MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing
Paper β’ 2607.05376 β’ Published β’ 7
Weights for MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing (ECCV 2026).
See the repository README for setup and inference.
Noncommercial use only, under PolyForm Noncommercial 1.0.0 plus the
repository's
ADDITIONAL_TERMS.md,
which also prohibit military and surveillance use and apply to these weights.
Commercial licensing: galf251@gmail.com. Note that the pipeline requires CUT3R
(CC BY-NC-SA 4.0), so a commercial license for MV-Forcing does not by itself
permit commercial use of the full pipeline. Third-party inventory:
NOTICE.md.
@article{fiebelman2026mv,
title={MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing},
author={Fiebelman, Gal and Averbuch-Elor, Hadar and Benaim, Sagie},
journal={arXiv preprint arXiv:2607.05376},
year={2026}
}