video generator on task tokens from a multimodal language model, with no masks, poses, or warping at inference.
International Journal of Computer Vision , volume =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
UniVVT: A Unified End-to-End Framework for High-Fidelity Video Virtual Try-on
UniVVT reports state-of-the-art video and image virtual try-on by conditioning a diffusion video generator on task tokens from a multimodal language model, with no masks, poses, or warping at inference.